nld-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10/checkpoint-7500/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.812358019082235,
  "eval_steps": 500,
  "global_step": 7500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 6.768713426589966,
      "epoch": 0.004543389368468878,
      "grad_norm": 1.1484375,
      "learning_rate": 2e-06,
      "loss": 6.7222,
      "mean_token_accuracy": 0.1095728911459446,
      "num_tokens": 10131.0,
      "step": 5
    },
    {
      "entropy": 6.720575475692749,
      "epoch": 0.009086778736937756,
      "grad_norm": 1.3203125,
      "learning_rate": 4.5e-06,
      "loss": 6.6561,
      "mean_token_accuracy": 0.10461715683341026,
      "num_tokens": 20042.0,
      "step": 10
    },
    {
      "entropy": 6.794066381454468,
      "epoch": 0.013630168105406633,
      "grad_norm": 1.21875,
      "learning_rate": 7e-06,
      "loss": 6.6777,
      "mean_token_accuracy": 0.10963529199361802,
      "num_tokens": 30596.0,
      "step": 15
    },
    {
      "entropy": 6.780745935440064,
      "epoch": 0.01817355747387551,
      "grad_norm": 1.1484375,
      "learning_rate": 9.5e-06,
      "loss": 6.6242,
      "mean_token_accuracy": 0.11041242256760597,
      "num_tokens": 40667.0,
      "step": 20
    },
    {
      "entropy": 6.800935125350952,
      "epoch": 0.02271694684234439,
      "grad_norm": 1.4140625,
      "learning_rate": 1.2e-05,
      "loss": 6.6384,
      "mean_token_accuracy": 0.10581032261252403,
      "num_tokens": 51027.0,
      "step": 25
    },
    {
      "entropy": 6.829052972793579,
      "epoch": 0.027260336210813266,
      "grad_norm": 1.109375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 6.7264,
      "mean_token_accuracy": 0.10166022032499314,
      "num_tokens": 61860.0,
      "step": 30
    },
    {
      "entropy": 6.852202463150024,
      "epoch": 0.03180372557928214,
      "grad_norm": 0.99609375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 6.5345,
      "mean_token_accuracy": 0.1121172659099102,
      "num_tokens": 72136.0,
      "step": 35
    },
    {
      "entropy": 6.821422100067139,
      "epoch": 0.03634711494775102,
      "grad_norm": 1.0703125,
      "learning_rate": 1.95e-05,
      "loss": 6.624,
      "mean_token_accuracy": 0.10981205776333809,
      "num_tokens": 82611.0,
      "step": 40
    },
    {
      "entropy": 6.870565414428711,
      "epoch": 0.0408905043162199,
      "grad_norm": 1.203125,
      "learning_rate": 2.2e-05,
      "loss": 6.6909,
      "mean_token_accuracy": 0.10643857344985008,
      "num_tokens": 92722.0,
      "step": 45
    },
    {
      "entropy": 6.841821718215942,
      "epoch": 0.04543389368468878,
      "grad_norm": 1.0703125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.6865,
      "mean_token_accuracy": 0.10719444304704666,
      "num_tokens": 103258.0,
      "step": 50
    },
    {
      "entropy": 6.887676668167114,
      "epoch": 0.049977283053157656,
      "grad_norm": 1.203125,
      "learning_rate": 2.7e-05,
      "loss": 6.7029,
      "mean_token_accuracy": 0.10421370416879654,
      "num_tokens": 114277.0,
      "step": 55
    },
    {
      "entropy": 6.867907333374023,
      "epoch": 0.05452067242162653,
      "grad_norm": 1.0703125,
      "learning_rate": 2.95e-05,
      "loss": 6.6028,
      "mean_token_accuracy": 0.10567226111888886,
      "num_tokens": 124098.0,
      "step": 60
    },
    {
      "entropy": 6.858053731918335,
      "epoch": 0.059064061790095414,
      "grad_norm": 1.1328125,
      "learning_rate": 3.2e-05,
      "loss": 6.6126,
      "mean_token_accuracy": 0.11433581411838531,
      "num_tokens": 133775.0,
      "step": 65
    },
    {
      "entropy": 6.852071142196655,
      "epoch": 0.06360745115856428,
      "grad_norm": 1.0390625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.6068,
      "mean_token_accuracy": 0.11076592430472373,
      "num_tokens": 144765.0,
      "step": 70
    },
    {
      "entropy": 6.78433780670166,
      "epoch": 0.06815084052703317,
      "grad_norm": 1.265625,
      "learning_rate": 3.7e-05,
      "loss": 6.6368,
      "mean_token_accuracy": 0.1169683001935482,
      "num_tokens": 154641.0,
      "step": 75
    },
    {
      "entropy": 6.829514408111573,
      "epoch": 0.07269422989550205,
      "grad_norm": 1.1953125,
      "learning_rate": 3.95e-05,
      "loss": 6.5928,
      "mean_token_accuracy": 0.11736959144473076,
      "num_tokens": 164728.0,
      "step": 80
    },
    {
      "entropy": 6.903993606567383,
      "epoch": 0.07723761926397092,
      "grad_norm": 1.09375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.5341,
      "mean_token_accuracy": 0.10781509801745415,
      "num_tokens": 174739.0,
      "step": 85
    },
    {
      "entropy": 6.8236547946929935,
      "epoch": 0.0817810086324398,
      "grad_norm": 1.015625,
      "learning_rate": 4.45e-05,
      "loss": 6.5946,
      "mean_token_accuracy": 0.1138243854045868,
      "num_tokens": 184580.0,
      "step": 90
    },
    {
      "entropy": 6.850541257858277,
      "epoch": 0.08632439800090867,
      "grad_norm": 1.1171875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 6.6544,
      "mean_token_accuracy": 0.1113918036222458,
      "num_tokens": 195243.0,
      "step": 95
    },
    {
      "entropy": 6.854983806610107,
      "epoch": 0.09086778736937756,
      "grad_norm": 1.1171875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.6038,
      "mean_token_accuracy": 0.11426160708069802,
      "num_tokens": 205751.0,
      "step": 100
    },
    {
      "entropy": 6.885841703414917,
      "epoch": 0.09541117673784644,
      "grad_norm": 1.109375,
      "learning_rate": 5.2e-05,
      "loss": 6.5497,
      "mean_token_accuracy": 0.11042018011212348,
      "num_tokens": 217283.0,
      "step": 105
    },
    {
      "entropy": 6.744452333450317,
      "epoch": 0.09995456610631531,
      "grad_norm": 1.1875,
      "learning_rate": 5.45e-05,
      "loss": 6.51,
      "mean_token_accuracy": 0.11327873021364213,
      "num_tokens": 226909.0,
      "step": 110
    },
    {
      "entropy": 6.87985577583313,
      "epoch": 0.10449795547478419,
      "grad_norm": 1.0546875,
      "learning_rate": 5.7e-05,
      "loss": 6.5855,
      "mean_token_accuracy": 0.11328148767352104,
      "num_tokens": 237343.0,
      "step": 115
    },
    {
      "entropy": 6.860165071487427,
      "epoch": 0.10904134484325306,
      "grad_norm": 1.09375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.5179,
      "mean_token_accuracy": 0.1166656605899334,
      "num_tokens": 246601.0,
      "step": 120
    },
    {
      "entropy": 6.857304906845092,
      "epoch": 0.11358473421172194,
      "grad_norm": 1.265625,
      "learning_rate": 6.2e-05,
      "loss": 6.5727,
      "mean_token_accuracy": 0.11443846747279167,
      "num_tokens": 256818.0,
      "step": 125
    },
    {
      "entropy": 6.8257914066314695,
      "epoch": 0.11812812358019083,
      "grad_norm": 1.046875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.4876,
      "mean_token_accuracy": 0.11914753541350365,
      "num_tokens": 267157.0,
      "step": 130
    },
    {
      "entropy": 6.724686956405639,
      "epoch": 0.1226715129486597,
      "grad_norm": 1.171875,
      "learning_rate": 6.7e-05,
      "loss": 6.4988,
      "mean_token_accuracy": 0.11270106509327889,
      "num_tokens": 277370.0,
      "step": 135
    },
    {
      "entropy": 6.806635904312134,
      "epoch": 0.12721490231712856,
      "grad_norm": 1.140625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.5163,
      "mean_token_accuracy": 0.11316018253564834,
      "num_tokens": 287851.0,
      "step": 140
    },
    {
      "entropy": 6.965766048431396,
      "epoch": 0.13175829168559747,
      "grad_norm": 1.1640625,
      "learning_rate": 7.2e-05,
      "loss": 6.6036,
      "mean_token_accuracy": 0.11733311712741852,
      "num_tokens": 299043.0,
      "step": 145
    },
    {
      "entropy": 6.744367074966431,
      "epoch": 0.13630168105406634,
      "grad_norm": 1.1796875,
      "learning_rate": 7.45e-05,
      "loss": 6.5794,
      "mean_token_accuracy": 0.11460703909397126,
      "num_tokens": 308525.0,
      "step": 150
    },
    {
      "entropy": 6.8237261295318605,
      "epoch": 0.14084507042253522,
      "grad_norm": 1.5546875,
      "learning_rate": 7.7e-05,
      "loss": 6.5493,
      "mean_token_accuracy": 0.11844029426574706,
      "num_tokens": 318301.0,
      "step": 155
    },
    {
      "entropy": 6.790290117263794,
      "epoch": 0.1453884597910041,
      "grad_norm": 1.203125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.6009,
      "mean_token_accuracy": 0.11239130273461342,
      "num_tokens": 329599.0,
      "step": 160
    },
    {
      "entropy": 6.843774080276489,
      "epoch": 0.14993184915947297,
      "grad_norm": 1.2109375,
      "learning_rate": 8.2e-05,
      "loss": 6.4774,
      "mean_token_accuracy": 0.11909914687275887,
      "num_tokens": 339446.0,
      "step": 165
    },
    {
      "entropy": 6.9367317199707035,
      "epoch": 0.15447523852794184,
      "grad_norm": 1.1953125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.6333,
      "mean_token_accuracy": 0.1179355539381504,
      "num_tokens": 349579.0,
      "step": 170
    },
    {
      "entropy": 6.8179216384887695,
      "epoch": 0.15901862789641072,
      "grad_norm": 1.1328125,
      "learning_rate": 8.7e-05,
      "loss": 6.5631,
      "mean_token_accuracy": 0.11152750179171562,
      "num_tokens": 359452.0,
      "step": 175
    },
    {
      "entropy": 6.830770826339721,
      "epoch": 0.1635620172648796,
      "grad_norm": 1.28125,
      "learning_rate": 8.95e-05,
      "loss": 6.5702,
      "mean_token_accuracy": 0.11352085247635842,
      "num_tokens": 369558.0,
      "step": 180
    },
    {
      "entropy": 6.878524255752564,
      "epoch": 0.16810540663334847,
      "grad_norm": 1.140625,
      "learning_rate": 9.2e-05,
      "loss": 6.5577,
      "mean_token_accuracy": 0.11322271078824997,
      "num_tokens": 380458.0,
      "step": 185
    },
    {
      "entropy": 6.9294568538665775,
      "epoch": 0.17264879600181735,
      "grad_norm": 1.1484375,
      "learning_rate": 9.45e-05,
      "loss": 6.5669,
      "mean_token_accuracy": 0.1112134151160717,
      "num_tokens": 390051.0,
      "step": 190
    },
    {
      "entropy": 6.777473735809326,
      "epoch": 0.17719218537028622,
      "grad_norm": 1.28125,
      "learning_rate": 9.7e-05,
      "loss": 6.5164,
      "mean_token_accuracy": 0.11806451380252839,
      "num_tokens": 400343.0,
      "step": 195
    },
    {
      "entropy": 6.889895868301392,
      "epoch": 0.18173557473875512,
      "grad_norm": 1.109375,
      "learning_rate": 9.95e-05,
      "loss": 6.5619,
      "mean_token_accuracy": 0.11565203443169594,
      "num_tokens": 410004.0,
      "step": 200
    },
    {
      "entropy": 6.7963909149169925,
      "epoch": 0.186278964107224,
      "grad_norm": 1.109375,
      "learning_rate": 0.000102,
      "loss": 6.5851,
      "mean_token_accuracy": 0.10973369553685189,
      "num_tokens": 421286.0,
      "step": 205
    },
    {
      "entropy": 6.9212535381317135,
      "epoch": 0.19082235347569287,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.6359,
      "mean_token_accuracy": 0.10912902057170867,
      "num_tokens": 432351.0,
      "step": 210
    },
    {
      "entropy": 6.842854261398315,
      "epoch": 0.19536574284416175,
      "grad_norm": 1.328125,
      "learning_rate": 0.000107,
      "loss": 6.627,
      "mean_token_accuracy": 0.11367111280560493,
      "num_tokens": 442843.0,
      "step": 215
    },
    {
      "entropy": 6.870680522918701,
      "epoch": 0.19990913221263062,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001095,
      "loss": 6.4973,
      "mean_token_accuracy": 0.11255754232406616,
      "num_tokens": 453370.0,
      "step": 220
    },
    {
      "entropy": 6.800969171524048,
      "epoch": 0.2044525215810995,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000112,
      "loss": 6.6865,
      "mean_token_accuracy": 0.10846318081021308,
      "num_tokens": 464715.0,
      "step": 225
    },
    {
      "entropy": 6.875910711288452,
      "epoch": 0.20899591094956838,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001145,
      "loss": 6.4927,
      "mean_token_accuracy": 0.11165988445281982,
      "num_tokens": 475086.0,
      "step": 230
    },
    {
      "entropy": 6.743642044067383,
      "epoch": 0.21353930031803725,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.6147,
      "mean_token_accuracy": 0.12273732423782349,
      "num_tokens": 486088.0,
      "step": 235
    },
    {
      "entropy": 6.893033027648926,
      "epoch": 0.21808268968650613,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.563,
      "mean_token_accuracy": 0.10623423308134079,
      "num_tokens": 496978.0,
      "step": 240
    },
    {
      "entropy": 6.730010509490967,
      "epoch": 0.222626079054975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000122,
      "loss": 6.3713,
      "mean_token_accuracy": 0.12526726275682448,
      "num_tokens": 507794.0,
      "step": 245
    },
    {
      "entropy": 6.739193630218506,
      "epoch": 0.22716946842344388,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001245,
      "loss": 6.5377,
      "mean_token_accuracy": 0.10970785096287727,
      "num_tokens": 517763.0,
      "step": 250
    },
    {
      "entropy": 6.85530915260315,
      "epoch": 0.23171285779191278,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000127,
      "loss": 6.4664,
      "mean_token_accuracy": 0.11921982616186141,
      "num_tokens": 527374.0,
      "step": 255
    },
    {
      "entropy": 6.741254997253418,
      "epoch": 0.23625624716038165,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001295,
      "loss": 6.5256,
      "mean_token_accuracy": 0.11517209336161613,
      "num_tokens": 537435.0,
      "step": 260
    },
    {
      "entropy": 6.810067796707154,
      "epoch": 0.24079963652885053,
      "grad_norm": 1.265625,
      "learning_rate": 0.000132,
      "loss": 6.4963,
      "mean_token_accuracy": 0.11652736142277717,
      "num_tokens": 547236.0,
      "step": 265
    },
    {
      "entropy": 6.788457632064819,
      "epoch": 0.2453430258973194,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.4943,
      "mean_token_accuracy": 0.11792855635285378,
      "num_tokens": 557130.0,
      "step": 270
    },
    {
      "entropy": 6.698097038269043,
      "epoch": 0.24988641526578828,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.5301,
      "mean_token_accuracy": 0.11694961413741112,
      "num_tokens": 567482.0,
      "step": 275
    },
    {
      "entropy": 6.759066724777222,
      "epoch": 0.25442980463425713,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001395,
      "loss": 6.3925,
      "mean_token_accuracy": 0.11411905512213708,
      "num_tokens": 577084.0,
      "step": 280
    },
    {
      "entropy": 6.76326265335083,
      "epoch": 0.25897319400272606,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.5115,
      "mean_token_accuracy": 0.11677431538701058,
      "num_tokens": 587980.0,
      "step": 285
    },
    {
      "entropy": 6.667721605300903,
      "epoch": 0.26351658337119493,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001445,
      "loss": 6.4219,
      "mean_token_accuracy": 0.12288179770112037,
      "num_tokens": 598989.0,
      "step": 290
    },
    {
      "entropy": 6.985043001174927,
      "epoch": 0.2680599727396638,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000147,
      "loss": 6.6076,
      "mean_token_accuracy": 0.11418735831975937,
      "num_tokens": 608599.0,
      "step": 295
    },
    {
      "entropy": 6.601436614990234,
      "epoch": 0.2726033621081327,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001495,
      "loss": 6.4418,
      "mean_token_accuracy": 0.11823201850056649,
      "num_tokens": 619267.0,
      "step": 300
    },
    {
      "entropy": 6.909968709945678,
      "epoch": 0.27714675147660156,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000152,
      "loss": 6.6386,
      "mean_token_accuracy": 0.12054958641529083,
      "num_tokens": 629991.0,
      "step": 305
    },
    {
      "entropy": 6.833962726593017,
      "epoch": 0.28169014084507044,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.5693,
      "mean_token_accuracy": 0.11267199367284775,
      "num_tokens": 641311.0,
      "step": 310
    },
    {
      "entropy": 6.812391519546509,
      "epoch": 0.2862335302135393,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000157,
      "loss": 6.5406,
      "mean_token_accuracy": 0.11235396489501,
      "num_tokens": 651812.0,
      "step": 315
    },
    {
      "entropy": 6.799654388427735,
      "epoch": 0.2907769195820082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001595,
      "loss": 6.5081,
      "mean_token_accuracy": 0.11410669162869454,
      "num_tokens": 662387.0,
      "step": 320
    },
    {
      "entropy": 6.780321025848389,
      "epoch": 0.29532030895047706,
      "grad_norm": 1.3125,
      "learning_rate": 0.000162,
      "loss": 6.4997,
      "mean_token_accuracy": 0.11494562402367592,
      "num_tokens": 673018.0,
      "step": 325
    },
    {
      "entropy": 6.777340126037598,
      "epoch": 0.29986369831894594,
      "grad_norm": 1.25,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.434,
      "mean_token_accuracy": 0.12099529802799225,
      "num_tokens": 683192.0,
      "step": 330
    },
    {
      "entropy": 6.75422306060791,
      "epoch": 0.3044070876874148,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.5675,
      "mean_token_accuracy": 0.1253136858344078,
      "num_tokens": 692747.0,
      "step": 335
    },
    {
      "entropy": 6.676179361343384,
      "epoch": 0.3089504770558837,
      "grad_norm": 0.96875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.5389,
      "mean_token_accuracy": 0.11221015229821205,
      "num_tokens": 705109.0,
      "step": 340
    },
    {
      "entropy": 6.835248422622681,
      "epoch": 0.31349386642435256,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.4544,
      "mean_token_accuracy": 0.11584834307432175,
      "num_tokens": 715658.0,
      "step": 345
    },
    {
      "entropy": 6.698165464401245,
      "epoch": 0.31803725579282144,
      "grad_norm": 1.25,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.4818,
      "mean_token_accuracy": 0.1224296160042286,
      "num_tokens": 724961.0,
      "step": 350
    },
    {
      "entropy": 6.6860706329345705,
      "epoch": 0.3225806451612903,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000177,
      "loss": 6.4358,
      "mean_token_accuracy": 0.1249639131128788,
      "num_tokens": 735416.0,
      "step": 355
    },
    {
      "entropy": 6.823048305511475,
      "epoch": 0.3271240345297592,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001795,
      "loss": 6.5408,
      "mean_token_accuracy": 0.12204954698681832,
      "num_tokens": 745005.0,
      "step": 360
    },
    {
      "entropy": 6.874798774719238,
      "epoch": 0.33166742389822806,
      "grad_norm": 1.34375,
      "learning_rate": 0.000182,
      "loss": 6.5836,
      "mean_token_accuracy": 0.11828166246414185,
      "num_tokens": 754743.0,
      "step": 365
    },
    {
      "entropy": 6.583785772323608,
      "epoch": 0.33621081326669694,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001845,
      "loss": 6.437,
      "mean_token_accuracy": 0.1200392097234726,
      "num_tokens": 764897.0,
      "step": 370
    },
    {
      "entropy": 6.818075704574585,
      "epoch": 0.3407542026351658,
      "grad_norm": 1.234375,
      "learning_rate": 0.000187,
      "loss": 6.511,
      "mean_token_accuracy": 0.11428396552801132,
      "num_tokens": 775302.0,
      "step": 375
    },
    {
      "entropy": 6.651220989227295,
      "epoch": 0.3452975920036347,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001895,
      "loss": 6.4495,
      "mean_token_accuracy": 0.12070777788758277,
      "num_tokens": 785694.0,
      "step": 380
    },
    {
      "entropy": 6.732499074935913,
      "epoch": 0.34984098137210357,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000192,
      "loss": 6.3957,
      "mean_token_accuracy": 0.12520513609051703,
      "num_tokens": 795045.0,
      "step": 385
    },
    {
      "entropy": 6.696088409423828,
      "epoch": 0.35438437074057244,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001945,
      "loss": 6.4921,
      "mean_token_accuracy": 0.1158549316227436,
      "num_tokens": 805579.0,
      "step": 390
    },
    {
      "entropy": 6.654592990875244,
      "epoch": 0.35892776010904137,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.3914,
      "mean_token_accuracy": 0.12176291272044182,
      "num_tokens": 816743.0,
      "step": 395
    },
    {
      "entropy": 6.768208026885986,
      "epoch": 0.36347114947751025,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.5313,
      "mean_token_accuracy": 0.11511081382632256,
      "num_tokens": 827217.0,
      "step": 400
    },
    {
      "entropy": 6.615100955963134,
      "epoch": 0.3680145388459791,
      "grad_norm": 1.28125,
      "learning_rate": 0.000202,
      "loss": 6.3835,
      "mean_token_accuracy": 0.1261222928762436,
      "num_tokens": 837555.0,
      "step": 405
    },
    {
      "entropy": 6.711054515838623,
      "epoch": 0.372557928214448,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.4678,
      "mean_token_accuracy": 0.11874150708317757,
      "num_tokens": 846726.0,
      "step": 410
    },
    {
      "entropy": 6.75947527885437,
      "epoch": 0.3771013175829169,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000207,
      "loss": 6.4571,
      "mean_token_accuracy": 0.12268495261669159,
      "num_tokens": 856535.0,
      "step": 415
    },
    {
      "entropy": 6.683750581741333,
      "epoch": 0.38164470695138575,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002095,
      "loss": 6.531,
      "mean_token_accuracy": 0.1186832346022129,
      "num_tokens": 866973.0,
      "step": 420
    },
    {
      "entropy": 6.770697355270386,
      "epoch": 0.3861880963198546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000212,
      "loss": 6.4716,
      "mean_token_accuracy": 0.11631686612963676,
      "num_tokens": 877873.0,
      "step": 425
    },
    {
      "entropy": 6.69672966003418,
      "epoch": 0.3907314856883235,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002145,
      "loss": 6.4657,
      "mean_token_accuracy": 0.11843668594956398,
      "num_tokens": 887135.0,
      "step": 430
    },
    {
      "entropy": 6.835273313522339,
      "epoch": 0.3952748750567924,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.4843,
      "mean_token_accuracy": 0.1205623060464859,
      "num_tokens": 897206.0,
      "step": 435
    },
    {
      "entropy": 6.631858015060425,
      "epoch": 0.39981826442526125,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002195,
      "loss": 6.4857,
      "mean_token_accuracy": 0.1176255464553833,
      "num_tokens": 906808.0,
      "step": 440
    },
    {
      "entropy": 6.630966377258301,
      "epoch": 0.4043616537937301,
      "grad_norm": 1.15625,
      "learning_rate": 0.000222,
      "loss": 6.3833,
      "mean_token_accuracy": 0.1273773953318596,
      "num_tokens": 917177.0,
      "step": 445
    },
    {
      "entropy": 6.6929981231689455,
      "epoch": 0.408905043162199,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002245,
      "loss": 6.4283,
      "mean_token_accuracy": 0.11764879748225213,
      "num_tokens": 927229.0,
      "step": 450
    },
    {
      "entropy": 6.704322004318238,
      "epoch": 0.4134484325306679,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.4261,
      "mean_token_accuracy": 0.12415409311652184,
      "num_tokens": 937832.0,
      "step": 455
    },
    {
      "entropy": 6.592918109893799,
      "epoch": 0.41799182189913675,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.3477,
      "mean_token_accuracy": 0.12181811258196831,
      "num_tokens": 948038.0,
      "step": 460
    },
    {
      "entropy": 6.663812637329102,
      "epoch": 0.4225352112676056,
      "grad_norm": 1.109375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.4591,
      "mean_token_accuracy": 0.11563793793320656,
      "num_tokens": 957924.0,
      "step": 465
    },
    {
      "entropy": 6.706633520126343,
      "epoch": 0.4270786006360745,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.4292,
      "mean_token_accuracy": 0.1167376309633255,
      "num_tokens": 968497.0,
      "step": 470
    },
    {
      "entropy": 6.624789667129517,
      "epoch": 0.4316219900045434,
      "grad_norm": 1.125,
      "learning_rate": 0.000237,
      "loss": 6.3514,
      "mean_token_accuracy": 0.12173967584967613,
      "num_tokens": 978878.0,
      "step": 475
    },
    {
      "entropy": 6.5878393173217775,
      "epoch": 0.43616537937301225,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002395,
      "loss": 6.4639,
      "mean_token_accuracy": 0.12137211859226227,
      "num_tokens": 989984.0,
      "step": 480
    },
    {
      "entropy": 6.6825902462005615,
      "epoch": 0.4407087687414811,
      "grad_norm": 1.1875,
      "learning_rate": 0.000242,
      "loss": 6.3862,
      "mean_token_accuracy": 0.12439708858728409,
      "num_tokens": 1000254.0,
      "step": 485
    },
    {
      "entropy": 6.580802011489868,
      "epoch": 0.44525215810995,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002445,
      "loss": 6.4602,
      "mean_token_accuracy": 0.11807657480239868,
      "num_tokens": 1010697.0,
      "step": 490
    },
    {
      "entropy": 6.8235663890838625,
      "epoch": 0.4497955474784189,
      "grad_norm": 1.171875,
      "learning_rate": 0.000247,
      "loss": 6.5558,
      "mean_token_accuracy": 0.1186821848154068,
      "num_tokens": 1020892.0,
      "step": 495
    },
    {
      "entropy": 6.656906843185425,
      "epoch": 0.45433893684688775,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002495,
      "loss": 6.4358,
      "mean_token_accuracy": 0.1271519921720028,
      "num_tokens": 1030380.0,
      "step": 500
    },
    {
      "epoch": 0.45433893684688775,
      "eval_entropy": 6.70977929470151,
      "eval_loss": 6.51187801361084,
      "eval_mean_token_accuracy": 0.11979533588990222,
      "eval_num_tokens": 1030380.0,
      "eval_runtime": 2.6058,
      "eval_samples_per_second": 1318.216,
      "eval_steps_per_second": 165.017,
      "step": 500
    },
    {
      "entropy": 6.737278318405151,
      "epoch": 0.45888232621535663,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000252,
      "loss": 6.4798,
      "mean_token_accuracy": 0.11720329746603966,
      "num_tokens": 1040301.0,
      "step": 505
    },
    {
      "entropy": 6.551387023925781,
      "epoch": 0.46342571558382556,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002545,
      "loss": 6.3367,
      "mean_token_accuracy": 0.12788272872567177,
      "num_tokens": 1051999.0,
      "step": 510
    },
    {
      "entropy": 6.841769361495972,
      "epoch": 0.46796910495229443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000257,
      "loss": 6.5815,
      "mean_token_accuracy": 0.11753517240285874,
      "num_tokens": 1063133.0,
      "step": 515
    },
    {
      "entropy": 6.560217952728271,
      "epoch": 0.4725124943207633,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002595,
      "loss": 6.3922,
      "mean_token_accuracy": 0.11554303765296936,
      "num_tokens": 1073972.0,
      "step": 520
    },
    {
      "entropy": 6.600145530700684,
      "epoch": 0.4770558836892322,
      "grad_norm": 1.59375,
      "learning_rate": 0.000262,
      "loss": 6.3869,
      "mean_token_accuracy": 0.11973805353045464,
      "num_tokens": 1084050.0,
      "step": 525
    },
    {
      "entropy": 6.673818635940552,
      "epoch": 0.48159927305770106,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.4799,
      "mean_token_accuracy": 0.12647488638758658,
      "num_tokens": 1094415.0,
      "step": 530
    },
    {
      "entropy": 6.534582090377808,
      "epoch": 0.48614266242616994,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.4305,
      "mean_token_accuracy": 0.11784208044409752,
      "num_tokens": 1105133.0,
      "step": 535
    },
    {
      "entropy": 6.662813949584961,
      "epoch": 0.4906860517946388,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.3911,
      "mean_token_accuracy": 0.12304398790001869,
      "num_tokens": 1115266.0,
      "step": 540
    },
    {
      "entropy": 6.537495183944702,
      "epoch": 0.4952294411631077,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.3602,
      "mean_token_accuracy": 0.12327887490391731,
      "num_tokens": 1124775.0,
      "step": 545
    },
    {
      "entropy": 6.56860499382019,
      "epoch": 0.49977283053157656,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002745,
      "loss": 6.3894,
      "mean_token_accuracy": 0.12164504677057267,
      "num_tokens": 1134498.0,
      "step": 550
    },
    {
      "entropy": 6.696820211410523,
      "epoch": 0.5043162199000454,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000277,
      "loss": 6.4295,
      "mean_token_accuracy": 0.11844469383358955,
      "num_tokens": 1146067.0,
      "step": 555
    },
    {
      "entropy": 6.667341804504394,
      "epoch": 0.5088596092685143,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002795,
      "loss": 6.4374,
      "mean_token_accuracy": 0.11883379444479943,
      "num_tokens": 1156349.0,
      "step": 560
    },
    {
      "entropy": 6.60188570022583,
      "epoch": 0.5134029986369832,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.3706,
      "mean_token_accuracy": 0.1226509042084217,
      "num_tokens": 1166851.0,
      "step": 565
    },
    {
      "entropy": 6.5608902931213375,
      "epoch": 0.5179463880054521,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002845,
      "loss": 6.3262,
      "mean_token_accuracy": 0.12422406151890755,
      "num_tokens": 1176899.0,
      "step": 570
    },
    {
      "entropy": 6.576656770706177,
      "epoch": 0.522489777373921,
      "grad_norm": 1.0625,
      "learning_rate": 0.000287,
      "loss": 6.4297,
      "mean_token_accuracy": 0.1186425231397152,
      "num_tokens": 1187383.0,
      "step": 575
    },
    {
      "entropy": 6.718804931640625,
      "epoch": 0.5270331667423899,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002895,
      "loss": 6.4153,
      "mean_token_accuracy": 0.12464964166283607,
      "num_tokens": 1197836.0,
      "step": 580
    },
    {
      "entropy": 6.5609543800354,
      "epoch": 0.5315765561108587,
      "grad_norm": 1.25,
      "learning_rate": 0.000292,
      "loss": 6.3661,
      "mean_token_accuracy": 0.1282975621521473,
      "num_tokens": 1208533.0,
      "step": 585
    },
    {
      "entropy": 6.60496711730957,
      "epoch": 0.5361199454793276,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002945,
      "loss": 6.4107,
      "mean_token_accuracy": 0.11816166192293168,
      "num_tokens": 1219397.0,
      "step": 590
    },
    {
      "entropy": 6.6025289535522464,
      "epoch": 0.5406633348477965,
      "grad_norm": 1.140625,
      "learning_rate": 0.000297,
      "loss": 6.4063,
      "mean_token_accuracy": 0.11670600771903991,
      "num_tokens": 1229991.0,
      "step": 595
    },
    {
      "entropy": 6.493481159210205,
      "epoch": 0.5452067242162654,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002995,
      "loss": 6.2923,
      "mean_token_accuracy": 0.11882022395730019,
      "num_tokens": 1240343.0,
      "step": 600
    },
    {
      "entropy": 6.529299545288086,
      "epoch": 0.5497501135847342,
      "grad_norm": 1.203125,
      "learning_rate": 0.000302,
      "loss": 6.3761,
      "mean_token_accuracy": 0.11741307675838471,
      "num_tokens": 1250432.0,
      "step": 605
    },
    {
      "entropy": 6.5340169906616214,
      "epoch": 0.5542935029532031,
      "grad_norm": 1.25,
      "learning_rate": 0.0003045,
      "loss": 6.3185,
      "mean_token_accuracy": 0.12538510411977768,
      "num_tokens": 1260744.0,
      "step": 610
    },
    {
      "entropy": 6.600591993331909,
      "epoch": 0.558836892321672,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000307,
      "loss": 6.339,
      "mean_token_accuracy": 0.12402234598994255,
      "num_tokens": 1272155.0,
      "step": 615
    },
    {
      "entropy": 6.398069381713867,
      "epoch": 0.5633802816901409,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003095,
      "loss": 6.2189,
      "mean_token_accuracy": 0.12761275619268417,
      "num_tokens": 1282240.0,
      "step": 620
    },
    {
      "entropy": 6.573273801803589,
      "epoch": 0.5679236710586097,
      "grad_norm": 1.21875,
      "learning_rate": 0.000312,
      "loss": 6.4098,
      "mean_token_accuracy": 0.12361594140529633,
      "num_tokens": 1292169.0,
      "step": 625
    },
    {
      "entropy": 6.485548496246338,
      "epoch": 0.5724670604270786,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003145,
      "loss": 6.3056,
      "mean_token_accuracy": 0.12872742787003516,
      "num_tokens": 1302010.0,
      "step": 630
    },
    {
      "entropy": 6.528713846206665,
      "epoch": 0.5770104497955475,
      "grad_norm": 1.15625,
      "learning_rate": 0.000317,
      "loss": 6.365,
      "mean_token_accuracy": 0.12618476673960685,
      "num_tokens": 1312596.0,
      "step": 635
    },
    {
      "entropy": 6.589738416671753,
      "epoch": 0.5815538391640164,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003195,
      "loss": 6.313,
      "mean_token_accuracy": 0.11931094974279403,
      "num_tokens": 1321753.0,
      "step": 640
    },
    {
      "entropy": 6.462535858154297,
      "epoch": 0.5860972285324852,
      "grad_norm": 1.171875,
      "learning_rate": 0.000322,
      "loss": 6.356,
      "mean_token_accuracy": 0.12229031547904015,
      "num_tokens": 1332059.0,
      "step": 645
    },
    {
      "entropy": 6.533907556533814,
      "epoch": 0.5906406179009541,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.3321,
      "mean_token_accuracy": 0.1260114513337612,
      "num_tokens": 1341517.0,
      "step": 650
    },
    {
      "entropy": 6.640655851364135,
      "epoch": 0.595184007269423,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4102,
      "mean_token_accuracy": 0.12594819515943528,
      "num_tokens": 1351439.0,
      "step": 655
    },
    {
      "entropy": 6.483338737487793,
      "epoch": 0.5997273966378919,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.2893,
      "mean_token_accuracy": 0.12250665053725243,
      "num_tokens": 1360745.0,
      "step": 660
    },
    {
      "entropy": 6.401842403411865,
      "epoch": 0.6042707860063607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.2989,
      "mean_token_accuracy": 0.1253798320889473,
      "num_tokens": 1371931.0,
      "step": 665
    },
    {
      "entropy": 6.429254722595215,
      "epoch": 0.6088141753748296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.2382,
      "mean_token_accuracy": 0.12485149502754211,
      "num_tokens": 1382040.0,
      "step": 670
    },
    {
      "entropy": 6.451748323440552,
      "epoch": 0.6133575647432985,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000337,
      "loss": 6.2984,
      "mean_token_accuracy": 0.12093956097960472,
      "num_tokens": 1392690.0,
      "step": 675
    },
    {
      "entropy": 6.548810720443726,
      "epoch": 0.6179009541117674,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003395,
      "loss": 6.3009,
      "mean_token_accuracy": 0.12423968389630317,
      "num_tokens": 1403060.0,
      "step": 680
    },
    {
      "entropy": 6.452625370025634,
      "epoch": 0.6224443434802363,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000342,
      "loss": 6.315,
      "mean_token_accuracy": 0.124802715331316,
      "num_tokens": 1412895.0,
      "step": 685
    },
    {
      "entropy": 6.463527679443359,
      "epoch": 0.6269877328487051,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.2829,
      "mean_token_accuracy": 0.1315432645380497,
      "num_tokens": 1423024.0,
      "step": 690
    },
    {
      "entropy": 6.438167715072632,
      "epoch": 0.631531122217174,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000347,
      "loss": 6.2903,
      "mean_token_accuracy": 0.12509298026561738,
      "num_tokens": 1433578.0,
      "step": 695
    },
    {
      "entropy": 6.441053771972657,
      "epoch": 0.6360745115856429,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003495,
      "loss": 6.3409,
      "mean_token_accuracy": 0.12466292232275009,
      "num_tokens": 1444473.0,
      "step": 700
    },
    {
      "entropy": 6.530716371536255,
      "epoch": 0.6406179009541118,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000352,
      "loss": 6.3658,
      "mean_token_accuracy": 0.12251063138246536,
      "num_tokens": 1454553.0,
      "step": 705
    },
    {
      "entropy": 6.492323493957519,
      "epoch": 0.6451612903225806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003545,
      "loss": 6.2283,
      "mean_token_accuracy": 0.12746653333306313,
      "num_tokens": 1464162.0,
      "step": 710
    },
    {
      "entropy": 6.370441341400147,
      "epoch": 0.6497046796910495,
      "grad_norm": 1.09375,
      "learning_rate": 0.000357,
      "loss": 6.2923,
      "mean_token_accuracy": 0.12994937747716903,
      "num_tokens": 1474691.0,
      "step": 715
    },
    {
      "entropy": 6.465943193435669,
      "epoch": 0.6542480690595184,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003595,
      "loss": 6.3715,
      "mean_token_accuracy": 0.12509030997753143,
      "num_tokens": 1484650.0,
      "step": 720
    },
    {
      "entropy": 6.523843812942505,
      "epoch": 0.6587914584279873,
      "grad_norm": 1.171875,
      "learning_rate": 0.000362,
      "loss": 6.3244,
      "mean_token_accuracy": 0.12303805649280548,
      "num_tokens": 1494788.0,
      "step": 725
    },
    {
      "entropy": 6.440337705612182,
      "epoch": 0.6633348477964561,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003645,
      "loss": 6.2751,
      "mean_token_accuracy": 0.13085415214300156,
      "num_tokens": 1506574.0,
      "step": 730
    },
    {
      "entropy": 6.369408988952637,
      "epoch": 0.667878237164925,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000367,
      "loss": 6.2929,
      "mean_token_accuracy": 0.12216815054416656,
      "num_tokens": 1517505.0,
      "step": 735
    },
    {
      "entropy": 6.457225704193116,
      "epoch": 0.6724216265333939,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003695,
      "loss": 6.1799,
      "mean_token_accuracy": 0.1351654462516308,
      "num_tokens": 1526662.0,
      "step": 740
    },
    {
      "entropy": 6.37555251121521,
      "epoch": 0.6769650159018628,
      "grad_norm": 1.203125,
      "learning_rate": 0.000372,
      "loss": 6.3026,
      "mean_token_accuracy": 0.1286969043314457,
      "num_tokens": 1536798.0,
      "step": 745
    },
    {
      "entropy": 6.40103816986084,
      "epoch": 0.6815084052703316,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003745,
      "loss": 6.2123,
      "mean_token_accuracy": 0.13174753487110138,
      "num_tokens": 1546673.0,
      "step": 750
    },
    {
      "entropy": 6.403094959259033,
      "epoch": 0.6860517946388005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000377,
      "loss": 6.236,
      "mean_token_accuracy": 0.13387759551405906,
      "num_tokens": 1556793.0,
      "step": 755
    },
    {
      "entropy": 6.438587808609009,
      "epoch": 0.6905951840072694,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003795,
      "loss": 6.283,
      "mean_token_accuracy": 0.12869395688176155,
      "num_tokens": 1566450.0,
      "step": 760
    },
    {
      "entropy": 6.352961111068725,
      "epoch": 0.6951385733757383,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000382,
      "loss": 6.3079,
      "mean_token_accuracy": 0.13057589232921601,
      "num_tokens": 1576353.0,
      "step": 765
    },
    {
      "entropy": 6.418825960159301,
      "epoch": 0.6996819627442071,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003845,
      "loss": 6.192,
      "mean_token_accuracy": 0.12764471173286437,
      "num_tokens": 1586318.0,
      "step": 770
    },
    {
      "entropy": 6.426858377456665,
      "epoch": 0.704225352112676,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.2869,
      "mean_token_accuracy": 0.12360998913645745,
      "num_tokens": 1596839.0,
      "step": 775
    },
    {
      "entropy": 6.35253381729126,
      "epoch": 0.7087687414811449,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.3133,
      "mean_token_accuracy": 0.12761991545557977,
      "num_tokens": 1606843.0,
      "step": 780
    },
    {
      "entropy": 6.324183559417724,
      "epoch": 0.7133121308496138,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.2747,
      "mean_token_accuracy": 0.12435603663325309,
      "num_tokens": 1617048.0,
      "step": 785
    },
    {
      "entropy": 6.413893175125122,
      "epoch": 0.7178555202180827,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.1888,
      "mean_token_accuracy": 0.13071049749851227,
      "num_tokens": 1627472.0,
      "step": 790
    },
    {
      "entropy": 6.403914165496826,
      "epoch": 0.7223989095865516,
      "grad_norm": 1.125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.2874,
      "mean_token_accuracy": 0.12295314446091651,
      "num_tokens": 1637716.0,
      "step": 795
    },
    {
      "entropy": 6.384047985076904,
      "epoch": 0.7269422989550205,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003995,
      "loss": 6.3057,
      "mean_token_accuracy": 0.12159052193164825,
      "num_tokens": 1648119.0,
      "step": 800
    },
    {
      "entropy": 6.422551822662354,
      "epoch": 0.7314856883234894,
      "grad_norm": 1.15625,
      "learning_rate": 0.000402,
      "loss": 6.263,
      "mean_token_accuracy": 0.12927623763680457,
      "num_tokens": 1658214.0,
      "step": 805
    },
    {
      "entropy": 6.272933769226074,
      "epoch": 0.7360290776919582,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004045,
      "loss": 6.222,
      "mean_token_accuracy": 0.12815281972289086,
      "num_tokens": 1668725.0,
      "step": 810
    },
    {
      "entropy": 6.38576192855835,
      "epoch": 0.7405724670604271,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.291,
      "mean_token_accuracy": 0.12685041427612304,
      "num_tokens": 1678891.0,
      "step": 815
    },
    {
      "entropy": 6.396528625488282,
      "epoch": 0.745115856428896,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004095,
      "loss": 6.1896,
      "mean_token_accuracy": 0.12780508771538734,
      "num_tokens": 1688749.0,
      "step": 820
    },
    {
      "entropy": 6.372272872924805,
      "epoch": 0.7496592457973649,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000412,
      "loss": 6.2922,
      "mean_token_accuracy": 0.12699832171201705,
      "num_tokens": 1699166.0,
      "step": 825
    },
    {
      "entropy": 6.382475662231445,
      "epoch": 0.7542026351658337,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004145,
      "loss": 6.2227,
      "mean_token_accuracy": 0.12901799380779266,
      "num_tokens": 1710690.0,
      "step": 830
    },
    {
      "entropy": 6.378543329238892,
      "epoch": 0.7587460245343026,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000417,
      "loss": 6.3242,
      "mean_token_accuracy": 0.12391438856720924,
      "num_tokens": 1721169.0,
      "step": 835
    },
    {
      "entropy": 6.325247430801392,
      "epoch": 0.7632894139027715,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004195,
      "loss": 6.1868,
      "mean_token_accuracy": 0.12631086334586145,
      "num_tokens": 1731523.0,
      "step": 840
    },
    {
      "entropy": 6.33087682723999,
      "epoch": 0.7678328032712404,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000422,
      "loss": 6.1862,
      "mean_token_accuracy": 0.12567417919635773,
      "num_tokens": 1742316.0,
      "step": 845
    },
    {
      "entropy": 6.3732359409332275,
      "epoch": 0.7723761926397092,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004245,
      "loss": 6.3318,
      "mean_token_accuracy": 0.12681131660938263,
      "num_tokens": 1753424.0,
      "step": 850
    },
    {
      "entropy": 6.253160429000855,
      "epoch": 0.7769195820081781,
      "grad_norm": 1.265625,
      "learning_rate": 0.000427,
      "loss": 6.1569,
      "mean_token_accuracy": 0.13139303475618364,
      "num_tokens": 1762814.0,
      "step": 855
    },
    {
      "entropy": 6.360691118240356,
      "epoch": 0.781462971376647,
      "grad_norm": 1.125,
      "learning_rate": 0.0004295,
      "loss": 6.3124,
      "mean_token_accuracy": 0.1301097348332405,
      "num_tokens": 1773096.0,
      "step": 860
    },
    {
      "entropy": 6.37277398109436,
      "epoch": 0.7860063607451159,
      "grad_norm": 1.0625,
      "learning_rate": 0.000432,
      "loss": 6.2246,
      "mean_token_accuracy": 0.1275457039475441,
      "num_tokens": 1783499.0,
      "step": 865
    },
    {
      "entropy": 6.531914615631104,
      "epoch": 0.7905497501135847,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004345,
      "loss": 6.4162,
      "mean_token_accuracy": 0.12034643739461899,
      "num_tokens": 1794107.0,
      "step": 870
    },
    {
      "entropy": 6.342921924591065,
      "epoch": 0.7950931394820536,
      "grad_norm": 1.25,
      "learning_rate": 0.000437,
      "loss": 6.2924,
      "mean_token_accuracy": 0.12306121364235878,
      "num_tokens": 1805151.0,
      "step": 875
    },
    {
      "entropy": 6.362142753601074,
      "epoch": 0.7996365288505225,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004395,
      "loss": 6.2709,
      "mean_token_accuracy": 0.12512775808572768,
      "num_tokens": 1815205.0,
      "step": 880
    },
    {
      "entropy": 6.332495164871216,
      "epoch": 0.8041799182189914,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000442,
      "loss": 6.2403,
      "mean_token_accuracy": 0.13088379427790642,
      "num_tokens": 1825301.0,
      "step": 885
    },
    {
      "entropy": 6.311787033081055,
      "epoch": 0.8087233075874602,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004445,
      "loss": 6.2579,
      "mean_token_accuracy": 0.12949514910578727,
      "num_tokens": 1835634.0,
      "step": 890
    },
    {
      "entropy": 6.42405252456665,
      "epoch": 0.8132666969559291,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000447,
      "loss": 6.2745,
      "mean_token_accuracy": 0.12240010797977448,
      "num_tokens": 1846855.0,
      "step": 895
    },
    {
      "entropy": 6.2657325744628904,
      "epoch": 0.817810086324398,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.2508,
      "mean_token_accuracy": 0.13143997862935067,
      "num_tokens": 1857747.0,
      "step": 900
    },
    {
      "entropy": 6.357445240020752,
      "epoch": 0.8223534756928669,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.3017,
      "mean_token_accuracy": 0.12524260133504866,
      "num_tokens": 1868098.0,
      "step": 905
    },
    {
      "entropy": 6.268481779098511,
      "epoch": 0.8268968650613358,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.2039,
      "mean_token_accuracy": 0.12827951461076736,
      "num_tokens": 1879682.0,
      "step": 910
    },
    {
      "entropy": 6.3160333156585695,
      "epoch": 0.8314402544298046,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.2286,
      "mean_token_accuracy": 0.12772295102477074,
      "num_tokens": 1890605.0,
      "step": 915
    },
    {
      "entropy": 6.30132942199707,
      "epoch": 0.8359836437982735,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.2498,
      "mean_token_accuracy": 0.13315400406718253,
      "num_tokens": 1900861.0,
      "step": 920
    },
    {
      "entropy": 6.333718204498291,
      "epoch": 0.8405270331667424,
      "grad_norm": 1.34375,
      "learning_rate": 0.000462,
      "loss": 6.2322,
      "mean_token_accuracy": 0.1283200278878212,
      "num_tokens": 1910867.0,
      "step": 925
    },
    {
      "entropy": 6.311747121810913,
      "epoch": 0.8450704225352113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004645,
      "loss": 6.2479,
      "mean_token_accuracy": 0.12160541415214539,
      "num_tokens": 1921424.0,
      "step": 930
    },
    {
      "entropy": 6.38282585144043,
      "epoch": 0.8496138119036801,
      "grad_norm": 1.203125,
      "learning_rate": 0.000467,
      "loss": 6.2312,
      "mean_token_accuracy": 0.13186774998903275,
      "num_tokens": 1930784.0,
      "step": 935
    },
    {
      "entropy": 6.240966749191284,
      "epoch": 0.854157201272149,
      "grad_norm": 1.125,
      "learning_rate": 0.0004695,
      "loss": 6.2259,
      "mean_token_accuracy": 0.1281438447535038,
      "num_tokens": 1940325.0,
      "step": 940
    },
    {
      "entropy": 6.260193490982056,
      "epoch": 0.8587005906406179,
      "grad_norm": 1.1875,
      "learning_rate": 0.000472,
      "loss": 6.238,
      "mean_token_accuracy": 0.12963936477899551,
      "num_tokens": 1951799.0,
      "step": 945
    },
    {
      "entropy": 6.3922779083251955,
      "epoch": 0.8632439800090868,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004745,
      "loss": 6.229,
      "mean_token_accuracy": 0.12269769310951233,
      "num_tokens": 1962311.0,
      "step": 950
    },
    {
      "entropy": 6.225095224380493,
      "epoch": 0.8677873693775556,
      "grad_norm": 1.078125,
      "learning_rate": 0.000477,
      "loss": 6.2281,
      "mean_token_accuracy": 0.1275744341313839,
      "num_tokens": 1974128.0,
      "step": 955
    },
    {
      "entropy": 6.168290805816651,
      "epoch": 0.8723307587460245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004795,
      "loss": 6.1582,
      "mean_token_accuracy": 0.1310737431049347,
      "num_tokens": 1983765.0,
      "step": 960
    },
    {
      "entropy": 6.315899515151978,
      "epoch": 0.8768741481144934,
      "grad_norm": 1.15625,
      "learning_rate": 0.000482,
      "loss": 6.235,
      "mean_token_accuracy": 0.12682206705212593,
      "num_tokens": 1994045.0,
      "step": 965
    },
    {
      "entropy": 6.336924314498901,
      "epoch": 0.8814175374829623,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004845,
      "loss": 6.3221,
      "mean_token_accuracy": 0.12170391902327538,
      "num_tokens": 2005353.0,
      "step": 970
    },
    {
      "entropy": 6.108946228027344,
      "epoch": 0.8859609268514311,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000487,
      "loss": 6.1282,
      "mean_token_accuracy": 0.13348021805286409,
      "num_tokens": 2015808.0,
      "step": 975
    },
    {
      "entropy": 6.315927267074585,
      "epoch": 0.8905043162199,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004895,
      "loss": 6.1734,
      "mean_token_accuracy": 0.13906338140368463,
      "num_tokens": 2025824.0,
      "step": 980
    },
    {
      "entropy": 6.163518285751342,
      "epoch": 0.8950477055883689,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000492,
      "loss": 6.197,
      "mean_token_accuracy": 0.1368340127170086,
      "num_tokens": 2036924.0,
      "step": 985
    },
    {
      "entropy": 6.243646335601807,
      "epoch": 0.8995910949568378,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004945,
      "loss": 6.228,
      "mean_token_accuracy": 0.13041988983750344,
      "num_tokens": 2047302.0,
      "step": 990
    },
    {
      "entropy": 6.261142539978027,
      "epoch": 0.9041344843253066,
      "grad_norm": 1.203125,
      "learning_rate": 0.000497,
      "loss": 6.2047,
      "mean_token_accuracy": 0.12715124264359473,
      "num_tokens": 2057972.0,
      "step": 995
    },
    {
      "entropy": 6.142115783691406,
      "epoch": 0.9086778736937755,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995,
      "loss": 6.0515,
      "mean_token_accuracy": 0.1383442610502243,
      "num_tokens": 2067916.0,
      "step": 1000
    },
    {
      "epoch": 0.9086778736937755,
      "eval_entropy": 6.119013550115186,
      "eval_loss": 6.310596466064453,
      "eval_mean_token_accuracy": 0.13073905674350816,
      "eval_num_tokens": 2067916.0,
      "eval_runtime": 2.5901,
      "eval_samples_per_second": 1326.19,
      "eval_steps_per_second": 166.015,
      "step": 1000
    },
    {
      "entropy": 6.095190238952637,
      "epoch": 0.9132212630622444,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004999998223471442,
      "loss": 6.2426,
      "mean_token_accuracy": 0.13274624422192574,
      "num_tokens": 2079119.0,
      "step": 1005
    },
    {
      "entropy": 6.302926826477051,
      "epoch": 0.9177646524307133,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999991006328981,
      "loss": 6.1263,
      "mean_token_accuracy": 0.1312612734735012,
      "num_tokens": 2090080.0,
      "step": 1010
    },
    {
      "entropy": 6.224363374710083,
      "epoch": 0.9223080417991822,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999978237557377,
      "loss": 6.1623,
      "mean_token_accuracy": 0.13245146051049234,
      "num_tokens": 2100541.0,
      "step": 1015
    },
    {
      "entropy": 6.144991064071656,
      "epoch": 0.9268514311676511,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999959917188137,
      "loss": 6.1332,
      "mean_token_accuracy": 0.1294662743806839,
      "num_tokens": 2111213.0,
      "step": 1020
    },
    {
      "entropy": 6.290520286560058,
      "epoch": 0.93139482053612,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999936045266462,
      "loss": 6.2263,
      "mean_token_accuracy": 0.1313753105700016,
      "num_tokens": 2121714.0,
      "step": 1025
    },
    {
      "entropy": 6.139681434631347,
      "epoch": 0.9359382099045889,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004999906621851255,
      "loss": 6.11,
      "mean_token_accuracy": 0.12752713337540628,
      "num_tokens": 2131226.0,
      "step": 1030
    },
    {
      "entropy": 6.254400205612183,
      "epoch": 0.9404815992730577,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999871647015115,
      "loss": 6.1907,
      "mean_token_accuracy": 0.12505661994218825,
      "num_tokens": 2143179.0,
      "step": 1035
    },
    {
      "entropy": 6.165806579589844,
      "epoch": 0.9450249886415266,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999831120844339,
      "loss": 6.1733,
      "mean_token_accuracy": 0.13424054831266402,
      "num_tokens": 2153646.0,
      "step": 1040
    },
    {
      "entropy": 6.0638185977935795,
      "epoch": 0.9495683780099955,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999785043438922,
      "loss": 6.1202,
      "mean_token_accuracy": 0.1302776090800762,
      "num_tokens": 2163267.0,
      "step": 1045
    },
    {
      "entropy": 6.2338563919067385,
      "epoch": 0.9541117673784644,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999733414912555,
      "loss": 6.024,
      "mean_token_accuracy": 0.13413071036338806,
      "num_tokens": 2173282.0,
      "step": 1050
    },
    {
      "entropy": 6.159117269515991,
      "epoch": 0.9586551567469332,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999676235392626,
      "loss": 6.1316,
      "mean_token_accuracy": 0.12614955380558968,
      "num_tokens": 2184202.0,
      "step": 1055
    },
    {
      "entropy": 6.249439859390259,
      "epoch": 0.9631985461154021,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999613505020219,
      "loss": 6.1008,
      "mean_token_accuracy": 0.13505923077464105,
      "num_tokens": 2194342.0,
      "step": 1060
    },
    {
      "entropy": 6.107129049301148,
      "epoch": 0.967741935483871,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999545223950116,
      "loss": 6.1139,
      "mean_token_accuracy": 0.13676788732409478,
      "num_tokens": 2204817.0,
      "step": 1065
    },
    {
      "entropy": 6.196682405471802,
      "epoch": 0.9722853248523399,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999471392350795,
      "loss": 6.1875,
      "mean_token_accuracy": 0.1350012630224228,
      "num_tokens": 2215281.0,
      "step": 1070
    },
    {
      "entropy": 6.2491254806518555,
      "epoch": 0.9768287142208087,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999392010404426,
      "loss": 6.1817,
      "mean_token_accuracy": 0.13481876030564308,
      "num_tokens": 2225784.0,
      "step": 1075
    },
    {
      "entropy": 6.119947338104248,
      "epoch": 0.9813721035892776,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999307078306876,
      "loss": 6.1209,
      "mean_token_accuracy": 0.13831015974283217,
      "num_tokens": 2235357.0,
      "step": 1080
    },
    {
      "entropy": 6.250767660140991,
      "epoch": 0.9859154929577465,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999216596267708,
      "loss": 6.1407,
      "mean_token_accuracy": 0.13157120868563651,
      "num_tokens": 2246438.0,
      "step": 1085
    },
    {
      "entropy": 6.163524055480957,
      "epoch": 0.9904588823262154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999120564510177,
      "loss": 6.1279,
      "mean_token_accuracy": 0.13285833448171616,
      "num_tokens": 2257528.0,
      "step": 1090
    },
    {
      "entropy": 6.125581979751587,
      "epoch": 0.9950022716946842,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499901898327123,
      "loss": 6.1579,
      "mean_token_accuracy": 0.13823469057679177,
      "num_tokens": 2267572.0,
      "step": 1095
    },
    {
      "entropy": 6.111034965515136,
      "epoch": 0.9995456610631531,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998911852801512,
      "loss": 6.0791,
      "mean_token_accuracy": 0.13840872496366502,
      "num_tokens": 2278520.0,
      "step": 1100
    },
    {
      "entropy": 6.206923537784153,
      "epoch": 1.003634711494775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998799173365355,
      "loss": 5.9924,
      "mean_token_accuracy": 0.138457167479727,
      "num_tokens": 2286991.0,
      "step": 1105
    },
    {
      "entropy": 6.174563884735107,
      "epoch": 1.008178100863244,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998680945240783,
      "loss": 6.0311,
      "mean_token_accuracy": 0.13416522592306138,
      "num_tokens": 2297857.0,
      "step": 1110
    },
    {
      "entropy": 6.178468990325928,
      "epoch": 1.0127214902317128,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998557168719513,
      "loss": 6.0813,
      "mean_token_accuracy": 0.13441063985228538,
      "num_tokens": 2308621.0,
      "step": 1115
    },
    {
      "entropy": 6.059679460525513,
      "epoch": 1.0172648796001817,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998427844106954,
      "loss": 5.881,
      "mean_token_accuracy": 0.14716010466217994,
      "num_tokens": 2318654.0,
      "step": 1120
    },
    {
      "entropy": 6.125048398971558,
      "epoch": 1.0218082689686505,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998292971722197,
      "loss": 5.9387,
      "mean_token_accuracy": 0.13893765211105347,
      "num_tokens": 2329064.0,
      "step": 1125
    },
    {
      "entropy": 6.071296548843383,
      "epoch": 1.0263516583371195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499815255189803,
      "loss": 5.9926,
      "mean_token_accuracy": 0.1351088397204876,
      "num_tokens": 2339843.0,
      "step": 1130
    },
    {
      "entropy": 6.1580180644989015,
      "epoch": 1.0308950477055883,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998006584980922,
      "loss": 5.9417,
      "mean_token_accuracy": 0.13985907807946205,
      "num_tokens": 2350715.0,
      "step": 1135
    },
    {
      "entropy": 6.097461557388305,
      "epoch": 1.0354384370740572,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997855071331034,
      "loss": 5.905,
      "mean_token_accuracy": 0.14022022858262062,
      "num_tokens": 2360635.0,
      "step": 1140
    },
    {
      "entropy": 6.042455005645752,
      "epoch": 1.039981826442526,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997698011322211,
      "loss": 5.9263,
      "mean_token_accuracy": 0.13829426020383834,
      "num_tokens": 2370627.0,
      "step": 1145
    },
    {
      "entropy": 6.115921592712402,
      "epoch": 1.044525215810995,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499753540534198,
      "loss": 5.9589,
      "mean_token_accuracy": 0.1388851523399353,
      "num_tokens": 2381790.0,
      "step": 1150
    },
    {
      "entropy": 5.961507940292359,
      "epoch": 1.049068605179464,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499736725379156,
      "loss": 5.8937,
      "mean_token_accuracy": 0.14004927650094032,
      "num_tokens": 2392685.0,
      "step": 1155
    },
    {
      "entropy": 6.0807301044464115,
      "epoch": 1.0536119945479328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997193557085844,
      "loss": 5.9553,
      "mean_token_accuracy": 0.13977984860539436,
      "num_tokens": 2403722.0,
      "step": 1160
    },
    {
      "entropy": 6.059891986846924,
      "epoch": 1.0581553839164017,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997014315653413,
      "loss": 6.0031,
      "mean_token_accuracy": 0.13041603118181228,
      "num_tokens": 2415311.0,
      "step": 1165
    },
    {
      "entropy": 6.061007785797119,
      "epoch": 1.0626987732848705,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996829529936527,
      "loss": 5.8757,
      "mean_token_accuracy": 0.14463189542293547,
      "num_tokens": 2425116.0,
      "step": 1170
    },
    {
      "entropy": 6.099240064620972,
      "epoch": 1.0672421626533395,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996639200391126,
      "loss": 6.0665,
      "mean_token_accuracy": 0.1358228489756584,
      "num_tokens": 2435916.0,
      "step": 1175
    },
    {
      "entropy": 5.997359848022461,
      "epoch": 1.0717855520218083,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996443327486831,
      "loss": 5.9479,
      "mean_token_accuracy": 0.13406080305576323,
      "num_tokens": 2446920.0,
      "step": 1180
    },
    {
      "entropy": 6.106053638458252,
      "epoch": 1.0763289413902772,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996241911706937,
      "loss": 5.8986,
      "mean_token_accuracy": 0.14392359107732772,
      "num_tokens": 2456367.0,
      "step": 1185
    },
    {
      "entropy": 5.986075735092163,
      "epoch": 1.080872330758746,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996034953548418,
      "loss": 5.8625,
      "mean_token_accuracy": 0.14260815605521202,
      "num_tokens": 2466790.0,
      "step": 1190
    },
    {
      "entropy": 6.0295631885528564,
      "epoch": 1.085415720127215,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004995822453521924,
      "loss": 5.874,
      "mean_token_accuracy": 0.14739150553941727,
      "num_tokens": 2476419.0,
      "step": 1195
    },
    {
      "entropy": 6.042729234695434,
      "epoch": 1.0899591094956838,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995604412151776,
      "loss": 6.0248,
      "mean_token_accuracy": 0.1344558708369732,
      "num_tokens": 2487295.0,
      "step": 1200
    },
    {
      "entropy": 6.019842672348022,
      "epoch": 1.0945024988641527,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499538082997597,
      "loss": 5.8945,
      "mean_token_accuracy": 0.13653064966201783,
      "num_tokens": 2497545.0,
      "step": 1205
    },
    {
      "entropy": 6.14437952041626,
      "epoch": 1.0990458882326215,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995151707546172,
      "loss": 6.0565,
      "mean_token_accuracy": 0.13368018791079522,
      "num_tokens": 2507805.0,
      "step": 1210
    },
    {
      "entropy": 6.041342830657959,
      "epoch": 1.1035892776010905,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994917045427721,
      "loss": 5.9199,
      "mean_token_accuracy": 0.1381150558590889,
      "num_tokens": 2518474.0,
      "step": 1215
    },
    {
      "entropy": 6.150459957122803,
      "epoch": 1.1081326669695593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994676844199621,
      "loss": 5.9511,
      "mean_token_accuracy": 0.14577605128288268,
      "num_tokens": 2528974.0,
      "step": 1220
    },
    {
      "entropy": 6.016877841949463,
      "epoch": 1.1126760563380282,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994431104454543,
      "loss": 6.0097,
      "mean_token_accuracy": 0.13516863361001014,
      "num_tokens": 2540675.0,
      "step": 1225
    },
    {
      "entropy": 6.028059244155884,
      "epoch": 1.117219445706497,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994179826798828,
      "loss": 5.8699,
      "mean_token_accuracy": 0.14476859122514724,
      "num_tokens": 2551367.0,
      "step": 1230
    },
    {
      "entropy": 5.999497795104981,
      "epoch": 1.121762835074966,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993923011852477,
      "loss": 5.9653,
      "mean_token_accuracy": 0.13774474412202836,
      "num_tokens": 2562096.0,
      "step": 1235
    },
    {
      "entropy": 6.10791974067688,
      "epoch": 1.1263062244434348,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993660660249158,
      "loss": 6.0367,
      "mean_token_accuracy": 0.13786499425768853,
      "num_tokens": 2572917.0,
      "step": 1240
    },
    {
      "entropy": 5.983006000518799,
      "epoch": 1.1308496138119037,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993392772636194,
      "loss": 5.8634,
      "mean_token_accuracy": 0.14501798897981644,
      "num_tokens": 2582971.0,
      "step": 1245
    },
    {
      "entropy": 6.108616685867309,
      "epoch": 1.1353930031803725,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993119349674573,
      "loss": 6.0225,
      "mean_token_accuracy": 0.13913898766040803,
      "num_tokens": 2593186.0,
      "step": 1250
    },
    {
      "entropy": 6.026316690444946,
      "epoch": 1.1399363925488415,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004992840392038938,
      "loss": 6.018,
      "mean_token_accuracy": 0.1349714659154415,
      "num_tokens": 2604359.0,
      "step": 1255
    },
    {
      "entropy": 6.056495094299317,
      "epoch": 1.1444797819173103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499255590041759,
      "loss": 5.9367,
      "mean_token_accuracy": 0.14222330152988433,
      "num_tokens": 2614436.0,
      "step": 1260
    },
    {
      "entropy": 5.957977342605591,
      "epoch": 1.1490231712857792,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992265875512485,
      "loss": 5.898,
      "mean_token_accuracy": 0.14035152941942214,
      "num_tokens": 2624175.0,
      "step": 1265
    },
    {
      "entropy": 6.0870380878448485,
      "epoch": 1.153566560654248,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991970318039228,
      "loss": 5.9292,
      "mean_token_accuracy": 0.14449749886989594,
      "num_tokens": 2634898.0,
      "step": 1270
    },
    {
      "entropy": 6.137653446197509,
      "epoch": 1.158109950022717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991669228727078,
      "loss": 5.9626,
      "mean_token_accuracy": 0.14636341333389283,
      "num_tokens": 2645590.0,
      "step": 1275
    },
    {
      "entropy": 6.0029479503631595,
      "epoch": 1.1626533393911858,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991362608318947,
      "loss": 5.944,
      "mean_token_accuracy": 0.14702457636594773,
      "num_tokens": 2656115.0,
      "step": 1280
    },
    {
      "entropy": 6.0235066413879395,
      "epoch": 1.1671967287596547,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991050457571384,
      "loss": 5.9247,
      "mean_token_accuracy": 0.14117783308029175,
      "num_tokens": 2666760.0,
      "step": 1285
    },
    {
      "entropy": 6.054644346237183,
      "epoch": 1.1717401181281235,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990732777254594,
      "loss": 5.9022,
      "mean_token_accuracy": 0.14372125118970872,
      "num_tokens": 2676763.0,
      "step": 1290
    },
    {
      "entropy": 5.955284786224365,
      "epoch": 1.1762835074965925,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990409568152422,
      "loss": 5.9051,
      "mean_token_accuracy": 0.1380102217197418,
      "num_tokens": 2686598.0,
      "step": 1295
    },
    {
      "entropy": 6.0025769710540775,
      "epoch": 1.1808268968650613,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990080831062352,
      "loss": 5.8913,
      "mean_token_accuracy": 0.1458988979458809,
      "num_tokens": 2697767.0,
      "step": 1300
    },
    {
      "entropy": 5.992935132980347,
      "epoch": 1.1853702862335302,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989746566795512,
      "loss": 5.8749,
      "mean_token_accuracy": 0.14183612391352654,
      "num_tokens": 2707081.0,
      "step": 1305
    },
    {
      "entropy": 5.9335216045379635,
      "epoch": 1.189913675601999,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004989406776176665,
      "loss": 5.9001,
      "mean_token_accuracy": 0.14268443137407302,
      "num_tokens": 2717312.0,
      "step": 1310
    },
    {
      "entropy": 6.063012170791626,
      "epoch": 1.194457064970468,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498906146004421,
      "loss": 5.9208,
      "mean_token_accuracy": 0.14291029423475266,
      "num_tokens": 2728108.0,
      "step": 1315
    },
    {
      "entropy": 6.0912096977233885,
      "epoch": 1.1990004543389368,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004988710619250181,
      "loss": 6.1112,
      "mean_token_accuracy": 0.14074144288897514,
      "num_tokens": 2738860.0,
      "step": 1320
    },
    {
      "entropy": 6.000742721557617,
      "epoch": 1.2035438437074057,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988354254660244,
      "loss": 5.9563,
      "mean_token_accuracy": 0.14386919811367987,
      "num_tokens": 2749786.0,
      "step": 1325
    },
    {
      "entropy": 6.006845712661743,
      "epoch": 1.2080872330758745,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987992367153691,
      "loss": 5.8626,
      "mean_token_accuracy": 0.14512528330087662,
      "num_tokens": 2760018.0,
      "step": 1330
    },
    {
      "entropy": 5.931078481674194,
      "epoch": 1.2126306224443435,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987624957623443,
      "loss": 5.9177,
      "mean_token_accuracy": 0.14297755360603331,
      "num_tokens": 2770572.0,
      "step": 1335
    },
    {
      "entropy": 5.937653732299805,
      "epoch": 1.2171740118128123,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987252026976049,
      "loss": 5.8826,
      "mean_token_accuracy": 0.15135369300842286,
      "num_tokens": 2780297.0,
      "step": 1340
    },
    {
      "entropy": 6.039012765884399,
      "epoch": 1.2217174011812812,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986873576131676,
      "loss": 5.9992,
      "mean_token_accuracy": 0.14602920040488243,
      "num_tokens": 2790454.0,
      "step": 1345
    },
    {
      "entropy": 6.017839241027832,
      "epoch": 1.22626079054975,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986489606024116,
      "loss": 5.85,
      "mean_token_accuracy": 0.14553693532943726,
      "num_tokens": 2800034.0,
      "step": 1350
    },
    {
      "entropy": 5.885445594787598,
      "epoch": 1.230804179918219,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986100117600776,
      "loss": 5.864,
      "mean_token_accuracy": 0.14794481843709945,
      "num_tokens": 2810157.0,
      "step": 1355
    },
    {
      "entropy": 5.973232746124268,
      "epoch": 1.235347569286688,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498570511182268,
      "loss": 5.8857,
      "mean_token_accuracy": 0.1394151568412781,
      "num_tokens": 2819404.0,
      "step": 1360
    },
    {
      "entropy": 5.941764545440674,
      "epoch": 1.2398909586551567,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004985304589664465,
      "loss": 5.8298,
      "mean_token_accuracy": 0.1434129849076271,
      "num_tokens": 2829551.0,
      "step": 1365
    },
    {
      "entropy": 5.947914123535156,
      "epoch": 1.2444343480236255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004984898552114381,
      "loss": 5.8903,
      "mean_token_accuracy": 0.14327759444713592,
      "num_tokens": 2840381.0,
      "step": 1370
    },
    {
      "entropy": 5.982251596450806,
      "epoch": 1.2489777373920945,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004984487000174283,
      "loss": 5.8865,
      "mean_token_accuracy": 0.13880339488387108,
      "num_tokens": 2850279.0,
      "step": 1375
    },
    {
      "entropy": 6.027106142044067,
      "epoch": 1.2535211267605635,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984069934859637,
      "loss": 5.9153,
      "mean_token_accuracy": 0.150995684415102,
      "num_tokens": 2860209.0,
      "step": 1380
    },
    {
      "entropy": 5.9328803539276125,
      "epoch": 1.2580645161290323,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004983647357199509,
      "loss": 5.9148,
      "mean_token_accuracy": 0.14855456426739694,
      "num_tokens": 2870932.0,
      "step": 1385
    },
    {
      "entropy": 6.015257740020752,
      "epoch": 1.262607905497501,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983219268236567,
      "loss": 5.9053,
      "mean_token_accuracy": 0.14229729101061822,
      "num_tokens": 2881139.0,
      "step": 1390
    },
    {
      "entropy": 5.992234373092652,
      "epoch": 1.26715129486597,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004982785669027079,
      "loss": 5.893,
      "mean_token_accuracy": 0.14636517614126204,
      "num_tokens": 2890774.0,
      "step": 1395
    },
    {
      "entropy": 5.978844785690308,
      "epoch": 1.271694684234439,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982346560640907,
      "loss": 5.8629,
      "mean_token_accuracy": 0.1498524270951748,
      "num_tokens": 2900259.0,
      "step": 1400
    },
    {
      "entropy": 5.957555437088013,
      "epoch": 1.2762380736029078,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004981901944161507,
      "loss": 5.8767,
      "mean_token_accuracy": 0.1461343213915825,
      "num_tokens": 2911365.0,
      "step": 1405
    },
    {
      "entropy": 5.930820655822754,
      "epoch": 1.2807814629713765,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004981451820685928,
      "loss": 5.8679,
      "mean_token_accuracy": 0.1443587377667427,
      "num_tokens": 2921189.0,
      "step": 1410
    },
    {
      "entropy": 5.872604942321777,
      "epoch": 1.2853248523398455,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004980996191324802,
      "loss": 5.8969,
      "mean_token_accuracy": 0.1431989386677742,
      "num_tokens": 2931271.0,
      "step": 1415
    },
    {
      "entropy": 6.04894380569458,
      "epoch": 1.2898682417083145,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004980535057202351,
      "loss": 5.8929,
      "mean_token_accuracy": 0.14426368549466134,
      "num_tokens": 2941557.0,
      "step": 1420
    },
    {
      "entropy": 5.822301912307739,
      "epoch": 1.2944116310767833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004980068419456378,
      "loss": 5.8086,
      "mean_token_accuracy": 0.15275876671075822,
      "num_tokens": 2951305.0,
      "step": 1425
    },
    {
      "entropy": 5.909662580490112,
      "epoch": 1.2989550204452522,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004979596279238265,
      "loss": 5.8986,
      "mean_token_accuracy": 0.15126647800207138,
      "num_tokens": 2961948.0,
      "step": 1430
    },
    {
      "entropy": 5.9860834121704105,
      "epoch": 1.303498409813721,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004979118637712969,
      "loss": 5.9588,
      "mean_token_accuracy": 0.13625321835279464,
      "num_tokens": 2973300.0,
      "step": 1435
    },
    {
      "entropy": 5.997512292861939,
      "epoch": 1.30804179918219,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004978635496059025,
      "loss": 5.9007,
      "mean_token_accuracy": 0.14264205247163772,
      "num_tokens": 2984114.0,
      "step": 1440
    },
    {
      "entropy": 5.804658126831055,
      "epoch": 1.3125851885506588,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004978146855468537,
      "loss": 5.7665,
      "mean_token_accuracy": 0.1514141008257866,
      "num_tokens": 2993940.0,
      "step": 1445
    },
    {
      "entropy": 5.981986856460571,
      "epoch": 1.3171285779191277,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004977652717147177,
      "loss": 5.8833,
      "mean_token_accuracy": 0.14869886934757232,
      "num_tokens": 3004093.0,
      "step": 1450
    },
    {
      "entropy": 5.826439237594604,
      "epoch": 1.3216719672875965,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004977153082314182,
      "loss": 5.8068,
      "mean_token_accuracy": 0.14922491908073426,
      "num_tokens": 3013537.0,
      "step": 1455
    },
    {
      "entropy": 5.9768884658813475,
      "epoch": 1.3262153566560655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000497664795220235,
      "loss": 5.8106,
      "mean_token_accuracy": 0.14891566783189775,
      "num_tokens": 3023502.0,
      "step": 1460
    },
    {
      "entropy": 5.868037366867066,
      "epoch": 1.3307587460245343,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976137328058042,
      "loss": 5.8706,
      "mean_token_accuracy": 0.1440548375248909,
      "num_tokens": 3033910.0,
      "step": 1465
    },
    {
      "entropy": 5.910422515869141,
      "epoch": 1.3353021353930032,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000497562121114117,
      "loss": 5.8994,
      "mean_token_accuracy": 0.14917545020580292,
      "num_tokens": 3043500.0,
      "step": 1470
    },
    {
      "entropy": 5.915622234344482,
      "epoch": 1.339845524761472,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004975099602725202,
      "loss": 5.8788,
      "mean_token_accuracy": 0.14382385686039925,
      "num_tokens": 3054776.0,
      "step": 1475
    },
    {
      "entropy": 5.952015781402588,
      "epoch": 1.344388914129941,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004974572504097154,
      "loss": 5.844,
      "mean_token_accuracy": 0.14768940806388856,
      "num_tokens": 3065403.0,
      "step": 1480
    },
    {
      "entropy": 5.887816143035889,
      "epoch": 1.3489323034984098,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004974039916557591,
      "loss": 5.8759,
      "mean_token_accuracy": 0.1451573684811592,
      "num_tokens": 3074661.0,
      "step": 1485
    },
    {
      "entropy": 5.878644323348999,
      "epoch": 1.3534756928668787,
      "grad_norm": 1.21875,
      "learning_rate": 0.000497350184142062,
      "loss": 5.8337,
      "mean_token_accuracy": 0.1441040262579918,
      "num_tokens": 3085319.0,
      "step": 1490
    },
    {
      "entropy": 5.879520559310913,
      "epoch": 1.3580190822353475,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004972958280013886,
      "loss": 5.8288,
      "mean_token_accuracy": 0.14350272119045257,
      "num_tokens": 3095115.0,
      "step": 1495
    },
    {
      "entropy": 5.940704250335694,
      "epoch": 1.3625624716038165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004972409233678575,
      "loss": 6.0048,
      "mean_token_accuracy": 0.1379089266061783,
      "num_tokens": 3104841.0,
      "step": 1500
    },
    {
      "epoch": 1.3625624716038165,
      "eval_entropy": 5.845436406690021,
      "eval_loss": 6.132750034332275,
      "eval_mean_token_accuracy": 0.14173085743258165,
      "eval_num_tokens": 3104841.0,
      "eval_runtime": 2.6014,
      "eval_samples_per_second": 1320.451,
      "eval_steps_per_second": 165.297,
      "step": 1500
    },
    {
      "entropy": 5.94781379699707,
      "epoch": 1.3671058609722853,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004971854703769401,
      "loss": 5.865,
      "mean_token_accuracy": 0.1451248086988926,
      "num_tokens": 3115098.0,
      "step": 1505
    },
    {
      "entropy": 5.878427028656006,
      "epoch": 1.3716492503407542,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004971294691654616,
      "loss": 5.7491,
      "mean_token_accuracy": 0.15491948276758194,
      "num_tokens": 3124831.0,
      "step": 1510
    },
    {
      "entropy": 5.854487037658691,
      "epoch": 1.376192639709223,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000497072919871599,
      "loss": 5.9248,
      "mean_token_accuracy": 0.14528753682971002,
      "num_tokens": 3136113.0,
      "step": 1515
    },
    {
      "entropy": 6.086994552612305,
      "epoch": 1.380736029077692,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004970158226348824,
      "loss": 5.839,
      "mean_token_accuracy": 0.15899223536252977,
      "num_tokens": 3145617.0,
      "step": 1520
    },
    {
      "entropy": 5.833227968215942,
      "epoch": 1.3852794184461608,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004969581775961932,
      "loss": 5.9398,
      "mean_token_accuracy": 0.13940114825963973,
      "num_tokens": 3156913.0,
      "step": 1525
    },
    {
      "entropy": 5.977892684936523,
      "epoch": 1.3898228078146297,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004968999848977651,
      "loss": 5.8803,
      "mean_token_accuracy": 0.14702537804841995,
      "num_tokens": 3167506.0,
      "step": 1530
    },
    {
      "entropy": 5.819863986968994,
      "epoch": 1.3943661971830985,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004968412446831827,
      "loss": 5.759,
      "mean_token_accuracy": 0.14905699491500854,
      "num_tokens": 3177956.0,
      "step": 1535
    },
    {
      "entropy": 5.9371013164520265,
      "epoch": 1.3989095865515675,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004967819570973816,
      "loss": 6.0081,
      "mean_token_accuracy": 0.14665073454380034,
      "num_tokens": 3189172.0,
      "step": 1540
    },
    {
      "entropy": 6.025431394577026,
      "epoch": 1.4034529759200365,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000496722122286648,
      "loss": 5.9232,
      "mean_token_accuracy": 0.14157174676656722,
      "num_tokens": 3200188.0,
      "step": 1545
    },
    {
      "entropy": 5.811288738250733,
      "epoch": 1.4079963652885052,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004966617403986185,
      "loss": 5.7726,
      "mean_token_accuracy": 0.14368995130062104,
      "num_tokens": 3210171.0,
      "step": 1550
    },
    {
      "entropy": 6.013456678390503,
      "epoch": 1.412539754656974,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004966008115822793,
      "loss": 5.9925,
      "mean_token_accuracy": 0.14135242402553558,
      "num_tokens": 3221201.0,
      "step": 1555
    },
    {
      "entropy": 5.860878372192383,
      "epoch": 1.417083144025443,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004965393359879662,
      "loss": 5.8405,
      "mean_token_accuracy": 0.1457865744829178,
      "num_tokens": 3231825.0,
      "step": 1560
    },
    {
      "entropy": 6.031340217590332,
      "epoch": 1.421626533393912,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004964773137673641,
      "loss": 6.0398,
      "mean_token_accuracy": 0.13429115936160088,
      "num_tokens": 3241860.0,
      "step": 1565
    },
    {
      "entropy": 5.798910617828369,
      "epoch": 1.4261699227623807,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004964147450735067,
      "loss": 5.6964,
      "mean_token_accuracy": 0.1542890965938568,
      "num_tokens": 3251893.0,
      "step": 1570
    },
    {
      "entropy": 5.867627334594727,
      "epoch": 1.4307133121308495,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000496351630060776,
      "loss": 5.8404,
      "mean_token_accuracy": 0.14542031958699225,
      "num_tokens": 3262748.0,
      "step": 1575
    },
    {
      "entropy": 6.006178951263427,
      "epoch": 1.4352567014993185,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004962879688849021,
      "loss": 5.9546,
      "mean_token_accuracy": 0.14313093349337577,
      "num_tokens": 3273035.0,
      "step": 1580
    },
    {
      "entropy": 5.92848391532898,
      "epoch": 1.4398000908677875,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004962237617029625,
      "loss": 5.8831,
      "mean_token_accuracy": 0.14664052724838256,
      "num_tokens": 3283413.0,
      "step": 1585
    },
    {
      "entropy": 5.916056871414185,
      "epoch": 1.4443434802362562,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961590086733821,
      "loss": 5.8681,
      "mean_token_accuracy": 0.147178016602993,
      "num_tokens": 3294628.0,
      "step": 1590
    },
    {
      "entropy": 5.8341459274292,
      "epoch": 1.448886869604725,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004960937099559326,
      "loss": 5.8509,
      "mean_token_accuracy": 0.15535128712654114,
      "num_tokens": 3304631.0,
      "step": 1595
    },
    {
      "entropy": 5.913455057144165,
      "epoch": 1.453430258973194,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960278657117321,
      "loss": 5.8617,
      "mean_token_accuracy": 0.14451269209384918,
      "num_tokens": 3315481.0,
      "step": 1600
    },
    {
      "entropy": 5.88686203956604,
      "epoch": 1.457973648341663,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004959614761032446,
      "loss": 5.8135,
      "mean_token_accuracy": 0.147933691740036,
      "num_tokens": 3326493.0,
      "step": 1605
    },
    {
      "entropy": 6.0231451988220215,
      "epoch": 1.4625170377101318,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049589454129428,
      "loss": 5.9445,
      "mean_token_accuracy": 0.1462418481707573,
      "num_tokens": 3338358.0,
      "step": 1610
    },
    {
      "entropy": 5.90168080329895,
      "epoch": 1.4670604270786005,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004958270614499932,
      "loss": 5.9121,
      "mean_token_accuracy": 0.14460725635290145,
      "num_tokens": 3347961.0,
      "step": 1615
    },
    {
      "entropy": 5.902929592132568,
      "epoch": 1.4716038164470695,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004957590367368841,
      "loss": 5.8387,
      "mean_token_accuracy": 0.14820509701967238,
      "num_tokens": 3358544.0,
      "step": 1620
    },
    {
      "entropy": 5.889368629455566,
      "epoch": 1.4761472058155385,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004956904673227968,
      "loss": 5.8445,
      "mean_token_accuracy": 0.14531347751617432,
      "num_tokens": 3368233.0,
      "step": 1625
    },
    {
      "entropy": 5.820783710479736,
      "epoch": 1.4806905951840073,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004956213533769195,
      "loss": 5.8111,
      "mean_token_accuracy": 0.1510622039437294,
      "num_tokens": 3379065.0,
      "step": 1630
    },
    {
      "entropy": 5.984694337844848,
      "epoch": 1.485233984552476,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004955516950697843,
      "loss": 5.8971,
      "mean_token_accuracy": 0.139473906904459,
      "num_tokens": 3389251.0,
      "step": 1635
    },
    {
      "entropy": 5.882978105545044,
      "epoch": 1.489777373920945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004954814925732657,
      "loss": 5.826,
      "mean_token_accuracy": 0.15168372839689254,
      "num_tokens": 3399660.0,
      "step": 1640
    },
    {
      "entropy": 5.889898157119751,
      "epoch": 1.494320763289414,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004954107460605816,
      "loss": 5.8601,
      "mean_token_accuracy": 0.1534946471452713,
      "num_tokens": 3409395.0,
      "step": 1645
    },
    {
      "entropy": 5.963047361373901,
      "epoch": 1.4988641526578828,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004953394557062922,
      "loss": 5.8645,
      "mean_token_accuracy": 0.14099805802106857,
      "num_tokens": 3419881.0,
      "step": 1650
    },
    {
      "entropy": 5.847076463699341,
      "epoch": 1.5034075420263515,
      "grad_norm": 1.0,
      "learning_rate": 0.0004952676216862989,
      "loss": 5.8645,
      "mean_token_accuracy": 0.14429206624627114,
      "num_tokens": 3430316.0,
      "step": 1655
    },
    {
      "entropy": 5.860261344909668,
      "epoch": 1.5079509313948205,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004951952441778453,
      "loss": 5.7421,
      "mean_token_accuracy": 0.1556067496538162,
      "num_tokens": 3440174.0,
      "step": 1660
    },
    {
      "entropy": 5.8579755306243895,
      "epoch": 1.5124943207632895,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004951223233595158,
      "loss": 5.9341,
      "mean_token_accuracy": 0.14123850762844087,
      "num_tokens": 3451141.0,
      "step": 1665
    },
    {
      "entropy": 5.965538597106933,
      "epoch": 1.5170377101317583,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000495048859411235,
      "loss": 5.8206,
      "mean_token_accuracy": 0.14651950299739838,
      "num_tokens": 3461767.0,
      "step": 1670
    },
    {
      "entropy": 5.793015766143799,
      "epoch": 1.521581099500227,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004949748525142681,
      "loss": 5.7692,
      "mean_token_accuracy": 0.15745436996221543,
      "num_tokens": 3470937.0,
      "step": 1675
    },
    {
      "entropy": 5.909920597076416,
      "epoch": 1.526124488868696,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004949003028512195,
      "loss": 5.8406,
      "mean_token_accuracy": 0.14674706086516381,
      "num_tokens": 3482130.0,
      "step": 1680
    },
    {
      "entropy": 5.864456653594971,
      "epoch": 1.530667878237165,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004948252106060334,
      "loss": 5.9126,
      "mean_token_accuracy": 0.1415312610566616,
      "num_tokens": 3491992.0,
      "step": 1685
    },
    {
      "entropy": 5.804475402832031,
      "epoch": 1.5352112676056338,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004947495759639923,
      "loss": 5.7755,
      "mean_token_accuracy": 0.15035406947135926,
      "num_tokens": 3501437.0,
      "step": 1690
    },
    {
      "entropy": 5.762890481948853,
      "epoch": 1.5397546569741025,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004946733991117171,
      "loss": 5.8417,
      "mean_token_accuracy": 0.147023569047451,
      "num_tokens": 3511625.0,
      "step": 1695
    },
    {
      "entropy": 5.87570858001709,
      "epoch": 1.5442980463425715,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004945966802371668,
      "loss": 5.6182,
      "mean_token_accuracy": 0.15641522854566575,
      "num_tokens": 3520428.0,
      "step": 1700
    },
    {
      "entropy": 5.840044355392456,
      "epoch": 1.5488414357110405,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004945194195296374,
      "loss": 5.8491,
      "mean_token_accuracy": 0.14945719689130782,
      "num_tokens": 3532218.0,
      "step": 1705
    },
    {
      "entropy": 5.903447675704956,
      "epoch": 1.5533848250795095,
      "grad_norm": 1.03125,
      "learning_rate": 0.000494441617179762,
      "loss": 5.8271,
      "mean_token_accuracy": 0.14525461941957474,
      "num_tokens": 3543177.0,
      "step": 1710
    },
    {
      "entropy": 5.880182600021362,
      "epoch": 1.5579282144479782,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004943632733795104,
      "loss": 5.7948,
      "mean_token_accuracy": 0.15421123802661896,
      "num_tokens": 3552833.0,
      "step": 1715
    },
    {
      "entropy": 5.73713264465332,
      "epoch": 1.562471603816447,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004942843883221879,
      "loss": 5.8598,
      "mean_token_accuracy": 0.1434869036078453,
      "num_tokens": 3563558.0,
      "step": 1720
    },
    {
      "entropy": 5.838307952880859,
      "epoch": 1.567014993184916,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004942049622024357,
      "loss": 5.8033,
      "mean_token_accuracy": 0.15330324470996856,
      "num_tokens": 3574993.0,
      "step": 1725
    },
    {
      "entropy": 5.911986684799194,
      "epoch": 1.571558382553385,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004941249952162298,
      "loss": 5.8293,
      "mean_token_accuracy": 0.14535781294107436,
      "num_tokens": 3584586.0,
      "step": 1730
    },
    {
      "entropy": 5.7583925247192385,
      "epoch": 1.5761017719218537,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004940444875608807,
      "loss": 5.771,
      "mean_token_accuracy": 0.15141805708408357,
      "num_tokens": 3595029.0,
      "step": 1735
    },
    {
      "entropy": 5.890980005264282,
      "epoch": 1.5806451612903225,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004939634394350332,
      "loss": 5.84,
      "mean_token_accuracy": 0.14627694189548493,
      "num_tokens": 3606034.0,
      "step": 1740
    },
    {
      "entropy": 5.831475687026978,
      "epoch": 1.5851885506587915,
      "grad_norm": 1.125,
      "learning_rate": 0.0004938818510386654,
      "loss": 5.7677,
      "mean_token_accuracy": 0.150801981985569,
      "num_tokens": 3615551.0,
      "step": 1745
    },
    {
      "entropy": 5.829084348678589,
      "epoch": 1.5897319400272605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004937997225730886,
      "loss": 5.9243,
      "mean_token_accuracy": 0.1494157448410988,
      "num_tokens": 3626811.0,
      "step": 1750
    },
    {
      "entropy": 5.850355720520019,
      "epoch": 1.5942753293957292,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004937170542409464,
      "loss": 5.7805,
      "mean_token_accuracy": 0.14997649937868118,
      "num_tokens": 3637338.0,
      "step": 1755
    },
    {
      "entropy": 5.743219995498658,
      "epoch": 1.598818718764198,
      "grad_norm": 0.984375,
      "learning_rate": 0.000493633846246215,
      "loss": 5.7073,
      "mean_token_accuracy": 0.15153501853346824,
      "num_tokens": 3648650.0,
      "step": 1760
    },
    {
      "entropy": 5.89624662399292,
      "epoch": 1.603362108132667,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004935500987942018,
      "loss": 5.8664,
      "mean_token_accuracy": 0.14928599148988725,
      "num_tokens": 3659331.0,
      "step": 1765
    },
    {
      "entropy": 5.901072025299072,
      "epoch": 1.607905497501136,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004934658120915452,
      "loss": 5.93,
      "mean_token_accuracy": 0.1412065401673317,
      "num_tokens": 3670757.0,
      "step": 1770
    },
    {
      "entropy": 5.891441011428833,
      "epoch": 1.6124488868696047,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004933809863462142,
      "loss": 5.7722,
      "mean_token_accuracy": 0.15418703854084015,
      "num_tokens": 3681286.0,
      "step": 1775
    },
    {
      "entropy": 5.819612312316894,
      "epoch": 1.6169922762380735,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004932956217675082,
      "loss": 5.7627,
      "mean_token_accuracy": 0.15468500554561615,
      "num_tokens": 3690733.0,
      "step": 1780
    },
    {
      "entropy": 5.880888891220093,
      "epoch": 1.6215356656065425,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004932097185660555,
      "loss": 5.9492,
      "mean_token_accuracy": 0.14442042708396913,
      "num_tokens": 3701263.0,
      "step": 1785
    },
    {
      "entropy": 5.902997541427612,
      "epoch": 1.6260790549750115,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004931232769538138,
      "loss": 5.8574,
      "mean_token_accuracy": 0.14900370091199874,
      "num_tokens": 3711578.0,
      "step": 1790
    },
    {
      "entropy": 5.941886234283447,
      "epoch": 1.6306224443434802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004930362971440692,
      "loss": 5.8372,
      "mean_token_accuracy": 0.14781564697623253,
      "num_tokens": 3722140.0,
      "step": 1795
    },
    {
      "entropy": 5.696824550628662,
      "epoch": 1.635165833711949,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004929487793514358,
      "loss": 5.5599,
      "mean_token_accuracy": 0.15967873632907867,
      "num_tokens": 3732019.0,
      "step": 1800
    },
    {
      "entropy": 5.805211877822876,
      "epoch": 1.639709223080418,
      "grad_norm": 1.140625,
      "learning_rate": 0.000492860723791855,
      "loss": 5.7642,
      "mean_token_accuracy": 0.15093351081013678,
      "num_tokens": 3741912.0,
      "step": 1805
    },
    {
      "entropy": 5.867282581329346,
      "epoch": 1.644252612448887,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004927721306825952,
      "loss": 5.7436,
      "mean_token_accuracy": 0.15467035323381423,
      "num_tokens": 3751365.0,
      "step": 1810
    },
    {
      "entropy": 5.756223678588867,
      "epoch": 1.6487960018173557,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004926830002422509,
      "loss": 5.7401,
      "mean_token_accuracy": 0.14856993556022643,
      "num_tokens": 3762160.0,
      "step": 1815
    },
    {
      "entropy": 5.841907358169555,
      "epoch": 1.6533393911858245,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492593332690743,
      "loss": 5.7751,
      "mean_token_accuracy": 0.15270450413227082,
      "num_tokens": 3772608.0,
      "step": 1820
    },
    {
      "entropy": 5.784197378158569,
      "epoch": 1.6578827805542935,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004925031282493167,
      "loss": 5.8244,
      "mean_token_accuracy": 0.14576919451355935,
      "num_tokens": 3783344.0,
      "step": 1825
    },
    {
      "entropy": 5.834315538406372,
      "epoch": 1.6624261699227625,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004924123871405428,
      "loss": 5.8314,
      "mean_token_accuracy": 0.1431602880358696,
      "num_tokens": 3794103.0,
      "step": 1830
    },
    {
      "entropy": 5.791322803497314,
      "epoch": 1.6669695592912313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004923211095883161,
      "loss": 5.7645,
      "mean_token_accuracy": 0.15026065930724145,
      "num_tokens": 3804537.0,
      "step": 1835
    },
    {
      "entropy": 5.845962476730347,
      "epoch": 1.6715129486597,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004922292958178546,
      "loss": 5.7828,
      "mean_token_accuracy": 0.14817606657743454,
      "num_tokens": 3814147.0,
      "step": 1840
    },
    {
      "entropy": 5.8800671100616455,
      "epoch": 1.676056338028169,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004921369460557,
      "loss": 5.8494,
      "mean_token_accuracy": 0.1477331094443798,
      "num_tokens": 3823640.0,
      "step": 1845
    },
    {
      "entropy": 5.768005895614624,
      "epoch": 1.680599727396638,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004920440605297159,
      "loss": 5.6941,
      "mean_token_accuracy": 0.1621052548289299,
      "num_tokens": 3833418.0,
      "step": 1850
    },
    {
      "entropy": 5.781481456756592,
      "epoch": 1.6851431167651068,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000491950639469088,
      "loss": 5.715,
      "mean_token_accuracy": 0.15754078924655915,
      "num_tokens": 3844998.0,
      "step": 1855
    },
    {
      "entropy": 5.709991407394409,
      "epoch": 1.6896865061335755,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004918566831043238,
      "loss": 5.6519,
      "mean_token_accuracy": 0.16085057556629181,
      "num_tokens": 3855549.0,
      "step": 1860
    },
    {
      "entropy": 5.748021173477173,
      "epoch": 1.6942298955020445,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000491762191667251,
      "loss": 5.77,
      "mean_token_accuracy": 0.14954300299286843,
      "num_tokens": 3865287.0,
      "step": 1865
    },
    {
      "entropy": 5.839926052093506,
      "epoch": 1.6987732848705135,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000491667165391018,
      "loss": 5.8156,
      "mean_token_accuracy": 0.15002627968788146,
      "num_tokens": 3875914.0,
      "step": 1870
    },
    {
      "entropy": 5.837830352783203,
      "epoch": 1.7033166742389823,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004915716045100927,
      "loss": 5.82,
      "mean_token_accuracy": 0.14646194875240326,
      "num_tokens": 3885196.0,
      "step": 1875
    },
    {
      "entropy": 5.858989715576172,
      "epoch": 1.707860063607451,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004914755092602621,
      "loss": 5.8509,
      "mean_token_accuracy": 0.14010663330554962,
      "num_tokens": 3895227.0,
      "step": 1880
    },
    {
      "entropy": 5.839261770248413,
      "epoch": 1.71240345297592,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004913788798786314,
      "loss": 5.7349,
      "mean_token_accuracy": 0.15189915597438813,
      "num_tokens": 3905926.0,
      "step": 1885
    },
    {
      "entropy": 5.73164644241333,
      "epoch": 1.716946842344389,
      "grad_norm": 1.125,
      "learning_rate": 0.0004912817166036243,
      "loss": 5.7775,
      "mean_token_accuracy": 0.1529722660779953,
      "num_tokens": 3916982.0,
      "step": 1890
    },
    {
      "entropy": 5.735852575302124,
      "epoch": 1.7214902317128578,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004911840196749813,
      "loss": 5.6888,
      "mean_token_accuracy": 0.15600222796201707,
      "num_tokens": 3926889.0,
      "step": 1895
    },
    {
      "entropy": 5.79035530090332,
      "epoch": 1.7260336210813265,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004910857893337602,
      "loss": 5.7173,
      "mean_token_accuracy": 0.1464507170021534,
      "num_tokens": 3936734.0,
      "step": 1900
    },
    {
      "entropy": 5.838219404220581,
      "epoch": 1.7305770104497955,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004909870258223342,
      "loss": 5.8571,
      "mean_token_accuracy": 0.1459001362323761,
      "num_tokens": 3948001.0,
      "step": 1905
    },
    {
      "entropy": 5.830196475982666,
      "epoch": 1.7351203998182645,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004908877293843926,
      "loss": 5.7328,
      "mean_token_accuracy": 0.14376608580350875,
      "num_tokens": 3957433.0,
      "step": 1910
    },
    {
      "entropy": 5.854920768737793,
      "epoch": 1.7396637891867333,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004907879002649395,
      "loss": 5.8053,
      "mean_token_accuracy": 0.14790061712265015,
      "num_tokens": 3967375.0,
      "step": 1915
    },
    {
      "entropy": 5.825519275665283,
      "epoch": 1.744207178555202,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004906875387102935,
      "loss": 5.7991,
      "mean_token_accuracy": 0.1441611260175705,
      "num_tokens": 3977190.0,
      "step": 1920
    },
    {
      "entropy": 5.819745349884033,
      "epoch": 1.748750567923671,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004905866449680864,
      "loss": 5.7739,
      "mean_token_accuracy": 0.1543112054467201,
      "num_tokens": 3987878.0,
      "step": 1925
    },
    {
      "entropy": 5.807843160629273,
      "epoch": 1.75329395729214,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004904852192872638,
      "loss": 5.7595,
      "mean_token_accuracy": 0.15661360770463945,
      "num_tokens": 3997166.0,
      "step": 1930
    },
    {
      "entropy": 5.839473009109497,
      "epoch": 1.757837346660609,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004903832619180834,
      "loss": 5.7745,
      "mean_token_accuracy": 0.14904787093400956,
      "num_tokens": 4007711.0,
      "step": 1935
    },
    {
      "entropy": 5.879887485504151,
      "epoch": 1.7623807360290777,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004902807731121147,
      "loss": 5.9244,
      "mean_token_accuracy": 0.14445645958185196,
      "num_tokens": 4018907.0,
      "step": 1940
    },
    {
      "entropy": 5.81760721206665,
      "epoch": 1.7669241253975465,
      "grad_norm": 1.375,
      "learning_rate": 0.0004901777531222389,
      "loss": 5.805,
      "mean_token_accuracy": 0.15347942411899568,
      "num_tokens": 4030006.0,
      "step": 1945
    },
    {
      "entropy": 5.913085842132569,
      "epoch": 1.7714675147660155,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900742022026474,
      "loss": 5.8894,
      "mean_token_accuracy": 0.15333996266126632,
      "num_tokens": 4040609.0,
      "step": 1950
    },
    {
      "entropy": 5.780558061599732,
      "epoch": 1.7760109041344845,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004899701206088419,
      "loss": 5.7283,
      "mean_token_accuracy": 0.14981921315193175,
      "num_tokens": 4050392.0,
      "step": 1955
    },
    {
      "entropy": 5.786962413787842,
      "epoch": 1.7805542935029532,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004898655085976333,
      "loss": 5.8167,
      "mean_token_accuracy": 0.14799166470766068,
      "num_tokens": 4060156.0,
      "step": 1960
    },
    {
      "entropy": 5.847379684448242,
      "epoch": 1.785097682871422,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004897603664271413,
      "loss": 5.694,
      "mean_token_accuracy": 0.1542102053761482,
      "num_tokens": 4069670.0,
      "step": 1965
    },
    {
      "entropy": 5.715523719787598,
      "epoch": 1.789641072239891,
      "grad_norm": 1.140625,
      "learning_rate": 0.000489654694356794,
      "loss": 5.7208,
      "mean_token_accuracy": 0.1497572034597397,
      "num_tokens": 4080738.0,
      "step": 1970
    },
    {
      "entropy": 5.706692314147949,
      "epoch": 1.79418446160836,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004895484926473265,
      "loss": 5.7011,
      "mean_token_accuracy": 0.15149858146905898,
      "num_tokens": 4091122.0,
      "step": 1975
    },
    {
      "entropy": 5.886227178573608,
      "epoch": 1.7987278509768287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000489441761560781,
      "loss": 5.8172,
      "mean_token_accuracy": 0.15150666385889053,
      "num_tokens": 4101055.0,
      "step": 1980
    },
    {
      "entropy": 5.796804714202881,
      "epoch": 1.8032712403452975,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000489334501360506,
      "loss": 5.7593,
      "mean_token_accuracy": 0.15707833468914031,
      "num_tokens": 4110631.0,
      "step": 1985
    },
    {
      "entropy": 5.802775764465332,
      "epoch": 1.8078146297137665,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004892267123111551,
      "loss": 5.7715,
      "mean_token_accuracy": 0.15272098332643508,
      "num_tokens": 4121120.0,
      "step": 1990
    },
    {
      "entropy": 5.787547445297241,
      "epoch": 1.8123580190822355,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004891183946786873,
      "loss": 5.7183,
      "mean_token_accuracy": 0.15237994641065597,
      "num_tokens": 4132106.0,
      "step": 1995
    },
    {
      "entropy": 5.80684905052185,
      "epoch": 1.8169014084507042,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004890095487303655,
      "loss": 5.7586,
      "mean_token_accuracy": 0.1490725338459015,
      "num_tokens": 4142216.0,
      "step": 2000
    },
    {
      "epoch": 1.8169014084507042,
      "eval_entropy": 5.790170495454655,
      "eval_loss": 5.993229389190674,
      "eval_mean_token_accuracy": 0.14785447165716525,
      "eval_num_tokens": 4142216.0,
      "eval_runtime": 2.6052,
      "eval_samples_per_second": 1318.527,
      "eval_steps_per_second": 165.056,
      "step": 2000
    },
    {
      "entropy": 5.849746036529541,
      "epoch": 1.821444797819173,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004889001747347562,
      "loss": 5.8059,
      "mean_token_accuracy": 0.15014866143465042,
      "num_tokens": 4152168.0,
      "step": 2005
    },
    {
      "entropy": 5.766653680801392,
      "epoch": 1.825988187187642,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000488790272961729,
      "loss": 5.8032,
      "mean_token_accuracy": 0.14618541896343232,
      "num_tokens": 4162085.0,
      "step": 2010
    },
    {
      "entropy": 5.875260972976685,
      "epoch": 1.830531576556111,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004886798436824556,
      "loss": 5.8125,
      "mean_token_accuracy": 0.1477835439145565,
      "num_tokens": 4171168.0,
      "step": 2015
    },
    {
      "entropy": 5.769736194610596,
      "epoch": 1.8350749659245797,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004885688871694091,
      "loss": 5.7614,
      "mean_token_accuracy": 0.15020202696323395,
      "num_tokens": 4181804.0,
      "step": 2020
    },
    {
      "entropy": 5.821374797821045,
      "epoch": 1.8396183552930485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004884574036963639,
      "loss": 5.8199,
      "mean_token_accuracy": 0.1513478934764862,
      "num_tokens": 4192483.0,
      "step": 2025
    },
    {
      "entropy": 5.816771841049194,
      "epoch": 1.8441617446615175,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004883453935383941,
      "loss": 5.7112,
      "mean_token_accuracy": 0.1539171442389488,
      "num_tokens": 4202214.0,
      "step": 2030
    },
    {
      "entropy": 5.823734378814697,
      "epoch": 1.8487051340299865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004882328569718738,
      "loss": 5.7831,
      "mean_token_accuracy": 0.15346236750483513,
      "num_tokens": 4212440.0,
      "step": 2035
    },
    {
      "entropy": 5.768901395797729,
      "epoch": 1.8532485233984552,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004881197942744758,
      "loss": 5.7498,
      "mean_token_accuracy": 0.1491874173283577,
      "num_tokens": 4222780.0,
      "step": 2040
    },
    {
      "entropy": 5.76951231956482,
      "epoch": 1.857791912766924,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004880062057251711,
      "loss": 5.6804,
      "mean_token_accuracy": 0.15827766805887222,
      "num_tokens": 4232895.0,
      "step": 2045
    },
    {
      "entropy": 5.765989923477173,
      "epoch": 1.862335302135393,
      "grad_norm": 1.140625,
      "learning_rate": 0.000487892091604228,
      "loss": 5.7712,
      "mean_token_accuracy": 0.1545240879058838,
      "num_tokens": 4244287.0,
      "step": 2050
    },
    {
      "entropy": 5.81574444770813,
      "epoch": 1.866878691503862,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048777745219321186,
      "loss": 5.7374,
      "mean_token_accuracy": 0.14745835065841675,
      "num_tokens": 4254327.0,
      "step": 2055
    },
    {
      "entropy": 5.822335624694825,
      "epoch": 1.8714220808723308,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048766228777498405,
      "loss": 5.7705,
      "mean_token_accuracy": 0.15245601162314415,
      "num_tokens": 4264540.0,
      "step": 2060
    },
    {
      "entropy": 5.770896577835083,
      "epoch": 1.8759654702407995,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048754659863370125,
      "loss": 5.7368,
      "mean_token_accuracy": 0.15067947059869766,
      "num_tokens": 4274946.0,
      "step": 2065
    },
    {
      "entropy": 5.72808837890625,
      "epoch": 1.8805088596092685,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048743038505481503,
      "loss": 5.6962,
      "mean_token_accuracy": 0.15184763371944426,
      "num_tokens": 4284395.0,
      "step": 2070
    },
    {
      "entropy": 5.802645063400268,
      "epoch": 1.8850522489777375,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004873136473250708,
      "loss": 5.7742,
      "mean_token_accuracy": 0.150366872549057,
      "num_tokens": 4294887.0,
      "step": 2075
    },
    {
      "entropy": 5.788530588150024,
      "epoch": 1.8895956383462063,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048719638573250715,
      "loss": 5.7803,
      "mean_token_accuracy": 0.14959521740674972,
      "num_tokens": 4304794.0,
      "step": 2080
    },
    {
      "entropy": 5.781466913223267,
      "epoch": 1.894139027714675,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004870786005664555,
      "loss": 5.6932,
      "mean_token_accuracy": 0.15308475941419603,
      "num_tokens": 4314810.0,
      "step": 2085
    },
    {
      "entropy": 5.787781238555908,
      "epoch": 1.898682417083144,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004869602921175392,
      "loss": 5.7473,
      "mean_token_accuracy": 0.14809799045324326,
      "num_tokens": 4325444.0,
      "step": 2090
    },
    {
      "entropy": 5.799683856964111,
      "epoch": 1.903225806451613,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004868414606776724,
      "loss": 5.8012,
      "mean_token_accuracy": 0.14584705978631973,
      "num_tokens": 4335620.0,
      "step": 2095
    },
    {
      "entropy": 5.8017576217651365,
      "epoch": 1.9077691958200818,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048672210654005996,
      "loss": 5.7571,
      "mean_token_accuracy": 0.15560859441757202,
      "num_tokens": 4345006.0,
      "step": 2100
    },
    {
      "entropy": 5.784935331344604,
      "epoch": 1.9123125851885505,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048660222999919616,
      "loss": 5.6956,
      "mean_token_accuracy": 0.15184667482972145,
      "num_tokens": 4354805.0,
      "step": 2105
    },
    {
      "entropy": 5.795891237258911,
      "epoch": 1.9168559745570195,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004864818313508647,
      "loss": 5.8479,
      "mean_token_accuracy": 0.1495343454182148,
      "num_tokens": 4365881.0,
      "step": 2110
    },
    {
      "entropy": 5.868531036376953,
      "epoch": 1.9213993639254885,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004863609108921372,
      "loss": 5.8125,
      "mean_token_accuracy": 0.1463113822042942,
      "num_tokens": 4376368.0,
      "step": 2115
    },
    {
      "entropy": 5.78206090927124,
      "epoch": 1.9259427532939573,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004862394689213728,
      "loss": 5.731,
      "mean_token_accuracy": 0.1575425609946251,
      "num_tokens": 4386695.0,
      "step": 2120
    },
    {
      "entropy": 5.824512624740601,
      "epoch": 1.930486142662426,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004861175057382175,
      "loss": 5.764,
      "mean_token_accuracy": 0.14746586680412294,
      "num_tokens": 4396837.0,
      "step": 2125
    },
    {
      "entropy": 5.728180027008056,
      "epoch": 1.935029532030895,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004859950216436034,
      "loss": 5.735,
      "mean_token_accuracy": 0.1538913369178772,
      "num_tokens": 4407310.0,
      "step": 2130
    },
    {
      "entropy": 5.785648822784424,
      "epoch": 1.939572921399364,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004858720169397478,
      "loss": 5.7763,
      "mean_token_accuracy": 0.15280596613883973,
      "num_tokens": 4417544.0,
      "step": 2135
    },
    {
      "entropy": 5.789197397232056,
      "epoch": 1.9441163107678328,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004857484919301525,
      "loss": 5.6953,
      "mean_token_accuracy": 0.1523209258913994,
      "num_tokens": 4428228.0,
      "step": 2140
    },
    {
      "entropy": 5.759953117370605,
      "epoch": 1.9486597001363015,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004856244469196034,
      "loss": 5.7348,
      "mean_token_accuracy": 0.15404450446367263,
      "num_tokens": 4439088.0,
      "step": 2145
    },
    {
      "entropy": 5.827485275268555,
      "epoch": 1.9532030895047705,
      "grad_norm": 1.09375,
      "learning_rate": 0.000485499882214169,
      "loss": 5.834,
      "mean_token_accuracy": 0.14901283532381057,
      "num_tokens": 4449715.0,
      "step": 2150
    },
    {
      "entropy": 5.700705814361572,
      "epoch": 1.9577464788732395,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004853747981212003,
      "loss": 5.6836,
      "mean_token_accuracy": 0.155790339410305,
      "num_tokens": 4460233.0,
      "step": 2155
    },
    {
      "entropy": 5.732491683959961,
      "epoch": 1.9622898682417085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004852491949493301,
      "loss": 5.622,
      "mean_token_accuracy": 0.1616998255252838,
      "num_tokens": 4471401.0,
      "step": 2160
    },
    {
      "entropy": 5.765408706665039,
      "epoch": 1.9668332576101772,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004851230730084716,
      "loss": 5.6844,
      "mean_token_accuracy": 0.16186238080263138,
      "num_tokens": 4481185.0,
      "step": 2165
    },
    {
      "entropy": 5.768443298339844,
      "epoch": 1.971376646978646,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048499643260981816,
      "loss": 5.7243,
      "mean_token_accuracy": 0.15362609475851058,
      "num_tokens": 4492342.0,
      "step": 2170
    },
    {
      "entropy": 5.788624048233032,
      "epoch": 1.975920036347115,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004848692740658424,
      "loss": 5.8147,
      "mean_token_accuracy": 0.15219077914953233,
      "num_tokens": 4504277.0,
      "step": 2175
    },
    {
      "entropy": 5.800276279449463,
      "epoch": 1.980463425715584,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004847415976902954,
      "loss": 5.7605,
      "mean_token_accuracy": 0.15419393926858901,
      "num_tokens": 4514178.0,
      "step": 2180
    },
    {
      "entropy": 5.759483575820923,
      "epoch": 1.9850068150840527,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048461340379820587,
      "loss": 5.7428,
      "mean_token_accuracy": 0.1533619150519371,
      "num_tokens": 4524559.0,
      "step": 2185
    },
    {
      "entropy": 5.756233596801758,
      "epoch": 1.9895502044525215,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048448469270587945,
      "loss": 5.6855,
      "mean_token_accuracy": 0.1572543576359749,
      "num_tokens": 4535168.0,
      "step": 2190
    },
    {
      "entropy": 5.80228271484375,
      "epoch": 1.9940935938209905,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048435546473089814,
      "loss": 5.7826,
      "mean_token_accuracy": 0.15085486620664595,
      "num_tokens": 4546105.0,
      "step": 2195
    },
    {
      "entropy": 5.812653303146362,
      "epoch": 1.9986369831894595,
      "grad_norm": 1.125,
      "learning_rate": 0.00048422572019211896,
      "loss": 5.708,
      "mean_token_accuracy": 0.1578806847333908,
      "num_tokens": 4556498.0,
      "step": 2200
    },
    {
      "entropy": 5.66848791970147,
      "epoch": 2.0027260336210815,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048409545940967375,
      "loss": 5.5154,
      "mean_token_accuracy": 0.1526057016518381,
      "num_tokens": 4565029.0,
      "step": 2205
    },
    {
      "entropy": 5.796992635726928,
      "epoch": 2.00726942298955,
      "grad_norm": 1.171875,
      "learning_rate": 0.000483964682704968,
      "loss": 5.4685,
      "mean_token_accuracy": 0.15541424453258515,
      "num_tokens": 4575232.0,
      "step": 2210
    },
    {
      "entropy": 5.7437793731689455,
      "epoch": 2.011812812358019,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048383339040068024,
      "loss": 5.4532,
      "mean_token_accuracy": 0.1595204696059227,
      "num_tokens": 4585818.0,
      "step": 2215
    },
    {
      "entropy": 5.6068517684936525,
      "epoch": 2.016356201726488,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004837015828207612,
      "loss": 5.4012,
      "mean_token_accuracy": 0.16608490496873857,
      "num_tokens": 4595823.0,
      "step": 2220
    },
    {
      "entropy": 5.728079986572266,
      "epoch": 2.020899591094957,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004835692602904329,
      "loss": 5.3915,
      "mean_token_accuracy": 0.1636677697300911,
      "num_tokens": 4606537.0,
      "step": 2225
    },
    {
      "entropy": 5.746958637237549,
      "epoch": 2.0254429804634255,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004834364231361882,
      "loss": 5.4224,
      "mean_token_accuracy": 0.1628384292125702,
      "num_tokens": 4616096.0,
      "step": 2230
    },
    {
      "entropy": 5.710427331924438,
      "epoch": 2.0299863698318945,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004833030716857894,
      "loss": 5.4293,
      "mean_token_accuracy": 0.16120587512850762,
      "num_tokens": 4626287.0,
      "step": 2235
    },
    {
      "entropy": 5.759665155410767,
      "epoch": 2.0345297592003635,
      "grad_norm": 1.125,
      "learning_rate": 0.00048316920626826807,
      "loss": 5.4402,
      "mean_token_accuracy": 0.1623578727245331,
      "num_tokens": 4636898.0,
      "step": 2240
    },
    {
      "entropy": 5.624807262420655,
      "epoch": 2.0390731485688325,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048303482721392363,
      "loss": 5.4013,
      "mean_token_accuracy": 0.16981358975172042,
      "num_tokens": 4646263.0,
      "step": 2245
    },
    {
      "entropy": 5.736131954193115,
      "epoch": 2.043616537937301,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048289993485432325,
      "loss": 5.4431,
      "mean_token_accuracy": 0.16511887609958648,
      "num_tokens": 4656463.0,
      "step": 2250
    },
    {
      "entropy": 5.68239917755127,
      "epoch": 2.04815992730577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048276452952230036,
      "loss": 5.4688,
      "mean_token_accuracy": 0.1624162033200264,
      "num_tokens": 4666824.0,
      "step": 2255
    },
    {
      "entropy": 5.599757385253906,
      "epoch": 2.052703316674239,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048262861155195407,
      "loss": 5.3639,
      "mean_token_accuracy": 0.16735220551490784,
      "num_tokens": 4676793.0,
      "step": 2260
    },
    {
      "entropy": 5.742352914810181,
      "epoch": 2.057246706042708,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004824921812786486,
      "loss": 5.5055,
      "mean_token_accuracy": 0.16184383034706115,
      "num_tokens": 4687124.0,
      "step": 2265
    },
    {
      "entropy": 5.761610460281372,
      "epoch": 2.0617900954111765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000482355239039012,
      "loss": 5.6481,
      "mean_token_accuracy": 0.1535782903432846,
      "num_tokens": 4698074.0,
      "step": 2270
    },
    {
      "entropy": 5.754100513458252,
      "epoch": 2.0663334847796455,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004822177851709356,
      "loss": 5.5246,
      "mean_token_accuracy": 0.15992826223373413,
      "num_tokens": 4709177.0,
      "step": 2275
    },
    {
      "entropy": 5.782966804504395,
      "epoch": 2.0708768741481145,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048207982001357327,
      "loss": 5.4517,
      "mean_token_accuracy": 0.1593527875840664,
      "num_tokens": 4719649.0,
      "step": 2280
    },
    {
      "entropy": 5.684721803665161,
      "epoch": 2.0754202635165835,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004819413439073402,
      "loss": 5.5078,
      "mean_token_accuracy": 0.16035877615213395,
      "num_tokens": 4731685.0,
      "step": 2285
    },
    {
      "entropy": 5.66634693145752,
      "epoch": 2.079963652885052,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004818023571939125,
      "loss": 5.4607,
      "mean_token_accuracy": 0.16363389641046525,
      "num_tokens": 4742854.0,
      "step": 2290
    },
    {
      "entropy": 5.704556083679199,
      "epoch": 2.084507042253521,
      "grad_norm": 1.265625,
      "learning_rate": 0.000481662860216226,
      "loss": 5.4302,
      "mean_token_accuracy": 0.1609123393893242,
      "num_tokens": 4754088.0,
      "step": 2295
    },
    {
      "entropy": 5.625769758224488,
      "epoch": 2.08905043162199,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048152285331847574,
      "loss": 5.4563,
      "mean_token_accuracy": 0.16070560812950135,
      "num_tokens": 4764090.0,
      "step": 2300
    },
    {
      "entropy": 5.688674736022949,
      "epoch": 2.093593820990459,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004813823368461148,
      "loss": 5.4389,
      "mean_token_accuracy": 0.1677153393626213,
      "num_tokens": 4773704.0,
      "step": 2305
    },
    {
      "entropy": 5.719896936416626,
      "epoch": 2.098137210358928,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004812413111458534,
      "loss": 5.4889,
      "mean_token_accuracy": 0.1603416234254837,
      "num_tokens": 4783678.0,
      "step": 2310
    },
    {
      "entropy": 5.613268756866455,
      "epoch": 2.1026805997273965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004810997765656587,
      "loss": 5.3904,
      "mean_token_accuracy": 0.1691882386803627,
      "num_tokens": 4794183.0,
      "step": 2315
    },
    {
      "entropy": 5.591863107681275,
      "epoch": 2.1072239890958655,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048095773345475294,
      "loss": 5.3512,
      "mean_token_accuracy": 0.16529901921749116,
      "num_tokens": 4803670.0,
      "step": 2320
    },
    {
      "entropy": 5.654296922683716,
      "epoch": 2.1117673784643345,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048081518216361367,
      "loss": 5.3684,
      "mean_token_accuracy": 0.1701149135828018,
      "num_tokens": 4813611.0,
      "step": 2325
    },
    {
      "entropy": 5.61386194229126,
      "epoch": 2.1163107678328035,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004806721230439719,
      "loss": 5.4503,
      "mean_token_accuracy": 0.15981992781162263,
      "num_tokens": 4824356.0,
      "step": 2330
    },
    {
      "entropy": 5.734505891799927,
      "epoch": 2.120854157201272,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004805285564488117,
      "loss": 5.4841,
      "mean_token_accuracy": 0.16397735029458999,
      "num_tokens": 4834410.0,
      "step": 2335
    },
    {
      "entropy": 5.604253244400025,
      "epoch": 2.125397546569741,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004803844827323696,
      "loss": 5.3483,
      "mean_token_accuracy": 0.16647956222295762,
      "num_tokens": 4844975.0,
      "step": 2340
    },
    {
      "entropy": 5.64118070602417,
      "epoch": 2.12994093593821,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004802399022501329,
      "loss": 5.468,
      "mean_token_accuracy": 0.16066430658102035,
      "num_tokens": 4856419.0,
      "step": 2345
    },
    {
      "entropy": 5.5977051734924315,
      "epoch": 2.134484325306679,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004800948153588397,
      "loss": 5.3846,
      "mean_token_accuracy": 0.16584357768297195,
      "num_tokens": 4867115.0,
      "step": 2350
    },
    {
      "entropy": 5.622549533843994,
      "epoch": 2.1390277146751475,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047994922241647746,
      "loss": 5.4392,
      "mean_token_accuracy": 0.16220083683729172,
      "num_tokens": 4876675.0,
      "step": 2355
    },
    {
      "entropy": 5.697238397598267,
      "epoch": 2.1435711040436165,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047980312378228244,
      "loss": 5.4672,
      "mean_token_accuracy": 0.16455530673265456,
      "num_tokens": 4887868.0,
      "step": 2360
    },
    {
      "entropy": 5.6144118309021,
      "epoch": 2.1481144934120855,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047965651981673823,
      "loss": 5.4278,
      "mean_token_accuracy": 0.1708490490913391,
      "num_tokens": 4896878.0,
      "step": 2365
    },
    {
      "entropy": 5.613104057312012,
      "epoch": 2.1526578827805545,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047950941088157576,
      "loss": 5.4043,
      "mean_token_accuracy": 0.16638759076595305,
      "num_tokens": 4906683.0,
      "step": 2370
    },
    {
      "entropy": 5.624288129806518,
      "epoch": 2.157201272149023,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004793617973397715,
      "loss": 5.4105,
      "mean_token_accuracy": 0.16810904890298844,
      "num_tokens": 4917184.0,
      "step": 2375
    },
    {
      "entropy": 5.664835643768311,
      "epoch": 2.161744661517492,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047921367955554745,
      "loss": 5.4157,
      "mean_token_accuracy": 0.16286967247724532,
      "num_tokens": 4927337.0,
      "step": 2380
    },
    {
      "entropy": 5.647590398788452,
      "epoch": 2.166288050885961,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004790650578943693,
      "loss": 5.4402,
      "mean_token_accuracy": 0.15885256677865983,
      "num_tokens": 4937661.0,
      "step": 2385
    },
    {
      "entropy": 5.667690324783325,
      "epoch": 2.17083144025443,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047891593272294636,
      "loss": 5.4757,
      "mean_token_accuracy": 0.15949209034442902,
      "num_tokens": 4948068.0,
      "step": 2390
    },
    {
      "entropy": 5.602111148834228,
      "epoch": 2.1753748296228985,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047876630440923,
      "loss": 5.4359,
      "mean_token_accuracy": 0.16286666691303253,
      "num_tokens": 4958703.0,
      "step": 2395
    },
    {
      "entropy": 5.604837799072266,
      "epoch": 2.1799182189913675,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004786161733224134,
      "loss": 5.4303,
      "mean_token_accuracy": 0.16746487617492675,
      "num_tokens": 4969190.0,
      "step": 2400
    },
    {
      "entropy": 5.608381891250611,
      "epoch": 2.1844616083598365,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047846553983293006,
      "loss": 5.3352,
      "mean_token_accuracy": 0.17012607008218766,
      "num_tokens": 4978663.0,
      "step": 2405
    },
    {
      "entropy": 5.718901681900024,
      "epoch": 2.1890049977283055,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047831440431245307,
      "loss": 5.5016,
      "mean_token_accuracy": 0.1635862797498703,
      "num_tokens": 4989839.0,
      "step": 2410
    },
    {
      "entropy": 5.570841836929321,
      "epoch": 2.193548387096774,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047816276713389424,
      "loss": 5.3359,
      "mean_token_accuracy": 0.16816314607858657,
      "num_tokens": 5000030.0,
      "step": 2415
    },
    {
      "entropy": 5.614529418945312,
      "epoch": 2.198091776465243,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004780106286714034,
      "loss": 5.5246,
      "mean_token_accuracy": 0.1619194909930229,
      "num_tokens": 5010229.0,
      "step": 2420
    },
    {
      "entropy": 5.72342209815979,
      "epoch": 2.202635165833712,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000477857989300367,
      "loss": 5.4352,
      "mean_token_accuracy": 0.16469539403915406,
      "num_tokens": 5021000.0,
      "step": 2425
    },
    {
      "entropy": 5.556600046157837,
      "epoch": 2.207178555202181,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047770484939740757,
      "loss": 5.3528,
      "mean_token_accuracy": 0.16867441087961196,
      "num_tokens": 5031319.0,
      "step": 2430
    },
    {
      "entropy": 5.554885625839233,
      "epoch": 2.2117219445706495,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004775512093403824,
      "loss": 5.3877,
      "mean_token_accuracy": 0.16844506412744523,
      "num_tokens": 5041892.0,
      "step": 2435
    },
    {
      "entropy": 5.641312980651856,
      "epoch": 2.2162653339391185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047739706950838327,
      "loss": 5.4625,
      "mean_token_accuracy": 0.16928391605615617,
      "num_tokens": 5052363.0,
      "step": 2440
    },
    {
      "entropy": 5.721763467788696,
      "epoch": 2.2208087233075875,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047724243028173484,
      "loss": 5.4828,
      "mean_token_accuracy": 0.16696733087301255,
      "num_tokens": 5063342.0,
      "step": 2445
    },
    {
      "entropy": 5.595815992355346,
      "epoch": 2.2253521126760565,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047708729204199406,
      "loss": 5.3949,
      "mean_token_accuracy": 0.15966860353946685,
      "num_tokens": 5073393.0,
      "step": 2450
    },
    {
      "entropy": 5.556210947036743,
      "epoch": 2.229895502044525,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000476931655171949,
      "loss": 5.3653,
      "mean_token_accuracy": 0.17017862498760222,
      "num_tokens": 5082417.0,
      "step": 2455
    },
    {
      "entropy": 5.705835771560669,
      "epoch": 2.234438891412994,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004767755200556183,
      "loss": 5.5362,
      "mean_token_accuracy": 0.15762273222208023,
      "num_tokens": 5093383.0,
      "step": 2460
    },
    {
      "entropy": 5.540210199356079,
      "epoch": 2.238982280781463,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004766188870782497,
      "loss": 5.3225,
      "mean_token_accuracy": 0.17020710408687592,
      "num_tokens": 5104346.0,
      "step": 2465
    },
    {
      "entropy": 5.558584880828858,
      "epoch": 2.243525670149932,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047646175662631967,
      "loss": 5.493,
      "mean_token_accuracy": 0.1641610309481621,
      "num_tokens": 5114375.0,
      "step": 2470
    },
    {
      "entropy": 5.656518793106079,
      "epoch": 2.2480690595184005,
      "grad_norm": 1.25,
      "learning_rate": 0.00047630412908753183,
      "loss": 5.4652,
      "mean_token_accuracy": 0.16726542860269547,
      "num_tokens": 5125035.0,
      "step": 2475
    },
    {
      "entropy": 5.6312737464904785,
      "epoch": 2.2526124488868695,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047614600485081657,
      "loss": 5.4692,
      "mean_token_accuracy": 0.1580390676856041,
      "num_tokens": 5135869.0,
      "step": 2480
    },
    {
      "entropy": 5.73362922668457,
      "epoch": 2.2571558382553385,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047598738430632963,
      "loss": 5.4279,
      "mean_token_accuracy": 0.16724849492311478,
      "num_tokens": 5146615.0,
      "step": 2485
    },
    {
      "entropy": 5.522592401504516,
      "epoch": 2.2616992276238075,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047582826784545146,
      "loss": 5.3366,
      "mean_token_accuracy": 0.16956065744161605,
      "num_tokens": 5157490.0,
      "step": 2490
    },
    {
      "entropy": 5.586135482788086,
      "epoch": 2.266242616992276,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047566865586078615,
      "loss": 5.4806,
      "mean_token_accuracy": 0.1588055059313774,
      "num_tokens": 5168693.0,
      "step": 2495
    },
    {
      "entropy": 5.658676815032959,
      "epoch": 2.270786006360745,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047550854874616047,
      "loss": 5.4435,
      "mean_token_accuracy": 0.1633205607533455,
      "num_tokens": 5179232.0,
      "step": 2500
    },
    {
      "epoch": 2.270786006360745,
      "eval_entropy": 5.410215771475504,
      "eval_loss": 5.926918983459473,
      "eval_mean_token_accuracy": 0.15270435387311979,
      "eval_num_tokens": 5179232.0,
      "eval_runtime": 2.6005,
      "eval_samples_per_second": 1320.907,
      "eval_steps_per_second": 165.354,
      "step": 2500
    },
    {
      "entropy": 5.575441551208496,
      "epoch": 2.275329395729214,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047534794689662257,
      "loss": 5.4851,
      "mean_token_accuracy": 0.16839685291051865,
      "num_tokens": 5188613.0,
      "step": 2505
    },
    {
      "entropy": 5.622188568115234,
      "epoch": 2.279872785097683,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004751868507084418,
      "loss": 5.3639,
      "mean_token_accuracy": 0.17478009462356567,
      "num_tokens": 5198441.0,
      "step": 2510
    },
    {
      "entropy": 5.584142398834229,
      "epoch": 2.284416174466152,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004750252605791068,
      "loss": 5.4392,
      "mean_token_accuracy": 0.16842508912086487,
      "num_tokens": 5208466.0,
      "step": 2515
    },
    {
      "entropy": 5.495612478256225,
      "epoch": 2.2889595638346205,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004748631769073254,
      "loss": 5.4106,
      "mean_token_accuracy": 0.16598602533340454,
      "num_tokens": 5218847.0,
      "step": 2520
    },
    {
      "entropy": 5.6372777938842775,
      "epoch": 2.2935029532030895,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004747006000930227,
      "loss": 5.4535,
      "mean_token_accuracy": 0.1546911284327507,
      "num_tokens": 5228870.0,
      "step": 2525
    },
    {
      "entropy": 5.6832057476043705,
      "epoch": 2.2980463425715585,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004745375305373409,
      "loss": 5.3881,
      "mean_token_accuracy": 0.1696573704481125,
      "num_tokens": 5238186.0,
      "step": 2530
    },
    {
      "entropy": 5.5625715255737305,
      "epoch": 2.302589731940027,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047437396864263804,
      "loss": 5.4582,
      "mean_token_accuracy": 0.16980416923761368,
      "num_tokens": 5247478.0,
      "step": 2535
    },
    {
      "entropy": 5.65769944190979,
      "epoch": 2.307133121308496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004742099148124866,
      "loss": 5.4044,
      "mean_token_accuracy": 0.1642364203929901,
      "num_tokens": 5258519.0,
      "step": 2540
    },
    {
      "entropy": 5.6592381477355955,
      "epoch": 2.311676510676965,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004740453694516734,
      "loss": 5.4296,
      "mean_token_accuracy": 0.16748272031545638,
      "num_tokens": 5268218.0,
      "step": 2545
    },
    {
      "entropy": 5.493631792068482,
      "epoch": 2.316219900045434,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047388033296619746,
      "loss": 5.3937,
      "mean_token_accuracy": 0.16241876482963563,
      "num_tokens": 5277926.0,
      "step": 2550
    },
    {
      "entropy": 5.5903431415557865,
      "epoch": 2.320763289413903,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004737148057632701,
      "loss": 5.4205,
      "mean_token_accuracy": 0.1694105461239815,
      "num_tokens": 5287735.0,
      "step": 2555
    },
    {
      "entropy": 5.570258474349975,
      "epoch": 2.3253066787823715,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004735487882513131,
      "loss": 5.3496,
      "mean_token_accuracy": 0.17214408963918687,
      "num_tokens": 5297120.0,
      "step": 2560
    },
    {
      "entropy": 5.45852484703064,
      "epoch": 2.3298500681508405,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004733822808399583,
      "loss": 5.3662,
      "mean_token_accuracy": 0.17136418521404267,
      "num_tokens": 5307448.0,
      "step": 2565
    },
    {
      "entropy": 5.61703200340271,
      "epoch": 2.3343934575193095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047321528394004615,
      "loss": 5.383,
      "mean_token_accuracy": 0.17348483204841614,
      "num_tokens": 5317914.0,
      "step": 2570
    },
    {
      "entropy": 5.543197679519653,
      "epoch": 2.3389368468877785,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047304779796362485,
      "loss": 5.3766,
      "mean_token_accuracy": 0.17093151062726974,
      "num_tokens": 5328471.0,
      "step": 2575
    },
    {
      "entropy": 5.516396713256836,
      "epoch": 2.343480236256247,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047287982332394943,
      "loss": 5.4029,
      "mean_token_accuracy": 0.16134533137083054,
      "num_tokens": 5339118.0,
      "step": 2580
    },
    {
      "entropy": 5.67015905380249,
      "epoch": 2.348023625624716,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047271136043548057,
      "loss": 5.5006,
      "mean_token_accuracy": 0.15954657047986984,
      "num_tokens": 5350183.0,
      "step": 2585
    },
    {
      "entropy": 5.617479610443115,
      "epoch": 2.352567014993185,
      "grad_norm": 1.125,
      "learning_rate": 0.0004725424097138838,
      "loss": 5.4171,
      "mean_token_accuracy": 0.16176852732896804,
      "num_tokens": 5361317.0,
      "step": 2590
    },
    {
      "entropy": 5.568935775756836,
      "epoch": 2.357110404361654,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047237297157602816,
      "loss": 5.423,
      "mean_token_accuracy": 0.1682720050215721,
      "num_tokens": 5371338.0,
      "step": 2595
    },
    {
      "entropy": 5.66186933517456,
      "epoch": 2.3616537937301225,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004722030464399854,
      "loss": 5.5323,
      "mean_token_accuracy": 0.16112422049045563,
      "num_tokens": 5382043.0,
      "step": 2600
    },
    {
      "entropy": 5.6375768184661865,
      "epoch": 2.3661971830985915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000472032634725029,
      "loss": 5.4346,
      "mean_token_accuracy": 0.16537451148033142,
      "num_tokens": 5392547.0,
      "step": 2605
    },
    {
      "entropy": 5.46543173789978,
      "epoch": 2.3707405724670605,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047186173685163273,
      "loss": 5.3103,
      "mean_token_accuracy": 0.17888594716787337,
      "num_tokens": 5402455.0,
      "step": 2610
    },
    {
      "entropy": 5.549544048309326,
      "epoch": 2.3752839618355295,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047169035324147023,
      "loss": 5.4883,
      "mean_token_accuracy": 0.16207679510116577,
      "num_tokens": 5413925.0,
      "step": 2615
    },
    {
      "entropy": 5.649377870559692,
      "epoch": 2.379827351203998,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047151848431741363,
      "loss": 5.5142,
      "mean_token_accuracy": 0.15564770698547364,
      "num_tokens": 5425344.0,
      "step": 2620
    },
    {
      "entropy": 5.551960802078247,
      "epoch": 2.384370740572467,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004713461305035322,
      "loss": 5.4531,
      "mean_token_accuracy": 0.16062026470899582,
      "num_tokens": 5435197.0,
      "step": 2625
    },
    {
      "entropy": 5.683890295028687,
      "epoch": 2.388914129940936,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000471173292225092,
      "loss": 5.441,
      "mean_token_accuracy": 0.166593936085701,
      "num_tokens": 5445683.0,
      "step": 2630
    },
    {
      "entropy": 5.553751707077026,
      "epoch": 2.393457519309405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047099996990855424,
      "loss": 5.3168,
      "mean_token_accuracy": 0.17248447388410568,
      "num_tokens": 5456846.0,
      "step": 2635
    },
    {
      "entropy": 5.486712694168091,
      "epoch": 2.3980009086778735,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047082616398157456,
      "loss": 5.3966,
      "mean_token_accuracy": 0.16698428243398666,
      "num_tokens": 5467701.0,
      "step": 2640
    },
    {
      "entropy": 5.618938636779785,
      "epoch": 2.4025442980463425,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004706518748730017,
      "loss": 5.4277,
      "mean_token_accuracy": 0.1629345163702965,
      "num_tokens": 5478246.0,
      "step": 2645
    },
    {
      "entropy": 5.579886436462402,
      "epoch": 2.4070876874148115,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004704771030128768,
      "loss": 5.4166,
      "mean_token_accuracy": 0.17258816808462143,
      "num_tokens": 5488107.0,
      "step": 2650
    },
    {
      "entropy": 5.611078310012817,
      "epoch": 2.4116310767832805,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047030184883243203,
      "loss": 5.4146,
      "mean_token_accuracy": 0.16715776920318604,
      "num_tokens": 5498521.0,
      "step": 2655
    },
    {
      "entropy": 5.542270088195801,
      "epoch": 2.416174466151749,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047012611276408964,
      "loss": 5.4674,
      "mean_token_accuracy": 0.15941065400838852,
      "num_tokens": 5509450.0,
      "step": 2660
    },
    {
      "entropy": 5.610725355148316,
      "epoch": 2.420717855520218,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004699498952414609,
      "loss": 5.3711,
      "mean_token_accuracy": 0.16671083718538285,
      "num_tokens": 5519662.0,
      "step": 2665
    },
    {
      "entropy": 5.5634044170379635,
      "epoch": 2.425261244888687,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046977319669934506,
      "loss": 5.3492,
      "mean_token_accuracy": 0.16818245649337768,
      "num_tokens": 5529499.0,
      "step": 2670
    },
    {
      "entropy": 5.544408178329467,
      "epoch": 2.429804634257156,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046959601757372814,
      "loss": 5.5263,
      "mean_token_accuracy": 0.1580025553703308,
      "num_tokens": 5540938.0,
      "step": 2675
    },
    {
      "entropy": 5.617356300354004,
      "epoch": 2.4343480236256245,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046941835830178216,
      "loss": 5.3945,
      "mean_token_accuracy": 0.16811649203300477,
      "num_tokens": 5552120.0,
      "step": 2680
    },
    {
      "entropy": 5.520823812484741,
      "epoch": 2.4388914129940935,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046924021932186357,
      "loss": 5.3612,
      "mean_token_accuracy": 0.16833633482456206,
      "num_tokens": 5561590.0,
      "step": 2685
    },
    {
      "entropy": 5.538807725906372,
      "epoch": 2.4434348023625625,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046906160107351263,
      "loss": 5.3543,
      "mean_token_accuracy": 0.17522021383047104,
      "num_tokens": 5571682.0,
      "step": 2690
    },
    {
      "entropy": 5.5401527881622314,
      "epoch": 2.4479781917310315,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046888250399745217,
      "loss": 5.3823,
      "mean_token_accuracy": 0.17362736463546752,
      "num_tokens": 5581299.0,
      "step": 2695
    },
    {
      "entropy": 5.500322532653809,
      "epoch": 2.4525215810995,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046870292853558646,
      "loss": 5.3322,
      "mean_token_accuracy": 0.17054419070482255,
      "num_tokens": 5590766.0,
      "step": 2700
    },
    {
      "entropy": 5.596356201171875,
      "epoch": 2.457064970467969,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046852287513099997,
      "loss": 5.4528,
      "mean_token_accuracy": 0.1649368464946747,
      "num_tokens": 5600898.0,
      "step": 2705
    },
    {
      "entropy": 5.5878630638122555,
      "epoch": 2.461608359836438,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004683423442279566,
      "loss": 5.4471,
      "mean_token_accuracy": 0.1655479371547699,
      "num_tokens": 5612052.0,
      "step": 2710
    },
    {
      "entropy": 5.576379442214966,
      "epoch": 2.466151749204907,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004681613362718985,
      "loss": 5.5255,
      "mean_token_accuracy": 0.16940135806798934,
      "num_tokens": 5622847.0,
      "step": 2715
    },
    {
      "entropy": 5.519678974151612,
      "epoch": 2.470695138573376,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004679798517094447,
      "loss": 5.3434,
      "mean_token_accuracy": 0.17103047668933868,
      "num_tokens": 5633134.0,
      "step": 2720
    },
    {
      "entropy": 5.61676058769226,
      "epoch": 2.4752385279418445,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046779789098839043,
      "loss": 5.5131,
      "mean_token_accuracy": 0.16538699269294738,
      "num_tokens": 5644223.0,
      "step": 2725
    },
    {
      "entropy": 5.535487318038941,
      "epoch": 2.4797819173103135,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046761545455770555,
      "loss": 5.441,
      "mean_token_accuracy": 0.16498867720365523,
      "num_tokens": 5653620.0,
      "step": 2730
    },
    {
      "entropy": 5.589802694320679,
      "epoch": 2.4843253066787825,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004674325428675339,
      "loss": 5.3628,
      "mean_token_accuracy": 0.16851337254047394,
      "num_tokens": 5663800.0,
      "step": 2735
    },
    {
      "entropy": 5.5757472038269045,
      "epoch": 2.488868696047251,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004672491563691919,
      "loss": 5.4142,
      "mean_token_accuracy": 0.17137184292078017,
      "num_tokens": 5674606.0,
      "step": 2740
    },
    {
      "entropy": 5.553486776351929,
      "epoch": 2.49341208541572,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046706529551516753,
      "loss": 5.5256,
      "mean_token_accuracy": 0.16617924422025682,
      "num_tokens": 5684975.0,
      "step": 2745
    },
    {
      "entropy": 5.5582334995269775,
      "epoch": 2.497955474784189,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046688096075911905,
      "loss": 5.3629,
      "mean_token_accuracy": 0.17054101526737214,
      "num_tokens": 5695336.0,
      "step": 2750
    },
    {
      "entropy": 5.629152488708496,
      "epoch": 2.502498864152658,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004666961525558743,
      "loss": 5.3853,
      "mean_token_accuracy": 0.17582341730594636,
      "num_tokens": 5706027.0,
      "step": 2755
    },
    {
      "entropy": 5.53211817741394,
      "epoch": 2.507042253521127,
      "grad_norm": 1.21875,
      "learning_rate": 0.000466510871361429,
      "loss": 5.4212,
      "mean_token_accuracy": 0.16578415632247925,
      "num_tokens": 5716026.0,
      "step": 2760
    },
    {
      "entropy": 5.543093681335449,
      "epoch": 2.5115856428895955,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046632511763294615,
      "loss": 5.4318,
      "mean_token_accuracy": 0.1688849925994873,
      "num_tokens": 5726055.0,
      "step": 2765
    },
    {
      "entropy": 5.624571752548218,
      "epoch": 2.5161290322580645,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004661388918287546,
      "loss": 5.4575,
      "mean_token_accuracy": 0.1618290051817894,
      "num_tokens": 5736562.0,
      "step": 2770
    },
    {
      "entropy": 5.6554388999938965,
      "epoch": 2.5206724216265335,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004659521944083481,
      "loss": 5.5098,
      "mean_token_accuracy": 0.1590184301137924,
      "num_tokens": 5747297.0,
      "step": 2775
    },
    {
      "entropy": 5.538739776611328,
      "epoch": 2.525215810995002,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004657650258323838,
      "loss": 5.478,
      "mean_token_accuracy": 0.16293535381555557,
      "num_tokens": 5756801.0,
      "step": 2780
    },
    {
      "entropy": 5.576935529708862,
      "epoch": 2.529759200363471,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046557738656268166,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1629885956645012,
      "num_tokens": 5767493.0,
      "step": 2785
    },
    {
      "entropy": 5.555518960952758,
      "epoch": 2.53430258973194,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000465389277062223,
      "loss": 5.3594,
      "mean_token_accuracy": 0.17765380591154098,
      "num_tokens": 5777539.0,
      "step": 2790
    },
    {
      "entropy": 5.553316736221314,
      "epoch": 2.538845979100409,
      "grad_norm": 1.25,
      "learning_rate": 0.00046520069779514917,
      "loss": 5.3735,
      "mean_token_accuracy": 0.17463183254003525,
      "num_tokens": 5787202.0,
      "step": 2795
    },
    {
      "entropy": 5.628433799743652,
      "epoch": 2.543389368468878,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004650116492267609,
      "loss": 5.4325,
      "mean_token_accuracy": 0.1649623319506645,
      "num_tokens": 5798023.0,
      "step": 2800
    },
    {
      "entropy": 5.5028839111328125,
      "epoch": 2.5479327578373465,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004648221318235165,
      "loss": 5.33,
      "mean_token_accuracy": 0.173333603143692,
      "num_tokens": 5807711.0,
      "step": 2805
    },
    {
      "entropy": 5.53934383392334,
      "epoch": 2.5524761472058155,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004646321460530316,
      "loss": 5.3492,
      "mean_token_accuracy": 0.17637757062911988,
      "num_tokens": 5817053.0,
      "step": 2810
    },
    {
      "entropy": 5.51912112236023,
      "epoch": 2.5570195365742845,
      "grad_norm": 1.421875,
      "learning_rate": 0.000464441692384077,
      "loss": 5.4475,
      "mean_token_accuracy": 0.1696236550807953,
      "num_tokens": 5827792.0,
      "step": 2815
    },
    {
      "entropy": 5.547548723220825,
      "epoch": 2.561562925942753,
      "grad_norm": 1.25,
      "learning_rate": 0.0004642507712865784,
      "loss": 5.3467,
      "mean_token_accuracy": 0.17662595361471176,
      "num_tokens": 5837702.0,
      "step": 2820
    },
    {
      "entropy": 5.595866680145264,
      "epoch": 2.566106315311222,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046405938323161453,
      "loss": 5.4854,
      "mean_token_accuracy": 0.16242269724607467,
      "num_tokens": 5847973.0,
      "step": 2825
    },
    {
      "entropy": 5.641831970214843,
      "epoch": 2.570649704679691,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046386752869141624,
      "loss": 5.4681,
      "mean_token_accuracy": 0.16168300807476044,
      "num_tokens": 5857591.0,
      "step": 2830
    },
    {
      "entropy": 5.531976175308228,
      "epoch": 2.57519309404816,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004636752081393658,
      "loss": 5.427,
      "mean_token_accuracy": 0.16628410071134567,
      "num_tokens": 5868616.0,
      "step": 2835
    },
    {
      "entropy": 5.584429121017456,
      "epoch": 2.579736483416629,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046348242204999487,
      "loss": 5.3655,
      "mean_token_accuracy": 0.16841525882482528,
      "num_tokens": 5878526.0,
      "step": 2840
    },
    {
      "entropy": 5.5044105529785154,
      "epoch": 2.5842798727850975,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046328917089898396,
      "loss": 5.3646,
      "mean_token_accuracy": 0.17235694974660873,
      "num_tokens": 5887844.0,
      "step": 2845
    },
    {
      "entropy": 5.548652219772339,
      "epoch": 2.5888232621535665,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004630954551631612,
      "loss": 5.4672,
      "mean_token_accuracy": 0.16923740059137343,
      "num_tokens": 5898523.0,
      "step": 2850
    },
    {
      "entropy": 5.582967376708984,
      "epoch": 2.5933666515220355,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046290127532050087,
      "loss": 5.4856,
      "mean_token_accuracy": 0.16219591945409775,
      "num_tokens": 5909671.0,
      "step": 2855
    },
    {
      "entropy": 5.532243728637695,
      "epoch": 2.5979100408905045,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004627066318501223,
      "loss": 5.3943,
      "mean_token_accuracy": 0.1625651866197586,
      "num_tokens": 5920451.0,
      "step": 2860
    },
    {
      "entropy": 5.5454424858093265,
      "epoch": 2.602453430258973,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046251152523228906,
      "loss": 5.4,
      "mean_token_accuracy": 0.17224705815315247,
      "num_tokens": 5931218.0,
      "step": 2865
    },
    {
      "entropy": 5.466064500808716,
      "epoch": 2.606996819627442,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004623159559484074,
      "loss": 5.409,
      "mean_token_accuracy": 0.16545435637235642,
      "num_tokens": 5941380.0,
      "step": 2870
    },
    {
      "entropy": 5.5833117961883545,
      "epoch": 2.611540208995911,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004621199244810248,
      "loss": 5.3604,
      "mean_token_accuracy": 0.1685359075665474,
      "num_tokens": 5951989.0,
      "step": 2875
    },
    {
      "entropy": 5.5274125099182125,
      "epoch": 2.61608359836438,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046192343131382964,
      "loss": 5.4266,
      "mean_token_accuracy": 0.16473983377218246,
      "num_tokens": 5961982.0,
      "step": 2880
    },
    {
      "entropy": 5.592204570770264,
      "epoch": 2.620626987732849,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046172647693164926,
      "loss": 5.4702,
      "mean_token_accuracy": 0.16245893388986588,
      "num_tokens": 5972184.0,
      "step": 2885
    },
    {
      "entropy": 5.5910967826843265,
      "epoch": 2.6251703771013175,
      "grad_norm": 1.25,
      "learning_rate": 0.0004615290618204489,
      "loss": 5.3963,
      "mean_token_accuracy": 0.16848855316638947,
      "num_tokens": 5981912.0,
      "step": 2890
    },
    {
      "entropy": 5.533260202407837,
      "epoch": 2.6297137664697865,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004613311864673309,
      "loss": 5.3817,
      "mean_token_accuracy": 0.17015166729688644,
      "num_tokens": 5992460.0,
      "step": 2895
    },
    {
      "entropy": 5.5702805519104,
      "epoch": 2.6342571558382555,
      "grad_norm": 1.375,
      "learning_rate": 0.00046113285136053286,
      "loss": 5.4398,
      "mean_token_accuracy": 0.16693648248910903,
      "num_tokens": 6002170.0,
      "step": 2900
    },
    {
      "entropy": 5.510522031784058,
      "epoch": 2.638800545206724,
      "grad_norm": 1.234375,
      "learning_rate": 0.000460934056989427,
      "loss": 5.349,
      "mean_token_accuracy": 0.16113705635070802,
      "num_tokens": 6012240.0,
      "step": 2905
    },
    {
      "entropy": 5.564363622665406,
      "epoch": 2.643343934575193,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004607348038445187,
      "loss": 5.3199,
      "mean_token_accuracy": 0.17848723530769348,
      "num_tokens": 6022289.0,
      "step": 2910
    },
    {
      "entropy": 5.595773458480835,
      "epoch": 2.647887323943662,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046053509241744526,
      "loss": 5.4483,
      "mean_token_accuracy": 0.1646754503250122,
      "num_tokens": 6032268.0,
      "step": 2915
    },
    {
      "entropy": 5.666926097869873,
      "epoch": 2.652430713312131,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046033492320097483,
      "loss": 5.5214,
      "mean_token_accuracy": 0.16039399057626724,
      "num_tokens": 6042574.0,
      "step": 2920
    },
    {
      "entropy": 5.6914349555969235,
      "epoch": 2.6569741026806,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004601342966890049,
      "loss": 5.4947,
      "mean_token_accuracy": 0.1591247946023941,
      "num_tokens": 6052508.0,
      "step": 2925
    },
    {
      "entropy": 5.463538599014282,
      "epoch": 2.6615174920490685,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045993321337656163,
      "loss": 5.3321,
      "mean_token_accuracy": 0.1798750251531601,
      "num_tokens": 6062541.0,
      "step": 2930
    },
    {
      "entropy": 5.529683589935303,
      "epoch": 2.6660608814175375,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000459731673759798,
      "loss": 5.501,
      "mean_token_accuracy": 0.1589680328965187,
      "num_tokens": 6072532.0,
      "step": 2935
    },
    {
      "entropy": 5.580250597000122,
      "epoch": 2.6706042707860065,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045952967833599303,
      "loss": 5.3838,
      "mean_token_accuracy": 0.17602769434452056,
      "num_tokens": 6082518.0,
      "step": 2940
    },
    {
      "entropy": 5.56395001411438,
      "epoch": 2.675147660154475,
      "grad_norm": 1.25,
      "learning_rate": 0.0004593272276035502,
      "loss": 5.3736,
      "mean_token_accuracy": 0.17306891083717346,
      "num_tokens": 6092356.0,
      "step": 2945
    },
    {
      "entropy": 5.564646244049072,
      "epoch": 2.679691049522944,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004591243220619969,
      "loss": 5.4849,
      "mean_token_accuracy": 0.16785520017147065,
      "num_tokens": 6103450.0,
      "step": 2950
    },
    {
      "entropy": 5.5815471649169925,
      "epoch": 2.684234438891413,
      "grad_norm": 1.484375,
      "learning_rate": 0.000458920962211982,
      "loss": 5.3605,
      "mean_token_accuracy": 0.17186060249805452,
      "num_tokens": 6113708.0,
      "step": 2955
    },
    {
      "entropy": 5.501417446136474,
      "epoch": 2.688777828259882,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004587171485552759,
      "loss": 5.4345,
      "mean_token_accuracy": 0.16605052649974822,
      "num_tokens": 6124508.0,
      "step": 2960
    },
    {
      "entropy": 5.609457397460938,
      "epoch": 2.693321217628351,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045851288159476854,
      "loss": 5.4159,
      "mean_token_accuracy": 0.17805173248052597,
      "num_tokens": 6135516.0,
      "step": 2965
    },
    {
      "entropy": 5.531106567382812,
      "epoch": 2.6978646069968195,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004583081618344684,
      "loss": 5.3405,
      "mean_token_accuracy": 0.16844869256019593,
      "num_tokens": 6146206.0,
      "step": 2970
    },
    {
      "entropy": 5.461429405212402,
      "epoch": 2.7024079963652885,
      "grad_norm": 1.265625,
      "learning_rate": 0.000458102989779501,
      "loss": 5.3038,
      "mean_token_accuracy": 0.17313728034496306,
      "num_tokens": 6156244.0,
      "step": 2975
    },
    {
      "entropy": 5.415205717086792,
      "epoch": 2.7069513857337575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004578973659361081,
      "loss": 5.3411,
      "mean_token_accuracy": 0.1671074226498604,
      "num_tokens": 6167091.0,
      "step": 2980
    },
    {
      "entropy": 5.516201972961426,
      "epoch": 2.711494775102226,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045769129081164596,
      "loss": 5.4108,
      "mean_token_accuracy": 0.17164330929517746,
      "num_tokens": 6177993.0,
      "step": 2985
    },
    {
      "entropy": 5.499716949462891,
      "epoch": 2.716038164470695,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004574847649145845,
      "loss": 5.4394,
      "mean_token_accuracy": 0.17122387140989304,
      "num_tokens": 6187963.0,
      "step": 2990
    },
    {
      "entropy": 5.5332225322723385,
      "epoch": 2.720581553839164,
      "grad_norm": 1.296875,
      "learning_rate": 0.000457277788754506,
      "loss": 5.419,
      "mean_token_accuracy": 0.16892076134681702,
      "num_tokens": 6198442.0,
      "step": 2995
    },
    {
      "entropy": 5.560413360595703,
      "epoch": 2.725124943207633,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045707036284210344,
      "loss": 5.4295,
      "mean_token_accuracy": 0.1676975205540657,
      "num_tokens": 6209332.0,
      "step": 3000
    },
    {
      "epoch": 2.725124943207633,
      "eval_entropy": 5.441976602687392,
      "eval_loss": 5.870466232299805,
      "eval_mean_token_accuracy": 0.15629731280685857,
      "eval_num_tokens": 6209332.0,
      "eval_runtime": 2.6005,
      "eval_samples_per_second": 1320.919,
      "eval_steps_per_second": 165.355,
      "step": 3000
    },
    {
      "entropy": 5.472497844696045,
      "epoch": 2.729668332576102,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004568624876891796,
      "loss": 5.3969,
      "mean_token_accuracy": 0.16789249032735826,
      "num_tokens": 6219467.0,
      "step": 3005
    },
    {
      "entropy": 5.526805019378662,
      "epoch": 2.7342117219445705,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045665416380864587,
      "loss": 5.4108,
      "mean_token_accuracy": 0.16932470202445984,
      "num_tokens": 6229277.0,
      "step": 3010
    },
    {
      "entropy": 5.555971479415893,
      "epoch": 2.7387551113130395,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004564453917145206,
      "loss": 5.3423,
      "mean_token_accuracy": 0.17621685266494752,
      "num_tokens": 6239191.0,
      "step": 3015
    },
    {
      "entropy": 5.5381481647491455,
      "epoch": 2.7432985006815085,
      "grad_norm": 1.25,
      "learning_rate": 0.0004562361719219283,
      "loss": 5.4587,
      "mean_token_accuracy": 0.16459385603666304,
      "num_tokens": 6249539.0,
      "step": 3020
    },
    {
      "entropy": 5.589121007919312,
      "epoch": 2.747841890049977,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045602650494709786,
      "loss": 5.4769,
      "mean_token_accuracy": 0.15975380837917327,
      "num_tokens": 6260896.0,
      "step": 3025
    },
    {
      "entropy": 5.573398494720459,
      "epoch": 2.752385279418446,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045581639130736173,
      "loss": 5.4287,
      "mean_token_accuracy": 0.1637202247977257,
      "num_tokens": 6271828.0,
      "step": 3030
    },
    {
      "entropy": 5.664141130447388,
      "epoch": 2.756928668786915,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004556058315211546,
      "loss": 5.4673,
      "mean_token_accuracy": 0.16932979524135588,
      "num_tokens": 6282698.0,
      "step": 3035
    },
    {
      "entropy": 5.551248216629029,
      "epoch": 2.761472058155384,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045539482610801156,
      "loss": 5.3764,
      "mean_token_accuracy": 0.17497740387916566,
      "num_tokens": 6291905.0,
      "step": 3040
    },
    {
      "entropy": 5.48179783821106,
      "epoch": 2.766015447523853,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004551833755885677,
      "loss": 5.4123,
      "mean_token_accuracy": 0.1668097898364067,
      "num_tokens": 6301579.0,
      "step": 3045
    },
    {
      "entropy": 5.518378877639771,
      "epoch": 2.7705588368923215,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004549714804845561,
      "loss": 5.3715,
      "mean_token_accuracy": 0.16212380230426787,
      "num_tokens": 6311209.0,
      "step": 3050
    },
    {
      "entropy": 5.491114473342895,
      "epoch": 2.7751022262607905,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004547591413188068,
      "loss": 5.3534,
      "mean_token_accuracy": 0.16724447011947632,
      "num_tokens": 6321601.0,
      "step": 3055
    },
    {
      "entropy": 5.594778347015381,
      "epoch": 2.7796456156292595,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004545463586152456,
      "loss": 5.4584,
      "mean_token_accuracy": 0.16972106993198394,
      "num_tokens": 6331070.0,
      "step": 3060
    },
    {
      "entropy": 5.598646020889282,
      "epoch": 2.784189004997728,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004543331328988927,
      "loss": 5.4393,
      "mean_token_accuracy": 0.16212439090013503,
      "num_tokens": 6341772.0,
      "step": 3065
    },
    {
      "entropy": 5.4714454174041744,
      "epoch": 2.788732394366197,
      "grad_norm": 1.25,
      "learning_rate": 0.0004541194646958614,
      "loss": 5.3942,
      "mean_token_accuracy": 0.16854195296764374,
      "num_tokens": 6352237.0,
      "step": 3070
    },
    {
      "entropy": 5.572699880599975,
      "epoch": 2.793275783734666,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045390535453335666,
      "loss": 5.5098,
      "mean_token_accuracy": 0.1657037004828453,
      "num_tokens": 6363645.0,
      "step": 3075
    },
    {
      "entropy": 5.5819111347198485,
      "epoch": 2.797819173103135,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045369080293967414,
      "loss": 5.3818,
      "mean_token_accuracy": 0.1713142290711403,
      "num_tokens": 6373559.0,
      "step": 3080
    },
    {
      "entropy": 5.572905397415161,
      "epoch": 2.802362562471604,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045347581044419846,
      "loss": 5.3914,
      "mean_token_accuracy": 0.16971153467893602,
      "num_tokens": 6384197.0,
      "step": 3085
    },
    {
      "entropy": 5.44353551864624,
      "epoch": 2.806905951840073,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045326037757740216,
      "loss": 5.3066,
      "mean_token_accuracy": 0.17053664550185205,
      "num_tokens": 6394834.0,
      "step": 3090
    },
    {
      "entropy": 5.472598361968994,
      "epoch": 2.8114493412085415,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004530445048708447,
      "loss": 5.3794,
      "mean_token_accuracy": 0.17127571552991866,
      "num_tokens": 6405515.0,
      "step": 3095
    },
    {
      "entropy": 5.5797937393188475,
      "epoch": 2.8159927305770105,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045282819285717024,
      "loss": 5.4198,
      "mean_token_accuracy": 0.17315615862607955,
      "num_tokens": 6415885.0,
      "step": 3100
    },
    {
      "entropy": 5.487499332427978,
      "epoch": 2.8205361199454795,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045261144207010736,
      "loss": 5.34,
      "mean_token_accuracy": 0.1776031032204628,
      "num_tokens": 6426053.0,
      "step": 3105
    },
    {
      "entropy": 5.544284534454346,
      "epoch": 2.825079509313948,
      "grad_norm": 1.5,
      "learning_rate": 0.00045239425304446684,
      "loss": 5.4591,
      "mean_token_accuracy": 0.16497154980897905,
      "num_tokens": 6436024.0,
      "step": 3110
    },
    {
      "entropy": 5.609182500839234,
      "epoch": 2.829622898682417,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004521766263161413,
      "loss": 5.4205,
      "mean_token_accuracy": 0.16752688437700272,
      "num_tokens": 6446847.0,
      "step": 3115
    },
    {
      "entropy": 5.473289823532104,
      "epoch": 2.834166288050886,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000451958562422103,
      "loss": 5.3313,
      "mean_token_accuracy": 0.17451879680156707,
      "num_tokens": 6456901.0,
      "step": 3120
    },
    {
      "entropy": 5.523167419433594,
      "epoch": 2.838709677419355,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045174006190040267,
      "loss": 5.3794,
      "mean_token_accuracy": 0.16624522507190703,
      "num_tokens": 6467308.0,
      "step": 3125
    },
    {
      "entropy": 5.510180139541626,
      "epoch": 2.843253066787824,
      "grad_norm": 1.28125,
      "learning_rate": 0.000451521125290169,
      "loss": 5.3414,
      "mean_token_accuracy": 0.17632320374250413,
      "num_tokens": 6477071.0,
      "step": 3130
    },
    {
      "entropy": 5.504618835449219,
      "epoch": 2.8477964561562925,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000451301753131606,
      "loss": 5.3967,
      "mean_token_accuracy": 0.16972736418247222,
      "num_tokens": 6488300.0,
      "step": 3135
    },
    {
      "entropy": 5.536497020721436,
      "epoch": 2.8523398455247615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045108194596599286,
      "loss": 5.4455,
      "mean_token_accuracy": 0.1610207736492157,
      "num_tokens": 6499515.0,
      "step": 3140
    },
    {
      "entropy": 5.490173101425171,
      "epoch": 2.8568832348932305,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004508617043356818,
      "loss": 5.3803,
      "mean_token_accuracy": 0.1675855800509453,
      "num_tokens": 6509167.0,
      "step": 3145
    },
    {
      "entropy": 5.514344978332519,
      "epoch": 2.861426624261699,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004506410287840972,
      "loss": 5.4204,
      "mean_token_accuracy": 0.16500514447689058,
      "num_tokens": 6520899.0,
      "step": 3150
    },
    {
      "entropy": 5.538662052154541,
      "epoch": 2.865970013630168,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000450419919855734,
      "loss": 5.4109,
      "mean_token_accuracy": 0.16202673316001892,
      "num_tokens": 6531293.0,
      "step": 3155
    },
    {
      "entropy": 5.6065332889556885,
      "epoch": 2.870513402998637,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004501983780961566,
      "loss": 5.4285,
      "mean_token_accuracy": 0.16660867929458617,
      "num_tokens": 6541039.0,
      "step": 3160
    },
    {
      "entropy": 5.4971050262451175,
      "epoch": 2.875056792367106,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004499764040519972,
      "loss": 5.3832,
      "mean_token_accuracy": 0.17120896875858307,
      "num_tokens": 6551948.0,
      "step": 3165
    },
    {
      "entropy": 5.511898946762085,
      "epoch": 2.879600181735575,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044975399827095467,
      "loss": 5.3115,
      "mean_token_accuracy": 0.17716679126024246,
      "num_tokens": 6562158.0,
      "step": 3170
    },
    {
      "entropy": 5.579851913452148,
      "epoch": 2.8841435711040435,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044953116130179313,
      "loss": 5.4085,
      "mean_token_accuracy": 0.16111837700009346,
      "num_tokens": 6572605.0,
      "step": 3175
    },
    {
      "entropy": 5.441836261749268,
      "epoch": 2.8886869604725125,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004493078936943408,
      "loss": 5.3846,
      "mean_token_accuracy": 0.17459261119365693,
      "num_tokens": 6583247.0,
      "step": 3180
    },
    {
      "entropy": 5.58513126373291,
      "epoch": 2.8932303498409815,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004490841959994882,
      "loss": 5.4919,
      "mean_token_accuracy": 0.16135846823453903,
      "num_tokens": 6593371.0,
      "step": 3185
    },
    {
      "entropy": 5.530911922454834,
      "epoch": 2.89777373920945,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044886006876918716,
      "loss": 5.2965,
      "mean_token_accuracy": 0.1691173195838928,
      "num_tokens": 6603777.0,
      "step": 3190
    },
    {
      "entropy": 5.606493377685547,
      "epoch": 2.902317128577919,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004486355125564493,
      "loss": 5.5567,
      "mean_token_accuracy": 0.16609083116054535,
      "num_tokens": 6615654.0,
      "step": 3195
    },
    {
      "entropy": 5.5463879108428955,
      "epoch": 2.906860517946388,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004484105279153449,
      "loss": 5.4119,
      "mean_token_accuracy": 0.16749218702316285,
      "num_tokens": 6626079.0,
      "step": 3200
    },
    {
      "entropy": 5.52779860496521,
      "epoch": 2.911403907314857,
      "grad_norm": 1.28125,
      "learning_rate": 0.000448185115401001,
      "loss": 5.3116,
      "mean_token_accuracy": 0.18257694989442824,
      "num_tokens": 6636047.0,
      "step": 3205
    },
    {
      "entropy": 5.494665479660034,
      "epoch": 2.915947296683326,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004479592755696008,
      "loss": 5.4977,
      "mean_token_accuracy": 0.1567457526922226,
      "num_tokens": 6646670.0,
      "step": 3210
    },
    {
      "entropy": 5.582828140258789,
      "epoch": 2.9204906860517945,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044773300897838137,
      "loss": 5.4899,
      "mean_token_accuracy": 0.16843743473291398,
      "num_tokens": 6658027.0,
      "step": 3215
    },
    {
      "entropy": 5.57952561378479,
      "epoch": 2.9250340754202635,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004475063161856332,
      "loss": 5.4095,
      "mean_token_accuracy": 0.1673821046948433,
      "num_tokens": 6668446.0,
      "step": 3220
    },
    {
      "entropy": 5.5199387550354,
      "epoch": 2.9295774647887325,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004472791977506983,
      "loss": 5.4117,
      "mean_token_accuracy": 0.17177652865648269,
      "num_tokens": 6679038.0,
      "step": 3225
    },
    {
      "entropy": 5.509660291671753,
      "epoch": 2.934120854157201,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044705165423396855,
      "loss": 5.3331,
      "mean_token_accuracy": 0.17690296173095704,
      "num_tokens": 6689207.0,
      "step": 3230
    },
    {
      "entropy": 5.469997262954712,
      "epoch": 2.93866424352567,
      "grad_norm": 1.359375,
      "learning_rate": 0.00044682368619688523,
      "loss": 5.3712,
      "mean_token_accuracy": 0.1679947555065155,
      "num_tokens": 6699264.0,
      "step": 3235
    },
    {
      "entropy": 5.485982608795166,
      "epoch": 2.943207632894139,
      "grad_norm": 1.25,
      "learning_rate": 0.0004465952942019367,
      "loss": 5.3969,
      "mean_token_accuracy": 0.16441409885883332,
      "num_tokens": 6709309.0,
      "step": 3240
    },
    {
      "entropy": 5.529012012481689,
      "epoch": 2.947751022262608,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044636647881265745,
      "loss": 5.385,
      "mean_token_accuracy": 0.1702449291944504,
      "num_tokens": 6719302.0,
      "step": 3245
    },
    {
      "entropy": 5.568566846847534,
      "epoch": 2.952294411631077,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004461372405936269,
      "loss": 5.3333,
      "mean_token_accuracy": 0.16730671972036362,
      "num_tokens": 6730668.0,
      "step": 3250
    },
    {
      "entropy": 5.4671618938446045,
      "epoch": 2.9568378009995455,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004459075801104673,
      "loss": 5.3489,
      "mean_token_accuracy": 0.17044253200292586,
      "num_tokens": 6740140.0,
      "step": 3255
    },
    {
      "entropy": 5.474802112579345,
      "epoch": 2.9613811903680145,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044567749792984325,
      "loss": 5.3597,
      "mean_token_accuracy": 0.1695680782198906,
      "num_tokens": 6751864.0,
      "step": 3260
    },
    {
      "entropy": 5.5141009330749515,
      "epoch": 2.9659245797364835,
      "grad_norm": 1.25,
      "learning_rate": 0.00044544699461945966,
      "loss": 5.4925,
      "mean_token_accuracy": 0.16408627927303315,
      "num_tokens": 6762375.0,
      "step": 3265
    },
    {
      "entropy": 5.54395604133606,
      "epoch": 2.970467969104952,
      "grad_norm": 1.25,
      "learning_rate": 0.00044521607074806037,
      "loss": 5.4166,
      "mean_token_accuracy": 0.17094823718070984,
      "num_tokens": 6772181.0,
      "step": 3270
    },
    {
      "entropy": 5.57254548072815,
      "epoch": 2.975011358473421,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004449847268854272,
      "loss": 5.4285,
      "mean_token_accuracy": 0.1662304550409317,
      "num_tokens": 6782402.0,
      "step": 3275
    },
    {
      "entropy": 5.484041738510132,
      "epoch": 2.97955474784189,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004447529636023782,
      "loss": 5.3829,
      "mean_token_accuracy": 0.1662940114736557,
      "num_tokens": 6792671.0,
      "step": 3280
    },
    {
      "entropy": 5.434405326843262,
      "epoch": 2.984098137210359,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044452078147076614,
      "loss": 5.2693,
      "mean_token_accuracy": 0.1770182028412819,
      "num_tokens": 6802111.0,
      "step": 3285
    },
    {
      "entropy": 5.504476499557495,
      "epoch": 2.988641526578828,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004442881810634774,
      "loss": 5.3511,
      "mean_token_accuracy": 0.17470426559448243,
      "num_tokens": 6812651.0,
      "step": 3290
    },
    {
      "entropy": 5.537303352355957,
      "epoch": 2.9931849159472965,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004440551629544303,
      "loss": 5.3924,
      "mean_token_accuracy": 0.16632127463817598,
      "num_tokens": 6823208.0,
      "step": 3295
    },
    {
      "entropy": 5.535593557357788,
      "epoch": 2.9977283053157655,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044382172771857383,
      "loss": 5.3789,
      "mean_token_accuracy": 0.17128838002681732,
      "num_tokens": 6833412.0,
      "step": 3300
    },
    {
      "entropy": 5.540709654490153,
      "epoch": 3.0018173557473875,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004435878759318864,
      "loss": 5.265,
      "mean_token_accuracy": 0.17887886530823177,
      "num_tokens": 6841977.0,
      "step": 3305
    },
    {
      "entropy": 5.5094132900238035,
      "epoch": 3.0063607451158565,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004433536081713738,
      "loss": 5.0816,
      "mean_token_accuracy": 0.17909709215164185,
      "num_tokens": 6852571.0,
      "step": 3310
    },
    {
      "entropy": 5.488974523544312,
      "epoch": 3.0109041344843255,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004431189250150686,
      "loss": 5.1075,
      "mean_token_accuracy": 0.17693999409675598,
      "num_tokens": 6861451.0,
      "step": 3315
    },
    {
      "entropy": 5.476665830612182,
      "epoch": 3.015447523852794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044288382704202816,
      "loss": 5.0814,
      "mean_token_accuracy": 0.1793387085199356,
      "num_tokens": 6872134.0,
      "step": 3320
    },
    {
      "entropy": 5.460846853256226,
      "epoch": 3.019990913221263,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044264831483233335,
      "loss": 5.1374,
      "mean_token_accuracy": 0.18726807236671447,
      "num_tokens": 6882477.0,
      "step": 3325
    },
    {
      "entropy": 5.439997911453247,
      "epoch": 3.024534302589732,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00044241238896708716,
      "loss": 5.073,
      "mean_token_accuracy": 0.18652296364307402,
      "num_tokens": 6892233.0,
      "step": 3330
    },
    {
      "entropy": 5.474528646469116,
      "epoch": 3.029077691958201,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004421760500284131,
      "loss": 5.0399,
      "mean_token_accuracy": 0.18419712036848068,
      "num_tokens": 6902090.0,
      "step": 3335
    },
    {
      "entropy": 5.481957197189331,
      "epoch": 3.0336210813266695,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004419392985994541,
      "loss": 5.1602,
      "mean_token_accuracy": 0.1848383754491806,
      "num_tokens": 6912295.0,
      "step": 3340
    },
    {
      "entropy": 5.552234935760498,
      "epoch": 3.0381644706951385,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044170213526437076,
      "loss": 5.1685,
      "mean_token_accuracy": 0.1694626122713089,
      "num_tokens": 6923064.0,
      "step": 3345
    },
    {
      "entropy": 5.526348495483399,
      "epoch": 3.0427078600636075,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044146456060833994,
      "loss": 5.1001,
      "mean_token_accuracy": 0.18083677887916566,
      "num_tokens": 6932825.0,
      "step": 3350
    },
    {
      "entropy": 5.368307161331177,
      "epoch": 3.0472512494320765,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044122657521755373,
      "loss": 5.0098,
      "mean_token_accuracy": 0.1848729133605957,
      "num_tokens": 6943188.0,
      "step": 3355
    },
    {
      "entropy": 5.403898143768311,
      "epoch": 3.051794638800545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044098817967921716,
      "loss": 5.1121,
      "mean_token_accuracy": 0.1794737011194229,
      "num_tokens": 6953864.0,
      "step": 3360
    },
    {
      "entropy": 5.474752712249756,
      "epoch": 3.056338028169014,
      "grad_norm": 1.25,
      "learning_rate": 0.00044074937458154763,
      "loss": 5.1022,
      "mean_token_accuracy": 0.1805911809206009,
      "num_tokens": 6963745.0,
      "step": 3365
    },
    {
      "entropy": 5.459611558914185,
      "epoch": 3.060881417537483,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044051016051377303,
      "loss": 5.122,
      "mean_token_accuracy": 0.1852324590086937,
      "num_tokens": 6973933.0,
      "step": 3370
    },
    {
      "entropy": 5.375621891021728,
      "epoch": 3.065424806905952,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004402705380661302,
      "loss": 5.0778,
      "mean_token_accuracy": 0.17894607186317443,
      "num_tokens": 6984314.0,
      "step": 3375
    },
    {
      "entropy": 5.492203569412231,
      "epoch": 3.0699681962744205,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044003050782986377,
      "loss": 5.2269,
      "mean_token_accuracy": 0.1728239580988884,
      "num_tokens": 6995921.0,
      "step": 3380
    },
    {
      "entropy": 5.586938333511353,
      "epoch": 3.0745115856428895,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004397900703972245,
      "loss": 5.1982,
      "mean_token_accuracy": 0.17636808156967163,
      "num_tokens": 7006910.0,
      "step": 3385
    },
    {
      "entropy": 5.467364645004272,
      "epoch": 3.0790549750113585,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004395492263614677,
      "loss": 5.0757,
      "mean_token_accuracy": 0.18143328577280043,
      "num_tokens": 7017229.0,
      "step": 3390
    },
    {
      "entropy": 5.436608743667603,
      "epoch": 3.0835983643798275,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043930797631685234,
      "loss": 5.1651,
      "mean_token_accuracy": 0.1796605408191681,
      "num_tokens": 7027934.0,
      "step": 3395
    },
    {
      "entropy": 5.451813697814941,
      "epoch": 3.088141753748296,
      "grad_norm": 1.375,
      "learning_rate": 0.00043906632085863876,
      "loss": 5.0813,
      "mean_token_accuracy": 0.18021165132522582,
      "num_tokens": 7037964.0,
      "step": 3400
    },
    {
      "entropy": 5.474063730239868,
      "epoch": 3.092685143116765,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043882426058308775,
      "loss": 5.0967,
      "mean_token_accuracy": 0.18498752415180206,
      "num_tokens": 7048275.0,
      "step": 3405
    },
    {
      "entropy": 5.366586685180664,
      "epoch": 3.097228532485234,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00043858179608745915,
      "loss": 5.1038,
      "mean_token_accuracy": 0.18191172629594804,
      "num_tokens": 7058488.0,
      "step": 3410
    },
    {
      "entropy": 5.509993934631348,
      "epoch": 3.101771921853703,
      "grad_norm": 1.375,
      "learning_rate": 0.0004383389279700098,
      "loss": 5.1491,
      "mean_token_accuracy": 0.18080411851406097,
      "num_tokens": 7069195.0,
      "step": 3415
    },
    {
      "entropy": 5.410095930099487,
      "epoch": 3.1063153112221715,
      "grad_norm": 1.515625,
      "learning_rate": 0.00043809565682999274,
      "loss": 5.0752,
      "mean_token_accuracy": 0.18355673998594285,
      "num_tokens": 7078858.0,
      "step": 3420
    },
    {
      "entropy": 5.434135675430298,
      "epoch": 3.1108587005906405,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00043785198326765535,
      "loss": 5.1029,
      "mean_token_accuracy": 0.17662401646375656,
      "num_tokens": 7088551.0,
      "step": 3425
    },
    {
      "entropy": 5.413759469985962,
      "epoch": 3.1154020899591095,
      "grad_norm": 1.25,
      "learning_rate": 0.00043760790788423785,
      "loss": 5.0901,
      "mean_token_accuracy": 0.1862627938389778,
      "num_tokens": 7098987.0,
      "step": 3430
    },
    {
      "entropy": 5.383126974105835,
      "epoch": 3.1199454793275785,
      "grad_norm": 1.375,
      "learning_rate": 0.00043736343128197207,
      "loss": 5.104,
      "mean_token_accuracy": 0.17984799593687056,
      "num_tokens": 7108337.0,
      "step": 3435
    },
    {
      "entropy": 5.464860963821411,
      "epoch": 3.124488868696047,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004371185540640796,
      "loss": 5.1439,
      "mean_token_accuracy": 0.18218799382448198,
      "num_tokens": 7118267.0,
      "step": 3440
    },
    {
      "entropy": 5.441714859008789,
      "epoch": 3.129032258064516,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043687327683477077,
      "loss": 5.1756,
      "mean_token_accuracy": 0.17961294054985047,
      "num_tokens": 7129780.0,
      "step": 3445
    },
    {
      "entropy": 5.458058166503906,
      "epoch": 3.133575647432985,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004366276001992427,
      "loss": 5.1413,
      "mean_token_accuracy": 0.1809666559100151,
      "num_tokens": 7139581.0,
      "step": 3450
    },
    {
      "entropy": 5.448301219940186,
      "epoch": 3.138119036801454,
      "grad_norm": 2.546875,
      "learning_rate": 0.000436381524763678,
      "loss": 5.1167,
      "mean_token_accuracy": 0.18601667881011963,
      "num_tokens": 7150093.0,
      "step": 3455
    },
    {
      "entropy": 5.389939212799073,
      "epoch": 3.142662426169923,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004361350511352435,
      "loss": 5.0676,
      "mean_token_accuracy": 0.1786562457680702,
      "num_tokens": 7159668.0,
      "step": 3460
    },
    {
      "entropy": 5.4402008056640625,
      "epoch": 3.1472058155383915,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004358881799220882,
      "loss": 5.1694,
      "mean_token_accuracy": 0.17355237156152725,
      "num_tokens": 7169713.0,
      "step": 3465
    },
    {
      "entropy": 5.41139783859253,
      "epoch": 3.1517492049068605,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043564091173334236,
      "loss": 5.1442,
      "mean_token_accuracy": 0.17021278142929078,
      "num_tokens": 7180681.0,
      "step": 3470
    },
    {
      "entropy": 5.462332677841187,
      "epoch": 3.1562925942753295,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043539324717911573,
      "loss": 5.02,
      "mean_token_accuracy": 0.19202615469694137,
      "num_tokens": 7189922.0,
      "step": 3475
    },
    {
      "entropy": 5.423739957809448,
      "epoch": 3.1608359836437985,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00043514518687049587,
      "loss": 5.1031,
      "mean_token_accuracy": 0.18684335500001908,
      "num_tokens": 7200388.0,
      "step": 3480
    },
    {
      "entropy": 5.453259754180908,
      "epoch": 3.165379373012267,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000434896731419547,
      "loss": 5.1431,
      "mean_token_accuracy": 0.18176197707653047,
      "num_tokens": 7211219.0,
      "step": 3485
    },
    {
      "entropy": 5.367387485504151,
      "epoch": 3.169922762380736,
      "grad_norm": 1.5,
      "learning_rate": 0.0004346478814393082,
      "loss": 4.942,
      "mean_token_accuracy": 0.1913049712777138,
      "num_tokens": 7220585.0,
      "step": 3490
    },
    {
      "entropy": 5.399928379058838,
      "epoch": 3.174466151749205,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004343986375437921,
      "loss": 5.1156,
      "mean_token_accuracy": 0.18510342985391617,
      "num_tokens": 7231123.0,
      "step": 3495
    },
    {
      "entropy": 5.409286117553711,
      "epoch": 3.179009541117674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043414900034798314,
      "loss": 5.1353,
      "mean_token_accuracy": 0.17770394682884216,
      "num_tokens": 7241802.0,
      "step": 3500
    },
    {
      "epoch": 3.179009541117674,
      "eval_entropy": 5.313468411911366,
      "eval_loss": 5.845177173614502,
      "eval_mean_token_accuracy": 0.15812873211364414,
      "eval_num_tokens": 7241802.0,
      "eval_runtime": 2.604,
      "eval_samples_per_second": 1319.121,
      "eval_steps_per_second": 165.13,
      "step": 3500
    },
    {
      "entropy": 5.459087657928467,
      "epoch": 3.1835529304861425,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004338989704678364,
      "loss": 5.1887,
      "mean_token_accuracy": 0.17418979108333588,
      "num_tokens": 7252014.0,
      "step": 3505
    },
    {
      "entropy": 5.4540942192077635,
      "epoch": 3.1880963198546115,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00043364854852027566,
      "loss": 5.1155,
      "mean_token_accuracy": 0.18346473425626755,
      "num_tokens": 7261738.0,
      "step": 3510
    },
    {
      "entropy": 5.43152346611023,
      "epoch": 3.1926397092230805,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004333977351231924,
      "loss": 5.1644,
      "mean_token_accuracy": 0.1816312238574028,
      "num_tokens": 7272342.0,
      "step": 3515
    },
    {
      "entropy": 5.32980899810791,
      "epoch": 3.1971830985915495,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043314653089544336,
      "loss": 5.0726,
      "mean_token_accuracy": 0.186733441054821,
      "num_tokens": 7282519.0,
      "step": 3520
    },
    {
      "entropy": 5.445510339736939,
      "epoch": 3.201726487960018,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004328949364568504,
      "loss": 5.1601,
      "mean_token_accuracy": 0.1821300521492958,
      "num_tokens": 7292675.0,
      "step": 3525
    },
    {
      "entropy": 5.5610816955566404,
      "epoch": 3.206269877328487,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004326429524281975,
      "loss": 5.1921,
      "mean_token_accuracy": 0.1733242854475975,
      "num_tokens": 7302220.0,
      "step": 3530
    },
    {
      "entropy": 5.450060939788818,
      "epoch": 3.210813266696956,
      "grad_norm": 1.375,
      "learning_rate": 0.0004323905794312303,
      "loss": 5.1087,
      "mean_token_accuracy": 0.18218960762023925,
      "num_tokens": 7311942.0,
      "step": 3535
    },
    {
      "entropy": 5.374439907073975,
      "epoch": 3.215356656065425,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043213781808865414,
      "loss": 5.0586,
      "mean_token_accuracy": 0.18685071170330048,
      "num_tokens": 7322083.0,
      "step": 3540
    },
    {
      "entropy": 5.366559839248657,
      "epoch": 3.2199000454338935,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004318846690241324,
      "loss": 5.0903,
      "mean_token_accuracy": 0.18440251797437668,
      "num_tokens": 7332291.0,
      "step": 3545
    },
    {
      "entropy": 5.409432220458984,
      "epoch": 3.2244434348023625,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043163113286228533,
      "loss": 5.0203,
      "mean_token_accuracy": 0.18952652364969252,
      "num_tokens": 7341702.0,
      "step": 3550
    },
    {
      "entropy": 5.401936340332031,
      "epoch": 3.2289868241708315,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004313772102286882,
      "loss": 5.0959,
      "mean_token_accuracy": 0.17808228582143784,
      "num_tokens": 7353332.0,
      "step": 3555
    },
    {
      "entropy": 5.32733302116394,
      "epoch": 3.2335302135393005,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004311229017498698,
      "loss": 5.0521,
      "mean_token_accuracy": 0.18500913679599762,
      "num_tokens": 7362965.0,
      "step": 3560
    },
    {
      "entropy": 5.446574020385742,
      "epoch": 3.238073602907769,
      "grad_norm": 1.328125,
      "learning_rate": 0.000430868208053311,
      "loss": 5.1895,
      "mean_token_accuracy": 0.17038872987031936,
      "num_tokens": 7374621.0,
      "step": 3565
    },
    {
      "entropy": 5.498407459259033,
      "epoch": 3.242616992276238,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043061312976744326,
      "loss": 5.1729,
      "mean_token_accuracy": 0.17913734167814255,
      "num_tokens": 7385826.0,
      "step": 3570
    },
    {
      "entropy": 5.432439231872559,
      "epoch": 3.247160381644707,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004303576675216469,
      "loss": 5.2191,
      "mean_token_accuracy": 0.17355716228485107,
      "num_tokens": 7397136.0,
      "step": 3575
    },
    {
      "entropy": 5.383523893356323,
      "epoch": 3.251703771013176,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043010182194624957,
      "loss": 5.0732,
      "mean_token_accuracy": 0.17954282313585282,
      "num_tokens": 7407126.0,
      "step": 3580
    },
    {
      "entropy": 5.414077377319336,
      "epoch": 3.2562471603816445,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004298455936725248,
      "loss": 5.0954,
      "mean_token_accuracy": 0.18238571435213088,
      "num_tokens": 7416719.0,
      "step": 3585
    },
    {
      "entropy": 5.381393194198608,
      "epoch": 3.2607905497501135,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004295889833326903,
      "loss": 5.128,
      "mean_token_accuracy": 0.18338222652673722,
      "num_tokens": 7426428.0,
      "step": 3590
    },
    {
      "entropy": 5.380539751052856,
      "epoch": 3.2653339391185825,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00042933199155990675,
      "loss": 5.1095,
      "mean_token_accuracy": 0.18013755232095718,
      "num_tokens": 7436877.0,
      "step": 3595
    },
    {
      "entropy": 5.530236911773682,
      "epoch": 3.2698773284870515,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004290746189882757,
      "loss": 5.2083,
      "mean_token_accuracy": 0.18230529725551606,
      "num_tokens": 7447726.0,
      "step": 3600
    },
    {
      "entropy": 5.434997701644898,
      "epoch": 3.27442071785552,
      "grad_norm": 2.109375,
      "learning_rate": 0.00042881686625283844,
      "loss": 5.1597,
      "mean_token_accuracy": 0.1741008758544922,
      "num_tokens": 7458201.0,
      "step": 3605
    },
    {
      "entropy": 5.441771125793457,
      "epoch": 3.278964107223989,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004285587339895742,
      "loss": 5.1529,
      "mean_token_accuracy": 0.17846778333187102,
      "num_tokens": 7467969.0,
      "step": 3610
    },
    {
      "entropy": 5.426313161849976,
      "epoch": 3.283507496592458,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042830022283539863,
      "loss": 5.2089,
      "mean_token_accuracy": 0.17596002221107482,
      "num_tokens": 7478545.0,
      "step": 3615
    },
    {
      "entropy": 5.4976049900054935,
      "epoch": 3.288050885960927,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004280413334281623,
      "loss": 5.1195,
      "mean_token_accuracy": 0.19189976751804352,
      "num_tokens": 7488318.0,
      "step": 3620
    },
    {
      "entropy": 5.52264723777771,
      "epoch": 3.292594275329396,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042778206640664915,
      "loss": 5.1514,
      "mean_token_accuracy": 0.18155771642923355,
      "num_tokens": 7497964.0,
      "step": 3625
    },
    {
      "entropy": 5.456651782989502,
      "epoch": 3.2971376646978645,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004275224224105748,
      "loss": 5.0865,
      "mean_token_accuracy": 0.18504077047109604,
      "num_tokens": 7508288.0,
      "step": 3630
    },
    {
      "entropy": 5.273019409179687,
      "epoch": 3.3016810540663335,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00042726240208058484,
      "loss": 5.1622,
      "mean_token_accuracy": 0.18209049850702286,
      "num_tokens": 7518716.0,
      "step": 3635
    },
    {
      "entropy": 5.479461908340454,
      "epoch": 3.3062244434348025,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004270020060582538,
      "loss": 5.1278,
      "mean_token_accuracy": 0.18468791395425796,
      "num_tokens": 7529165.0,
      "step": 3640
    },
    {
      "entropy": 5.37908353805542,
      "epoch": 3.310767832803271,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042674123498608285,
      "loss": 5.0995,
      "mean_token_accuracy": 0.17720864415168763,
      "num_tokens": 7539975.0,
      "step": 3645
    },
    {
      "entropy": 5.419282579421997,
      "epoch": 3.31531122217174,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00042648008950749877,
      "loss": 5.1253,
      "mean_token_accuracy": 0.1843060150742531,
      "num_tokens": 7549997.0,
      "step": 3650
    },
    {
      "entropy": 5.3928557395935055,
      "epoch": 3.319854611540209,
      "grad_norm": 1.53125,
      "learning_rate": 0.000426218570266852,
      "loss": 5.1369,
      "mean_token_accuracy": 0.18057385534048082,
      "num_tokens": 7559631.0,
      "step": 3655
    },
    {
      "entropy": 5.321335124969482,
      "epoch": 3.324398000908678,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042595667790941536,
      "loss": 5.085,
      "mean_token_accuracy": 0.1799587219953537,
      "num_tokens": 7569523.0,
      "step": 3660
    },
    {
      "entropy": 5.407132625579834,
      "epoch": 3.328941390277147,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042569441308138204,
      "loss": 5.068,
      "mean_token_accuracy": 0.1892345815896988,
      "num_tokens": 7580339.0,
      "step": 3665
    },
    {
      "entropy": 5.378613471984863,
      "epoch": 3.3334847796456155,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004254317764298645,
      "loss": 5.1348,
      "mean_token_accuracy": 0.18169948309659958,
      "num_tokens": 7591265.0,
      "step": 3670
    },
    {
      "entropy": 5.4041777610778805,
      "epoch": 3.3380281690140845,
      "grad_norm": 1.375,
      "learning_rate": 0.0004251687686028927,
      "loss": 5.1745,
      "mean_token_accuracy": 0.18491033017635344,
      "num_tokens": 7600518.0,
      "step": 3675
    },
    {
      "entropy": 5.469471406936646,
      "epoch": 3.3425715583825535,
      "grad_norm": 1.46875,
      "learning_rate": 0.00042490539024941216,
      "loss": 5.2818,
      "mean_token_accuracy": 0.17820267975330353,
      "num_tokens": 7611568.0,
      "step": 3680
    },
    {
      "entropy": 5.4590137004852295,
      "epoch": 3.347114947751022,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042464164201928286,
      "loss": 5.2074,
      "mean_token_accuracy": 0.18618203550577164,
      "num_tokens": 7621514.0,
      "step": 3685
    },
    {
      "entropy": 5.431461572647095,
      "epoch": 3.351658337119491,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00042437752456327734,
      "loss": 5.1676,
      "mean_token_accuracy": 0.18389933556318283,
      "num_tokens": 7632249.0,
      "step": 3690
    },
    {
      "entropy": 5.41857943534851,
      "epoch": 3.35620172648796,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042411303853307907,
      "loss": 5.1625,
      "mean_token_accuracy": 0.17991655617952346,
      "num_tokens": 7643539.0,
      "step": 3695
    },
    {
      "entropy": 5.419274997711182,
      "epoch": 3.360745115856429,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004238481845812811,
      "loss": 5.1548,
      "mean_token_accuracy": 0.17926737517118455,
      "num_tokens": 7654089.0,
      "step": 3700
    },
    {
      "entropy": 5.424456310272217,
      "epoch": 3.365288505224898,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042358296336138436,
      "loss": 5.0705,
      "mean_token_accuracy": 0.18664083778858184,
      "num_tokens": 7665421.0,
      "step": 3705
    },
    {
      "entropy": 5.358485746383667,
      "epoch": 3.3698318945933665,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004233173755277956,
      "loss": 5.1545,
      "mean_token_accuracy": 0.18582481890916824,
      "num_tokens": 7675971.0,
      "step": 3710
    },
    {
      "entropy": 5.400942754745484,
      "epoch": 3.3743752839618355,
      "grad_norm": 1.375,
      "learning_rate": 0.00042305142173582657,
      "loss": 5.1206,
      "mean_token_accuracy": 0.18684681504964828,
      "num_tokens": 7687444.0,
      "step": 3715
    },
    {
      "entropy": 5.431884622573852,
      "epoch": 3.3789186733303045,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042278510264169165,
      "loss": 5.1643,
      "mean_token_accuracy": 0.17761679887771606,
      "num_tokens": 7697700.0,
      "step": 3720
    },
    {
      "entropy": 5.397919416427612,
      "epoch": 3.3834620626987735,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042251841890250696,
      "loss": 5.059,
      "mean_token_accuracy": 0.17931891232728958,
      "num_tokens": 7708118.0,
      "step": 3725
    },
    {
      "entropy": 5.388166713714599,
      "epoch": 3.388005452067242,
      "grad_norm": 1.28125,
      "learning_rate": 0.000422251371176288,
      "loss": 5.0886,
      "mean_token_accuracy": 0.1841619625687599,
      "num_tokens": 7718558.0,
      "step": 3730
    },
    {
      "entropy": 5.4224144458770756,
      "epoch": 3.392548841435711,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042198396012194854,
      "loss": 5.1382,
      "mean_token_accuracy": 0.18727860748767852,
      "num_tokens": 7727873.0,
      "step": 3735
    },
    {
      "entropy": 5.486453485488892,
      "epoch": 3.39709223080418,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042171618639929865,
      "loss": 5.2016,
      "mean_token_accuracy": 0.17451137006282808,
      "num_tokens": 7737885.0,
      "step": 3740
    },
    {
      "entropy": 5.394660997390747,
      "epoch": 3.401635620172649,
      "grad_norm": 1.375,
      "learning_rate": 0.00042144805066904345,
      "loss": 5.1002,
      "mean_token_accuracy": 0.1860642671585083,
      "num_tokens": 7748423.0,
      "step": 3745
    },
    {
      "entropy": 5.412290239334107,
      "epoch": 3.4061790095411175,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042117955359278125,
      "loss": 5.2128,
      "mean_token_accuracy": 0.18704972118139268,
      "num_tokens": 7758879.0,
      "step": 3750
    },
    {
      "entropy": 5.491046285629272,
      "epoch": 3.4107223989095865,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042091069583300177,
      "loss": 5.1995,
      "mean_token_accuracy": 0.17518750727176666,
      "num_tokens": 7768765.0,
      "step": 3755
    },
    {
      "entropy": 5.353374290466308,
      "epoch": 3.4152657882780555,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004206414780530849,
      "loss": 5.0954,
      "mean_token_accuracy": 0.18172456473112106,
      "num_tokens": 7778680.0,
      "step": 3760
    },
    {
      "entropy": 5.382426071166992,
      "epoch": 3.4198091776465245,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004203719009172987,
      "loss": 5.1307,
      "mean_token_accuracy": 0.18617654144763945,
      "num_tokens": 7787423.0,
      "step": 3765
    },
    {
      "entropy": 5.421677970886231,
      "epoch": 3.424352567014993,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000420101965090798,
      "loss": 5.149,
      "mean_token_accuracy": 0.18102338612079621,
      "num_tokens": 7797358.0,
      "step": 3770
    },
    {
      "entropy": 5.490096235275269,
      "epoch": 3.428895956383462,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004198316712396226,
      "loss": 5.167,
      "mean_token_accuracy": 0.17768990397453308,
      "num_tokens": 7807865.0,
      "step": 3775
    },
    {
      "entropy": 5.442388582229614,
      "epoch": 3.433439345751931,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041956102003069573,
      "loss": 5.1171,
      "mean_token_accuracy": 0.1787397965788841,
      "num_tokens": 7818363.0,
      "step": 3780
    },
    {
      "entropy": 5.390089797973633,
      "epoch": 3.4379827351204,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004192900121318223,
      "loss": 5.123,
      "mean_token_accuracy": 0.1854868695139885,
      "num_tokens": 7827956.0,
      "step": 3785
    },
    {
      "entropy": 5.427382755279541,
      "epoch": 3.4425261244888685,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004190186482116874,
      "loss": 5.2473,
      "mean_token_accuracy": 0.1723328113555908,
      "num_tokens": 7838963.0,
      "step": 3790
    },
    {
      "entropy": 5.403203773498535,
      "epoch": 3.4470695138573375,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004187469289398545,
      "loss": 5.0979,
      "mean_token_accuracy": 0.1880764052271843,
      "num_tokens": 7850230.0,
      "step": 3795
    },
    {
      "entropy": 5.367580413818359,
      "epoch": 3.4516129032258065,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041847485498676385,
      "loss": 5.0296,
      "mean_token_accuracy": 0.18469575345516204,
      "num_tokens": 7861514.0,
      "step": 3800
    },
    {
      "entropy": 5.3711700439453125,
      "epoch": 3.4561562925942755,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041820242702373094,
      "loss": 5.1605,
      "mean_token_accuracy": 0.1821753591299057,
      "num_tokens": 7871175.0,
      "step": 3805
    },
    {
      "entropy": 5.384536361694336,
      "epoch": 3.460699681962744,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004179296457229448,
      "loss": 5.1821,
      "mean_token_accuracy": 0.177734437584877,
      "num_tokens": 7882312.0,
      "step": 3810
    },
    {
      "entropy": 5.426406764984131,
      "epoch": 3.465243071331213,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004176565117574659,
      "loss": 5.1066,
      "mean_token_accuracy": 0.18111293017864227,
      "num_tokens": 7892904.0,
      "step": 3815
    },
    {
      "entropy": 5.429520034790039,
      "epoch": 3.469786460699682,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004173830258012255,
      "loss": 5.1749,
      "mean_token_accuracy": 0.17856699526309966,
      "num_tokens": 7903127.0,
      "step": 3820
    },
    {
      "entropy": 5.408882665634155,
      "epoch": 3.474329850068151,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004171091885290226,
      "loss": 5.1742,
      "mean_token_accuracy": 0.17925683557987213,
      "num_tokens": 7914041.0,
      "step": 3825
    },
    {
      "entropy": 5.370095109939575,
      "epoch": 3.4788732394366195,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004168350006165238,
      "loss": 5.1651,
      "mean_token_accuracy": 0.17356557995080948,
      "num_tokens": 7924334.0,
      "step": 3830
    },
    {
      "entropy": 5.497560214996338,
      "epoch": 3.4834166288050885,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041656046274026034,
      "loss": 5.1818,
      "mean_token_accuracy": 0.18052341043949127,
      "num_tokens": 7935973.0,
      "step": 3835
    },
    {
      "entropy": 5.513583564758301,
      "epoch": 3.4879600181735575,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004162855755776272,
      "loss": 5.1546,
      "mean_token_accuracy": 0.18214287459850312,
      "num_tokens": 7946780.0,
      "step": 3840
    },
    {
      "entropy": 5.418990850448608,
      "epoch": 3.4925034075420265,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041601033980688106,
      "loss": 5.1972,
      "mean_token_accuracy": 0.18121838718652725,
      "num_tokens": 7957243.0,
      "step": 3845
    },
    {
      "entropy": 5.561002779006958,
      "epoch": 3.497046796910495,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004157347561071389,
      "loss": 5.3004,
      "mean_token_accuracy": 0.17513239830732347,
      "num_tokens": 7968160.0,
      "step": 3850
    },
    {
      "entropy": 5.387019920349121,
      "epoch": 3.501590186278964,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041545882515837604,
      "loss": 5.1991,
      "mean_token_accuracy": 0.17311442643404007,
      "num_tokens": 7977910.0,
      "step": 3855
    },
    {
      "entropy": 5.468769073486328,
      "epoch": 3.506133575647433,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00041518254764142477,
      "loss": 5.1493,
      "mean_token_accuracy": 0.17978033423423767,
      "num_tokens": 7988239.0,
      "step": 3860
    },
    {
      "entropy": 5.402276706695557,
      "epoch": 3.510676965015902,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00041490592423797226,
      "loss": 5.1235,
      "mean_token_accuracy": 0.18456532806158066,
      "num_tokens": 7998128.0,
      "step": 3865
    },
    {
      "entropy": 5.3204028606414795,
      "epoch": 3.515220354384371,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004146289556305594,
      "loss": 5.0988,
      "mean_token_accuracy": 0.18473509401082994,
      "num_tokens": 8008997.0,
      "step": 3870
    },
    {
      "entropy": 5.4050568580627445,
      "epoch": 3.5197637437528395,
      "grad_norm": 1.25,
      "learning_rate": 0.0004143516425025786,
      "loss": 5.1754,
      "mean_token_accuracy": 0.18440306633710862,
      "num_tokens": 8018864.0,
      "step": 3875
    },
    {
      "entropy": 5.446254634857178,
      "epoch": 3.5243071331213085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004140739855382724,
      "loss": 5.2073,
      "mean_token_accuracy": 0.1731021970510483,
      "num_tokens": 8029420.0,
      "step": 3880
    },
    {
      "entropy": 5.465769958496094,
      "epoch": 3.5288505224897775,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041379598542273197,
      "loss": 5.2406,
      "mean_token_accuracy": 0.17769331037998198,
      "num_tokens": 8040184.0,
      "step": 3885
    },
    {
      "entropy": 5.491438245773315,
      "epoch": 3.533393911858246,
      "grad_norm": 1.4375,
      "learning_rate": 0.00041351764284189464,
      "loss": 5.1647,
      "mean_token_accuracy": 0.1828169494867325,
      "num_tokens": 8050291.0,
      "step": 3890
    },
    {
      "entropy": 5.3932257175445555,
      "epoch": 3.537937301226715,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004132389584825433,
      "loss": 5.1504,
      "mean_token_accuracy": 0.18313387632369996,
      "num_tokens": 8060399.0,
      "step": 3895
    },
    {
      "entropy": 5.381469249725342,
      "epoch": 3.542480690595184,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041295993303230383,
      "loss": 5.1512,
      "mean_token_accuracy": 0.18886881470680236,
      "num_tokens": 8070986.0,
      "step": 3900
    },
    {
      "entropy": 5.452720308303833,
      "epoch": 3.547024079963653,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004126805671796438,
      "loss": 5.1088,
      "mean_token_accuracy": 0.1841768279671669,
      "num_tokens": 8080872.0,
      "step": 3905
    },
    {
      "entropy": 5.460559749603272,
      "epoch": 3.551567469332122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004124008616138707,
      "loss": 5.2131,
      "mean_token_accuracy": 0.17973160445690156,
      "num_tokens": 8091618.0,
      "step": 3910
    },
    {
      "entropy": 5.436691570281982,
      "epoch": 3.5561108587005905,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041212081702513015,
      "loss": 5.1481,
      "mean_token_accuracy": 0.1762390747666359,
      "num_tokens": 8101904.0,
      "step": 3915
    },
    {
      "entropy": 5.32893967628479,
      "epoch": 3.5606542480690595,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004118404341044044,
      "loss": 5.1033,
      "mean_token_accuracy": 0.18413669764995574,
      "num_tokens": 8112879.0,
      "step": 3920
    },
    {
      "entropy": 5.342809438705444,
      "epoch": 3.5651976374375285,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00041155971354351034,
      "loss": 5.1289,
      "mean_token_accuracy": 0.191913703083992,
      "num_tokens": 8123104.0,
      "step": 3925
    },
    {
      "entropy": 5.377686929702759,
      "epoch": 3.569741026805997,
      "grad_norm": 1.8125,
      "learning_rate": 0.00041127865603509815,
      "loss": 5.1217,
      "mean_token_accuracy": 0.18634375035762787,
      "num_tokens": 8132983.0,
      "step": 3930
    },
    {
      "entropy": 5.3964580535888675,
      "epoch": 3.574284416174466,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004109972622726493,
      "loss": 5.0894,
      "mean_token_accuracy": 0.18219338357448578,
      "num_tokens": 8143344.0,
      "step": 3935
    },
    {
      "entropy": 5.445954608917236,
      "epoch": 3.578827805542935,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041071553295047485,
      "loss": 5.1359,
      "mean_token_accuracy": 0.18180595338344574,
      "num_tokens": 8153561.0,
      "step": 3940
    },
    {
      "entropy": 5.34302921295166,
      "epoch": 3.583371194911404,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004104334687637139,
      "loss": 5.11,
      "mean_token_accuracy": 0.18399689793586732,
      "num_tokens": 8163713.0,
      "step": 3945
    },
    {
      "entropy": 5.2936101913452145,
      "epoch": 3.587914584279873,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041015107040833187,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19066124707460402,
      "num_tokens": 8174540.0,
      "step": 3950
    },
    {
      "entropy": 5.366580772399902,
      "epoch": 3.5924579736483415,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004098683385811185,
      "loss": 5.0824,
      "mean_token_accuracy": 0.18540124595165253,
      "num_tokens": 8185254.0,
      "step": 3955
    },
    {
      "entropy": 5.47531476020813,
      "epoch": 3.5970013630168105,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00040958527397968666,
      "loss": 5.1518,
      "mean_token_accuracy": 0.1791024163365364,
      "num_tokens": 8195205.0,
      "step": 3960
    },
    {
      "entropy": 5.415002536773682,
      "epoch": 3.6015447523852795,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004093018773024699,
      "loss": 5.2177,
      "mean_token_accuracy": 0.178206005692482,
      "num_tokens": 8205298.0,
      "step": 3965
    },
    {
      "entropy": 5.311694574356079,
      "epoch": 3.606088141753748,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004090181492487215,
      "loss": 5.139,
      "mean_token_accuracy": 0.1830941304564476,
      "num_tokens": 8215677.0,
      "step": 3970
    },
    {
      "entropy": 5.3269463062286375,
      "epoch": 3.610631531122217,
      "grad_norm": 1.40625,
      "learning_rate": 0.00040873409051851223,
      "loss": 5.1021,
      "mean_token_accuracy": 0.18233130872249603,
      "num_tokens": 8226938.0,
      "step": 3975
    },
    {
      "entropy": 5.449914836883545,
      "epoch": 3.615174920490686,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004084497018127286,
      "loss": 5.1411,
      "mean_token_accuracy": 0.17622751891613006,
      "num_tokens": 8238718.0,
      "step": 3980
    },
    {
      "entropy": 5.392190217971802,
      "epoch": 3.619718309859155,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004081649838330717,
      "loss": 5.0881,
      "mean_token_accuracy": 0.18608758896589278,
      "num_tokens": 8249562.0,
      "step": 3985
    },
    {
      "entropy": 5.4045857906341555,
      "epoch": 3.624261699227624,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040787993728205464,
      "loss": 5.1814,
      "mean_token_accuracy": 0.1786041796207428,
      "num_tokens": 8259736.0,
      "step": 3990
    },
    {
      "entropy": 5.439118814468384,
      "epoch": 3.6288050885960925,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00040759456286300154,
      "loss": 5.153,
      "mean_token_accuracy": 0.18659620583057404,
      "num_tokens": 8269933.0,
      "step": 3995
    },
    {
      "entropy": 5.480475950241089,
      "epoch": 3.6333484779645615,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040730886128004544,
      "loss": 5.1947,
      "mean_token_accuracy": 0.180545412003994,
      "num_tokens": 8279950.0,
      "step": 4000
    },
    {
      "epoch": 3.6333484779645615,
      "eval_entropy": 5.35828699511151,
      "eval_loss": 5.805370330810547,
      "eval_mean_token_accuracy": 0.16226694103936817,
      "eval_num_tokens": 8279950.0,
      "eval_runtime": 2.6016,
      "eval_samples_per_second": 1320.347,
      "eval_steps_per_second": 165.284,
      "step": 4000
    },
    {
      "entropy": 5.456209182739258,
      "epoch": 3.6378918673330305,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004070228332381266,
      "loss": 5.1388,
      "mean_token_accuracy": 0.1835985854268074,
      "num_tokens": 8289575.0,
      "step": 4005
    },
    {
      "entropy": 5.3499644756317135,
      "epoch": 3.6424352567014995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004067364794429908,
      "loss": 5.171,
      "mean_token_accuracy": 0.1789727121591568,
      "num_tokens": 8300809.0,
      "step": 4010
    },
    {
      "entropy": 5.424611616134643,
      "epoch": 3.646978646069968,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004064498006011874,
      "loss": 5.16,
      "mean_token_accuracy": 0.18115946501493455,
      "num_tokens": 8310666.0,
      "step": 4015
    },
    {
      "entropy": 5.4026069164276125,
      "epoch": 3.651522035438437,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040616279742006813,
      "loss": 5.1155,
      "mean_token_accuracy": 0.17834189534187317,
      "num_tokens": 8320823.0,
      "step": 4020
    },
    {
      "entropy": 5.439352893829346,
      "epoch": 3.656065424806906,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00040587547060778466,
      "loss": 5.1693,
      "mean_token_accuracy": 0.18191349059343337,
      "num_tokens": 8330343.0,
      "step": 4025
    },
    {
      "entropy": 5.3338744163513185,
      "epoch": 3.660608814175375,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004055878208732874,
      "loss": 5.064,
      "mean_token_accuracy": 0.1878042623400688,
      "num_tokens": 8341444.0,
      "step": 4030
    },
    {
      "entropy": 5.418221044540405,
      "epoch": 3.665152203543844,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040529984892632363,
      "loss": 5.1634,
      "mean_token_accuracy": 0.1753498524427414,
      "num_tokens": 8351654.0,
      "step": 4035
    },
    {
      "entropy": 5.394876766204834,
      "epoch": 3.6696955929123125,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004050115554774353,
      "loss": 5.1405,
      "mean_token_accuracy": 0.1814640522003174,
      "num_tokens": 8361534.0,
      "step": 4040
    },
    {
      "entropy": 5.360474348068237,
      "epoch": 3.6742389822807815,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004047229412379578,
      "loss": 5.1602,
      "mean_token_accuracy": 0.18625522404909134,
      "num_tokens": 8371955.0,
      "step": 4045
    },
    {
      "entropy": 5.4311682224273685,
      "epoch": 3.6787823716492505,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004044340069200183,
      "loss": 5.1842,
      "mean_token_accuracy": 0.18379201143980026,
      "num_tokens": 8383014.0,
      "step": 4050
    },
    {
      "entropy": 5.386659717559814,
      "epoch": 3.683325761017719,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040414475323653326,
      "loss": 5.152,
      "mean_token_accuracy": 0.1857267290353775,
      "num_tokens": 8393997.0,
      "step": 4055
    },
    {
      "entropy": 5.351330137252807,
      "epoch": 3.687869150386188,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004038551809012076,
      "loss": 5.1256,
      "mean_token_accuracy": 0.17812232673168182,
      "num_tokens": 8404712.0,
      "step": 4060
    },
    {
      "entropy": 5.4687525749206545,
      "epoch": 3.692412539754657,
      "grad_norm": 1.375,
      "learning_rate": 0.0004035652906285322,
      "loss": 5.247,
      "mean_token_accuracy": 0.17488803565502167,
      "num_tokens": 8416610.0,
      "step": 4065
    },
    {
      "entropy": 5.387298965454102,
      "epoch": 3.696955929123126,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040327508313378244,
      "loss": 5.1137,
      "mean_token_accuracy": 0.18516299724578858,
      "num_tokens": 8427096.0,
      "step": 4070
    },
    {
      "entropy": 5.448663330078125,
      "epoch": 3.701499318491595,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004029845591330164,
      "loss": 5.1328,
      "mean_token_accuracy": 0.17973324358463288,
      "num_tokens": 8438748.0,
      "step": 4075
    },
    {
      "entropy": 5.500763368606568,
      "epoch": 3.7060427078600635,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040269371934307336,
      "loss": 5.2151,
      "mean_token_accuracy": 0.181552018225193,
      "num_tokens": 8449293.0,
      "step": 4080
    },
    {
      "entropy": 5.418303155899048,
      "epoch": 3.7105860972285325,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00040240256448157146,
      "loss": 5.1381,
      "mean_token_accuracy": 0.17968246191740037,
      "num_tokens": 8459465.0,
      "step": 4085
    },
    {
      "entropy": 5.2986259937286375,
      "epoch": 3.7151294865970015,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004021110952669065,
      "loss": 5.1588,
      "mean_token_accuracy": 0.18253383338451384,
      "num_tokens": 8469589.0,
      "step": 4090
    },
    {
      "entropy": 5.3797430992126465,
      "epoch": 3.71967287596547,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004018193124182497,
      "loss": 5.1159,
      "mean_token_accuracy": 0.17903668731451033,
      "num_tokens": 8480201.0,
      "step": 4095
    },
    {
      "entropy": 5.399144172668457,
      "epoch": 3.724216265333939,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004015272166555462,
      "loss": 5.1432,
      "mean_token_accuracy": 0.1827358439564705,
      "num_tokens": 8492495.0,
      "step": 4100
    },
    {
      "entropy": 5.3984918117523195,
      "epoch": 3.728759654702408,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004012348086995132,
      "loss": 5.1794,
      "mean_token_accuracy": 0.18082428574562073,
      "num_tokens": 8503429.0,
      "step": 4105
    },
    {
      "entropy": 5.473850536346435,
      "epoch": 3.733303044070877,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004009420892716385,
      "loss": 5.174,
      "mean_token_accuracy": 0.17961965948343278,
      "num_tokens": 8513169.0,
      "step": 4110
    },
    {
      "entropy": 5.399222278594971,
      "epoch": 3.737846433439346,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004006490590941781,
      "loss": 5.1136,
      "mean_token_accuracy": 0.18261404037475587,
      "num_tokens": 8523192.0,
      "step": 4115
    },
    {
      "entropy": 5.327436351776123,
      "epoch": 3.7423898228078145,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004003557188901547,
      "loss": 5.0356,
      "mean_token_accuracy": 0.18966284543275833,
      "num_tokens": 8534410.0,
      "step": 4120
    },
    {
      "entropy": 5.328780555725098,
      "epoch": 3.7469332121762835,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004000620693833562,
      "loss": 5.1587,
      "mean_token_accuracy": 0.1865798845887184,
      "num_tokens": 8544317.0,
      "step": 4125
    },
    {
      "entropy": 5.338295316696167,
      "epoch": 3.7514766015447525,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003997681112983335,
      "loss": 5.0427,
      "mean_token_accuracy": 0.18043473809957505,
      "num_tokens": 8554730.0,
      "step": 4130
    },
    {
      "entropy": 5.349508476257324,
      "epoch": 3.756019990913221,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039947384536039905,
      "loss": 5.1038,
      "mean_token_accuracy": 0.18118361979722977,
      "num_tokens": 8564909.0,
      "step": 4135
    },
    {
      "entropy": 5.415514230728149,
      "epoch": 3.76056338028169,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039917927229562474,
      "loss": 5.1163,
      "mean_token_accuracy": 0.18912000954151154,
      "num_tokens": 8575290.0,
      "step": 4140
    },
    {
      "entropy": 5.390664386749267,
      "epoch": 3.765106769650159,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039888439283084035,
      "loss": 5.1579,
      "mean_token_accuracy": 0.18464586287736892,
      "num_tokens": 8585521.0,
      "step": 4145
    },
    {
      "entropy": 5.363944339752197,
      "epoch": 3.769650159018628,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039858920769363164,
      "loss": 5.0679,
      "mean_token_accuracy": 0.18088057339191438,
      "num_tokens": 8595049.0,
      "step": 4150
    },
    {
      "entropy": 5.342890739440918,
      "epoch": 3.774193548387097,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003982937176123386,
      "loss": 5.1528,
      "mean_token_accuracy": 0.18042400032281875,
      "num_tokens": 8604707.0,
      "step": 4155
    },
    {
      "entropy": 5.425475311279297,
      "epoch": 3.7787369377555655,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039799792331605364,
      "loss": 5.1126,
      "mean_token_accuracy": 0.1773986279964447,
      "num_tokens": 8615573.0,
      "step": 4160
    },
    {
      "entropy": 5.412333011627197,
      "epoch": 3.7832803271240345,
      "grad_norm": 1.546875,
      "learning_rate": 0.00039770182553461965,
      "loss": 5.1343,
      "mean_token_accuracy": 0.18454237282276154,
      "num_tokens": 8625799.0,
      "step": 4165
    },
    {
      "entropy": 5.374884843826294,
      "epoch": 3.7878237164925035,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00039740542499862866,
      "loss": 5.1108,
      "mean_token_accuracy": 0.18394931256771088,
      "num_tokens": 8635351.0,
      "step": 4170
    },
    {
      "entropy": 5.365823125839233,
      "epoch": 3.792367105860972,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003971087224394194,
      "loss": 5.2322,
      "mean_token_accuracy": 0.18046795725822448,
      "num_tokens": 8645765.0,
      "step": 4175
    },
    {
      "entropy": 5.436018180847168,
      "epoch": 3.796910495229441,
      "grad_norm": 1.375,
      "learning_rate": 0.0003968117185890761,
      "loss": 5.163,
      "mean_token_accuracy": 0.17681157290935517,
      "num_tokens": 8656170.0,
      "step": 4180
    },
    {
      "entropy": 5.384992551803589,
      "epoch": 3.80145388459791,
      "grad_norm": 1.78125,
      "learning_rate": 0.00039651441418042606,
      "loss": 5.0982,
      "mean_token_accuracy": 0.1894101619720459,
      "num_tokens": 8665704.0,
      "step": 4185
    },
    {
      "entropy": 5.41967134475708,
      "epoch": 3.805997273966379,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003962168099470384,
      "loss": 5.1689,
      "mean_token_accuracy": 0.18105421662330629,
      "num_tokens": 8676236.0,
      "step": 4190
    },
    {
      "entropy": 5.498148822784424,
      "epoch": 3.810540663334848,
      "grad_norm": 1.3125,
      "learning_rate": 0.000395918906623222,
      "loss": 5.1905,
      "mean_token_accuracy": 0.18399525433778763,
      "num_tokens": 8687242.0,
      "step": 4195
    },
    {
      "entropy": 5.377608585357666,
      "epoch": 3.8150840527033165,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003956207049440238,
      "loss": 5.1625,
      "mean_token_accuracy": 0.18241188675165176,
      "num_tokens": 8696967.0,
      "step": 4200
    },
    {
      "entropy": 5.367107295989991,
      "epoch": 3.8196274420717855,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003953222056452265,
      "loss": 5.0912,
      "mean_token_accuracy": 0.19142119884490966,
      "num_tokens": 8707832.0,
      "step": 4205
    },
    {
      "entropy": 5.319425582885742,
      "epoch": 3.8241708314402545,
      "grad_norm": 2.0625,
      "learning_rate": 0.0003950234094633479,
      "loss": 5.1455,
      "mean_token_accuracy": 0.1790603965520859,
      "num_tokens": 8717592.0,
      "step": 4210
    },
    {
      "entropy": 5.465997409820557,
      "epoch": 3.828714220808723,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039472431713563746,
      "loss": 5.1663,
      "mean_token_accuracy": 0.18136731684207916,
      "num_tokens": 8728899.0,
      "step": 4215
    },
    {
      "entropy": 5.3928474426269535,
      "epoch": 3.833257610177192,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000394424929400076,
      "loss": 5.1661,
      "mean_token_accuracy": 0.18066652864217758,
      "num_tokens": 8740092.0,
      "step": 4220
    },
    {
      "entropy": 5.397383117675782,
      "epoch": 3.837800999545661,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003941252469953729,
      "loss": 5.1464,
      "mean_token_accuracy": 0.18307925313711165,
      "num_tokens": 8750256.0,
      "step": 4225
    },
    {
      "entropy": 5.392667484283447,
      "epoch": 3.84234438891413,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003938252706609647,
      "loss": 5.1321,
      "mean_token_accuracy": 0.17983932197093963,
      "num_tokens": 8760546.0,
      "step": 4230
    },
    {
      "entropy": 5.467023324966431,
      "epoch": 3.846887778282599,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003935250011370132,
      "loss": 5.2278,
      "mean_token_accuracy": 0.17523436844348908,
      "num_tokens": 8770017.0,
      "step": 4235
    },
    {
      "entropy": 5.408269929885864,
      "epoch": 3.851431167651068,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003932244391644037,
      "loss": 5.1454,
      "mean_token_accuracy": 0.18087756037712097,
      "num_tokens": 8780597.0,
      "step": 4240
    },
    {
      "entropy": 5.394955205917358,
      "epoch": 3.8559745570195365,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003929235854847428,
      "loss": 5.1333,
      "mean_token_accuracy": 0.17753371745347976,
      "num_tokens": 8791508.0,
      "step": 4245
    },
    {
      "entropy": 5.456591320037842,
      "epoch": 3.8605179463880055,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039262244084035724,
      "loss": 5.143,
      "mean_token_accuracy": 0.1854093477129936,
      "num_tokens": 8803013.0,
      "step": 4250
    },
    {
      "entropy": 5.423864507675171,
      "epoch": 3.8650613357564745,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039232100597429143,
      "loss": 5.1417,
      "mean_token_accuracy": 0.1862639605998993,
      "num_tokens": 8812550.0,
      "step": 4255
    },
    {
      "entropy": 5.297554349899292,
      "epoch": 3.869604725124943,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039201928163030584,
      "loss": 5.0828,
      "mean_token_accuracy": 0.18953992277383805,
      "num_tokens": 8822700.0,
      "step": 4260
    },
    {
      "entropy": 5.3732829093933105,
      "epoch": 3.874148114493412,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039171726855287544,
      "loss": 5.1603,
      "mean_token_accuracy": 0.18270649760961533,
      "num_tokens": 8832859.0,
      "step": 4265
    },
    {
      "entropy": 5.374744319915772,
      "epoch": 3.878691503861881,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039141496748718735,
      "loss": 5.1184,
      "mean_token_accuracy": 0.1862223893404007,
      "num_tokens": 8843216.0,
      "step": 4270
    },
    {
      "entropy": 5.39664363861084,
      "epoch": 3.88323489323035,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003911123791791395,
      "loss": 5.0605,
      "mean_token_accuracy": 0.18764829337596894,
      "num_tokens": 8854791.0,
      "step": 4275
    },
    {
      "entropy": 5.346702718734742,
      "epoch": 3.887778282598819,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003908095043753384,
      "loss": 5.1149,
      "mean_token_accuracy": 0.18119485229253768,
      "num_tokens": 8865191.0,
      "step": 4280
    },
    {
      "entropy": 5.3600047588348385,
      "epoch": 3.8923216719672875,
      "grad_norm": 1.59375,
      "learning_rate": 0.00039050634382309753,
      "loss": 5.1193,
      "mean_token_accuracy": 0.1875446006655693,
      "num_tokens": 8875067.0,
      "step": 4285
    },
    {
      "entropy": 5.37948899269104,
      "epoch": 3.8968650613357565,
      "grad_norm": 1.65625,
      "learning_rate": 0.0003902028982704355,
      "loss": 5.143,
      "mean_token_accuracy": 0.18528290688991547,
      "num_tokens": 8884391.0,
      "step": 4290
    },
    {
      "entropy": 5.443444538116455,
      "epoch": 3.9014084507042255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003898991684660738,
      "loss": 5.231,
      "mean_token_accuracy": 0.17830575704574586,
      "num_tokens": 8895077.0,
      "step": 4295
    },
    {
      "entropy": 5.400945138931275,
      "epoch": 3.905951840072694,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003895951551594358,
      "loss": 5.1626,
      "mean_token_accuracy": 0.18274886459112166,
      "num_tokens": 8906754.0,
      "step": 4300
    },
    {
      "entropy": 5.370799493789673,
      "epoch": 3.910495229441163,
      "grad_norm": 1.4375,
      "learning_rate": 0.000389290859100644,
      "loss": 5.1114,
      "mean_token_accuracy": 0.18343409299850463,
      "num_tokens": 8917839.0,
      "step": 4305
    },
    {
      "entropy": 5.40346622467041,
      "epoch": 3.915038618809632,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003889862810405187,
      "loss": 5.1559,
      "mean_token_accuracy": 0.19021583944559098,
      "num_tokens": 8928555.0,
      "step": 4310
    },
    {
      "entropy": 5.388493967056275,
      "epoch": 3.919582008178101,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000388681421730576,
      "loss": 5.1163,
      "mean_token_accuracy": 0.17484002262353898,
      "num_tokens": 8938811.0,
      "step": 4315
    },
    {
      "entropy": 5.47366943359375,
      "epoch": 3.92412539754657,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003883762819230259,
      "loss": 5.2389,
      "mean_token_accuracy": 0.17729849815368653,
      "num_tokens": 8949302.0,
      "step": 4320
    },
    {
      "entropy": 5.4163741111755375,
      "epoch": 3.9286687869150385,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038807086237077057,
      "loss": 5.1088,
      "mean_token_accuracy": 0.18246268332004548,
      "num_tokens": 8959242.0,
      "step": 4325
    },
    {
      "entropy": 5.41228346824646,
      "epoch": 3.9332121762835075,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038776516382740237,
      "loss": 5.1574,
      "mean_token_accuracy": 0.18202644735574722,
      "num_tokens": 8969696.0,
      "step": 4330
    },
    {
      "entropy": 5.337202453613282,
      "epoch": 3.9377555656519765,
      "grad_norm": 1.9375,
      "learning_rate": 0.00038745918704720205,
      "loss": 5.1144,
      "mean_token_accuracy": 0.18682635128498076,
      "num_tokens": 8979458.0,
      "step": 4335
    },
    {
      "entropy": 5.347159099578858,
      "epoch": 3.942298955020445,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038715293278513693,
      "loss": 5.054,
      "mean_token_accuracy": 0.18760739266872406,
      "num_tokens": 8989851.0,
      "step": 4340
    },
    {
      "entropy": 5.448424053192139,
      "epoch": 3.946842344388914,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00038684640179685896,
      "loss": 5.285,
      "mean_token_accuracy": 0.17836133390665054,
      "num_tokens": 9000605.0,
      "step": 4345
    },
    {
      "entropy": 5.429435443878174,
      "epoch": 3.951385733757383,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003865395948387029,
      "loss": 5.1214,
      "mean_token_accuracy": 0.1846381887793541,
      "num_tokens": 9010578.0,
      "step": 4350
    },
    {
      "entropy": 5.385301828384399,
      "epoch": 3.955929123125852,
      "grad_norm": 1.578125,
      "learning_rate": 0.00038623251266768437,
      "loss": 5.0812,
      "mean_token_accuracy": 0.18727062940597533,
      "num_tokens": 9020697.0,
      "step": 4355
    },
    {
      "entropy": 5.386028385162353,
      "epoch": 3.960472512494321,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038592515604149815,
      "loss": 5.1609,
      "mean_token_accuracy": 0.17969390004873276,
      "num_tokens": 9030795.0,
      "step": 4360
    },
    {
      "entropy": 5.322641372680664,
      "epoch": 3.9650159018627895,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003856175257185162,
      "loss": 5.1277,
      "mean_token_accuracy": 0.17829941660165788,
      "num_tokens": 9041683.0,
      "step": 4365
    },
    {
      "entropy": 5.36119589805603,
      "epoch": 3.9695592912312585,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038530962245778564,
      "loss": 5.0816,
      "mean_token_accuracy": 0.186088490486145,
      "num_tokens": 9051167.0,
      "step": 4370
    },
    {
      "entropy": 5.358445405960083,
      "epoch": 3.9741026805997275,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003850014470190272,
      "loss": 5.1065,
      "mean_token_accuracy": 0.1862456262111664,
      "num_tokens": 9060514.0,
      "step": 4375
    },
    {
      "entropy": 5.402066802978515,
      "epoch": 3.978646069968196,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003846930001626332,
      "loss": 5.1363,
      "mean_token_accuracy": 0.17752173840999602,
      "num_tokens": 9070774.0,
      "step": 4380
    },
    {
      "entropy": 5.385051536560058,
      "epoch": 3.983189459336665,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003843842826496656,
      "loss": 5.0681,
      "mean_token_accuracy": 0.18622306287288665,
      "num_tokens": 9080718.0,
      "step": 4385
    },
    {
      "entropy": 5.261904287338257,
      "epoch": 3.987732848705134,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000384075295241854,
      "loss": 5.071,
      "mean_token_accuracy": 0.1758256122469902,
      "num_tokens": 9091183.0,
      "step": 4390
    },
    {
      "entropy": 5.377840423583985,
      "epoch": 3.992276238073603,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003837660387015944,
      "loss": 5.1282,
      "mean_token_accuracy": 0.18445907533168793,
      "num_tokens": 9101173.0,
      "step": 4395
    },
    {
      "entropy": 5.361019611358643,
      "epoch": 3.996819627442072,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003834565137919463,
      "loss": 5.1517,
      "mean_token_accuracy": 0.18319814056158065,
      "num_tokens": 9110724.0,
      "step": 4400
    },
    {
      "entropy": 5.390623728434245,
      "epoch": 4.000908677873694,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003831467212766318,
      "loss": 5.132,
      "mean_token_accuracy": 0.18870513968997532,
      "num_tokens": 9120410.0,
      "step": 4405
    },
    {
      "entropy": 5.398123264312744,
      "epoch": 4.005452067242163,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003828366619200331,
      "loss": 4.9732,
      "mean_token_accuracy": 0.1886272519826889,
      "num_tokens": 9130235.0,
      "step": 4410
    },
    {
      "entropy": 5.3377598285675045,
      "epoch": 4.009995456610632,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003825263364871908,
      "loss": 4.7304,
      "mean_token_accuracy": 0.20754059553146362,
      "num_tokens": 9140740.0,
      "step": 4415
    },
    {
      "entropy": 5.368291425704956,
      "epoch": 4.0145388459791,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003822157457438022,
      "loss": 4.9153,
      "mean_token_accuracy": 0.19531329423189164,
      "num_tokens": 9151271.0,
      "step": 4420
    },
    {
      "entropy": 5.324364852905274,
      "epoch": 4.019082235347569,
      "grad_norm": 1.375,
      "learning_rate": 0.00038190489045621893,
      "loss": 4.945,
      "mean_token_accuracy": 0.1921374171972275,
      "num_tokens": 9162520.0,
      "step": 4425
    },
    {
      "entropy": 5.397511291503906,
      "epoch": 4.023625624716038,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003815937713914455,
      "loss": 4.9288,
      "mean_token_accuracy": 0.1870266541838646,
      "num_tokens": 9173920.0,
      "step": 4430
    },
    {
      "entropy": 5.402887964248658,
      "epoch": 4.028169014084507,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038128238931713743,
      "loss": 4.9149,
      "mean_token_accuracy": 0.190347558259964,
      "num_tokens": 9184507.0,
      "step": 4435
    },
    {
      "entropy": 5.299306058883667,
      "epoch": 4.032712403452976,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038097074500159883,
      "loss": 4.9099,
      "mean_token_accuracy": 0.19192416816949845,
      "num_tokens": 9194971.0,
      "step": 4440
    },
    {
      "entropy": 5.285967206954956,
      "epoch": 4.037255792821445,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038065883921378123,
      "loss": 4.7599,
      "mean_token_accuracy": 0.20512468367815018,
      "num_tokens": 9205348.0,
      "step": 4445
    },
    {
      "entropy": 5.382515907287598,
      "epoch": 4.041799182189914,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003803466727232811,
      "loss": 5.0088,
      "mean_token_accuracy": 0.1923065721988678,
      "num_tokens": 9216301.0,
      "step": 4450
    },
    {
      "entropy": 5.353657817840576,
      "epoch": 4.046342571558383,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003800342463003383,
      "loss": 4.8575,
      "mean_token_accuracy": 0.1982938066124916,
      "num_tokens": 9226778.0,
      "step": 4455
    },
    {
      "entropy": 5.381874465942383,
      "epoch": 4.050885960926851,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037972156071583386,
      "loss": 4.9597,
      "mean_token_accuracy": 0.19581509232521058,
      "num_tokens": 9238309.0,
      "step": 4460
    },
    {
      "entropy": 5.314351511001587,
      "epoch": 4.05542935029532,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003794086167412884,
      "loss": 4.8988,
      "mean_token_accuracy": 0.2002609521150589,
      "num_tokens": 9247720.0,
      "step": 4465
    },
    {
      "entropy": 5.474795722961426,
      "epoch": 4.059972739663789,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003790954151488601,
      "loss": 4.9605,
      "mean_token_accuracy": 0.19488334208726882,
      "num_tokens": 9257519.0,
      "step": 4470
    },
    {
      "entropy": 5.386325979232788,
      "epoch": 4.064516129032258,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003787819567113427,
      "loss": 4.9417,
      "mean_token_accuracy": 0.1862177863717079,
      "num_tokens": 9268554.0,
      "step": 4475
    },
    {
      "entropy": 5.35470142364502,
      "epoch": 4.069059518400727,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003784682422021638,
      "loss": 4.9071,
      "mean_token_accuracy": 0.19462098479270934,
      "num_tokens": 9279354.0,
      "step": 4480
    },
    {
      "entropy": 5.352110528945923,
      "epoch": 4.073602907769196,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003781542723953828,
      "loss": 4.9135,
      "mean_token_accuracy": 0.20052281767129898,
      "num_tokens": 9290174.0,
      "step": 4485
    },
    {
      "entropy": 5.310688829421997,
      "epoch": 4.078146297137665,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00037784004806568885,
      "loss": 4.8646,
      "mean_token_accuracy": 0.19931808561086656,
      "num_tokens": 9300151.0,
      "step": 4490
    },
    {
      "entropy": 5.250537252426147,
      "epoch": 4.082689686506134,
      "grad_norm": 1.5,
      "learning_rate": 0.00037752556998839934,
      "loss": 4.9016,
      "mean_token_accuracy": 0.19709861129522324,
      "num_tokens": 9309868.0,
      "step": 4495
    },
    {
      "entropy": 5.4299815654754635,
      "epoch": 4.087233075874602,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003772108389394576,
      "loss": 4.9129,
      "mean_token_accuracy": 0.18753276765346527,
      "num_tokens": 9319753.0,
      "step": 4500
    },
    {
      "epoch": 4.087233075874602,
      "eval_entropy": 5.316827168575553,
      "eval_loss": 5.770237445831299,
      "eval_mean_token_accuracy": 0.16390313416026359,
      "eval_num_tokens": 9319753.0,
      "eval_runtime": 2.6064,
      "eval_samples_per_second": 1317.919,
      "eval_steps_per_second": 164.98,
      "step": 4500
    },
    {
      "entropy": 5.354657363891602,
      "epoch": 4.091776465243071,
      "grad_norm": 1.359375,
      "learning_rate": 0.00037689585569543137,
      "loss": 4.87,
      "mean_token_accuracy": 0.19929929971694946,
      "num_tokens": 9330293.0,
      "step": 4505
    },
    {
      "entropy": 5.246643829345703,
      "epoch": 4.09631985461154,
      "grad_norm": 1.71875,
      "learning_rate": 0.0003765806210335103,
      "loss": 4.8653,
      "mean_token_accuracy": 0.2043078511953354,
      "num_tokens": 9340791.0,
      "step": 4510
    },
    {
      "entropy": 5.248399829864502,
      "epoch": 4.100863243980009,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00037626513573150473,
      "loss": 4.8468,
      "mean_token_accuracy": 0.2090988740324974,
      "num_tokens": 9350685.0,
      "step": 4515
    },
    {
      "entropy": 5.334291791915893,
      "epoch": 4.105406633348478,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037594940056784315,
      "loss": 4.9036,
      "mean_token_accuracy": 0.1951570600271225,
      "num_tokens": 9360474.0,
      "step": 4520
    },
    {
      "entropy": 5.31734881401062,
      "epoch": 4.109950022716947,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003756334163215708,
      "loss": 4.9218,
      "mean_token_accuracy": 0.18663069009780883,
      "num_tokens": 9371389.0,
      "step": 4525
    },
    {
      "entropy": 5.366529130935669,
      "epoch": 4.114493412085416,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037531718377234723,
      "loss": 4.9168,
      "mean_token_accuracy": 0.19261382669210433,
      "num_tokens": 9381359.0,
      "step": 4530
    },
    {
      "entropy": 5.372603559494019,
      "epoch": 4.119036801453885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003750007037004451,
      "loss": 4.9081,
      "mean_token_accuracy": 0.19330682158470153,
      "num_tokens": 9391171.0,
      "step": 4535
    },
    {
      "entropy": 5.312482643127441,
      "epoch": 4.123580190822353,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003746839768867473,
      "loss": 4.9747,
      "mean_token_accuracy": 0.19027969241142273,
      "num_tokens": 9401912.0,
      "step": 4540
    },
    {
      "entropy": 5.370792198181152,
      "epoch": 4.128123580190822,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0003743670041127458,
      "loss": 4.934,
      "mean_token_accuracy": 0.1935926452279091,
      "num_tokens": 9413353.0,
      "step": 4545
    },
    {
      "entropy": 5.403579616546631,
      "epoch": 4.132666969559291,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003740497861605395,
      "loss": 4.8881,
      "mean_token_accuracy": 0.19373160153627395,
      "num_tokens": 9424236.0,
      "step": 4550
    },
    {
      "entropy": 5.2409583568573,
      "epoch": 4.13721035892776,
      "grad_norm": 1.328125,
      "learning_rate": 0.000373732323812832,
      "loss": 4.8888,
      "mean_token_accuracy": 0.19512705951929094,
      "num_tokens": 9434291.0,
      "step": 4555
    },
    {
      "entropy": 5.298729372024536,
      "epoch": 4.141753748296229,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00037341461785293025,
      "loss": 4.9339,
      "mean_token_accuracy": 0.1922853872179985,
      "num_tokens": 9445087.0,
      "step": 4560
    },
    {
      "entropy": 5.318222379684448,
      "epoch": 4.146297137664698,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003730966690647421,
      "loss": 4.8611,
      "mean_token_accuracy": 0.20207712799310684,
      "num_tokens": 9455479.0,
      "step": 4565
    },
    {
      "entropy": 5.300402545928955,
      "epoch": 4.150840527033167,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037277847823277444,
      "loss": 4.96,
      "mean_token_accuracy": 0.18749491274356841,
      "num_tokens": 9467709.0,
      "step": 4570
    },
    {
      "entropy": 5.355503892898559,
      "epoch": 4.155383916401636,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00037246004614213164,
      "loss": 4.9684,
      "mean_token_accuracy": 0.18480307161808013,
      "num_tokens": 9478004.0,
      "step": 4575
    },
    {
      "entropy": 5.335785579681397,
      "epoch": 4.159927305770104,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003721413735785132,
      "loss": 4.8295,
      "mean_token_accuracy": 0.20304200053215027,
      "num_tokens": 9486855.0,
      "step": 4580
    },
    {
      "entropy": 5.2842244625091555,
      "epoch": 4.164470695138573,
      "grad_norm": 1.453125,
      "learning_rate": 0.000371822461328212,
      "loss": 4.9754,
      "mean_token_accuracy": 0.18710989505052567,
      "num_tokens": 9497321.0,
      "step": 4585
    },
    {
      "entropy": 5.385933780670166,
      "epoch": 4.169014084507042,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037150331017811233,
      "loss": 4.9124,
      "mean_token_accuracy": 0.19370098114013673,
      "num_tokens": 9507299.0,
      "step": 4590
    },
    {
      "entropy": 5.318790006637573,
      "epoch": 4.173557473875511,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003711839209156878,
      "loss": 4.8352,
      "mean_token_accuracy": 0.20159285068511962,
      "num_tokens": 9517800.0,
      "step": 4595
    },
    {
      "entropy": 5.342543315887451,
      "epoch": 4.17810086324398,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003708642943289999,
      "loss": 4.8805,
      "mean_token_accuracy": 0.20139130502939223,
      "num_tokens": 9528215.0,
      "step": 4600
    },
    {
      "entropy": 5.2149711608886715,
      "epoch": 4.182644252612449,
      "grad_norm": 1.375,
      "learning_rate": 0.0003705444312066953,
      "loss": 4.8811,
      "mean_token_accuracy": 0.20167567431926728,
      "num_tokens": 9539515.0,
      "step": 4605
    },
    {
      "entropy": 5.272196912765503,
      "epoch": 4.187187641980918,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003702243323380045,
      "loss": 4.9267,
      "mean_token_accuracy": 0.19090737402439117,
      "num_tokens": 9548709.0,
      "step": 4610
    },
    {
      "entropy": 5.292052793502807,
      "epoch": 4.191731031349387,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036990399851273965,
      "loss": 4.9421,
      "mean_token_accuracy": 0.19588428139686584,
      "num_tokens": 9558578.0,
      "step": 4615
    },
    {
      "entropy": 5.31192364692688,
      "epoch": 4.196274420717856,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003695834305212925,
      "loss": 4.8921,
      "mean_token_accuracy": 0.19787038564682008,
      "num_tokens": 9568538.0,
      "step": 4620
    },
    {
      "entropy": 5.393190526962281,
      "epoch": 4.200817810086324,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036926262915463294,
      "loss": 4.9214,
      "mean_token_accuracy": 0.1884934663772583,
      "num_tokens": 9578819.0,
      "step": 4625
    },
    {
      "entropy": 5.28897123336792,
      "epoch": 4.205361199454793,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036894159520430616,
      "loss": 4.8529,
      "mean_token_accuracy": 0.1972287952899933,
      "num_tokens": 9589060.0,
      "step": 4630
    },
    {
      "entropy": 5.291293334960938,
      "epoch": 4.209904588823262,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036862032946243173,
      "loss": 4.9374,
      "mean_token_accuracy": 0.18974538892507553,
      "num_tokens": 9599332.0,
      "step": 4635
    },
    {
      "entropy": 5.394422101974487,
      "epoch": 4.214447978191731,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036829883272170087,
      "loss": 4.9352,
      "mean_token_accuracy": 0.19593384861946106,
      "num_tokens": 9610429.0,
      "step": 4640
    },
    {
      "entropy": 5.330147218704224,
      "epoch": 4.2189913675602,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036797710577537486,
      "loss": 4.9026,
      "mean_token_accuracy": 0.20103822350502015,
      "num_tokens": 9620187.0,
      "step": 4645
    },
    {
      "entropy": 5.228457927703857,
      "epoch": 4.223534756928669,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036765514941728295,
      "loss": 4.7991,
      "mean_token_accuracy": 0.2090354949235916,
      "num_tokens": 9629322.0,
      "step": 4650
    },
    {
      "entropy": 5.331879377365112,
      "epoch": 4.228078146297138,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003673329644418204,
      "loss": 4.9117,
      "mean_token_accuracy": 0.20229984819889069,
      "num_tokens": 9639157.0,
      "step": 4655
    },
    {
      "entropy": 5.277870559692383,
      "epoch": 4.232621535665607,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036701055164394664,
      "loss": 4.9106,
      "mean_token_accuracy": 0.19558399468660354,
      "num_tokens": 9649382.0,
      "step": 4660
    },
    {
      "entropy": 5.347518491744995,
      "epoch": 4.237164925034075,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036668791181918313,
      "loss": 4.9081,
      "mean_token_accuracy": 0.20189401954412461,
      "num_tokens": 9659239.0,
      "step": 4665
    },
    {
      "entropy": 5.40248064994812,
      "epoch": 4.241708314402544,
      "grad_norm": 1.5,
      "learning_rate": 0.0003663650457636117,
      "loss": 5.0039,
      "mean_token_accuracy": 0.18968315571546554,
      "num_tokens": 9669308.0,
      "step": 4670
    },
    {
      "entropy": 5.2448783874511715,
      "epoch": 4.246251703771013,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003660419542738722,
      "loss": 4.912,
      "mean_token_accuracy": 0.1881134405732155,
      "num_tokens": 9679825.0,
      "step": 4675
    },
    {
      "entropy": 5.331082534790039,
      "epoch": 4.250795093139482,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003657186381471608,
      "loss": 4.9213,
      "mean_token_accuracy": 0.19403446465730667,
      "num_tokens": 9689709.0,
      "step": 4680
    },
    {
      "entropy": 5.301193618774414,
      "epoch": 4.255338482507951,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00036539509818122765,
      "loss": 4.8869,
      "mean_token_accuracy": 0.19520628601312637,
      "num_tokens": 9699186.0,
      "step": 4685
    },
    {
      "entropy": 5.298560237884521,
      "epoch": 4.25988187187642,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003650713351743758,
      "loss": 4.8455,
      "mean_token_accuracy": 0.2010258600115776,
      "num_tokens": 9709967.0,
      "step": 4690
    },
    {
      "entropy": 5.37402663230896,
      "epoch": 4.264425261244889,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003647473499254581,
      "loss": 4.9892,
      "mean_token_accuracy": 0.1879942387342453,
      "num_tokens": 9720604.0,
      "step": 4695
    },
    {
      "entropy": 5.363400030136108,
      "epoch": 4.268968650613358,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000364423143233876,
      "loss": 5.0371,
      "mean_token_accuracy": 0.19304635375738144,
      "num_tokens": 9731962.0,
      "step": 4700
    },
    {
      "entropy": 5.345454359054566,
      "epoch": 4.273512039981826,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036409871589957723,
      "loss": 4.9088,
      "mean_token_accuracy": 0.20067238360643386,
      "num_tokens": 9742254.0,
      "step": 4705
    },
    {
      "entropy": 5.368442106246948,
      "epoch": 4.278055429350295,
      "grad_norm": 1.1875,
      "learning_rate": 0.000363774068723054,
      "loss": 4.9891,
      "mean_token_accuracy": 0.18724262565374375,
      "num_tokens": 9753166.0,
      "step": 4710
    },
    {
      "entropy": 5.247803831100464,
      "epoch": 4.282598818718764,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036344920250534106,
      "loss": 4.9049,
      "mean_token_accuracy": 0.19792281538248063,
      "num_tokens": 9764595.0,
      "step": 4715
    },
    {
      "entropy": 5.383092641830444,
      "epoch": 4.287142208087233,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003631241180480133,
      "loss": 4.9906,
      "mean_token_accuracy": 0.19331517219543456,
      "num_tokens": 9775829.0,
      "step": 4720
    },
    {
      "entropy": 5.258446168899536,
      "epoch": 4.291685597455702,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003627988161531845,
      "loss": 4.8562,
      "mean_token_accuracy": 0.19776624143123628,
      "num_tokens": 9786068.0,
      "step": 4725
    },
    {
      "entropy": 5.27303376197815,
      "epoch": 4.296228986824171,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003624732976235046,
      "loss": 4.9244,
      "mean_token_accuracy": 0.1953245773911476,
      "num_tokens": 9795757.0,
      "step": 4730
    },
    {
      "entropy": 5.326236200332642,
      "epoch": 4.30077237619264,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003621475632621584,
      "loss": 4.9586,
      "mean_token_accuracy": 0.19193981736898422,
      "num_tokens": 9806232.0,
      "step": 4735
    },
    {
      "entropy": 5.265902423858643,
      "epoch": 4.305315765561109,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003618216138728627,
      "loss": 4.9107,
      "mean_token_accuracy": 0.20131682902574538,
      "num_tokens": 9816251.0,
      "step": 4740
    },
    {
      "entropy": 5.309978246688843,
      "epoch": 4.309859154929577,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036149545025986567,
      "loss": 4.8357,
      "mean_token_accuracy": 0.19742674380540848,
      "num_tokens": 9825935.0,
      "step": 4745
    },
    {
      "entropy": 5.241865015029907,
      "epoch": 4.314402544298046,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036116907322794335,
      "loss": 5.0009,
      "mean_token_accuracy": 0.18832280188798906,
      "num_tokens": 9836575.0,
      "step": 4750
    },
    {
      "entropy": 5.365031719207764,
      "epoch": 4.318945933666515,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036084248358239857,
      "loss": 4.9476,
      "mean_token_accuracy": 0.19455160051584244,
      "num_tokens": 9846872.0,
      "step": 4755
    },
    {
      "entropy": 5.345770263671875,
      "epoch": 4.323489323034984,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036051568212905897,
      "loss": 4.933,
      "mean_token_accuracy": 0.19071408361196518,
      "num_tokens": 9858170.0,
      "step": 4760
    },
    {
      "entropy": 5.237937688827515,
      "epoch": 4.328032712403453,
      "grad_norm": 1.875,
      "learning_rate": 0.0003601886696742745,
      "loss": 4.8685,
      "mean_token_accuracy": 0.1951388120651245,
      "num_tokens": 9867497.0,
      "step": 4765
    },
    {
      "entropy": 5.240127658843994,
      "epoch": 4.332576101771922,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000359861447024916,
      "loss": 4.8419,
      "mean_token_accuracy": 0.19798148721456527,
      "num_tokens": 9878749.0,
      "step": 4770
    },
    {
      "entropy": 5.332259893417358,
      "epoch": 4.337119491140391,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00035953401498837285,
      "loss": 4.8228,
      "mean_token_accuracy": 0.20691393464803695,
      "num_tokens": 9888438.0,
      "step": 4775
    },
    {
      "entropy": 5.247419548034668,
      "epoch": 4.34166288050886,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035920637437255096,
      "loss": 4.9666,
      "mean_token_accuracy": 0.19074664860963822,
      "num_tokens": 9898657.0,
      "step": 4780
    },
    {
      "entropy": 5.258979845046997,
      "epoch": 4.346206269877328,
      "grad_norm": 1.390625,
      "learning_rate": 0.000358878525985871,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20148461312055588,
      "num_tokens": 9909012.0,
      "step": 4785
    },
    {
      "entropy": 5.390817737579345,
      "epoch": 4.350749659245797,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003585504706372663,
      "loss": 4.9764,
      "mean_token_accuracy": 0.19087023884058,
      "num_tokens": 9919515.0,
      "step": 4790
    },
    {
      "entropy": 5.259331941604614,
      "epoch": 4.355293048614266,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003582222091361808,
      "loss": 4.7949,
      "mean_token_accuracy": 0.20567953437566758,
      "num_tokens": 9928877.0,
      "step": 4795
    },
    {
      "entropy": 5.251314544677735,
      "epoch": 4.359836437982735,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003578937422925671,
      "loss": 4.8787,
      "mean_token_accuracy": 0.1995159700512886,
      "num_tokens": 9938344.0,
      "step": 4800
    },
    {
      "entropy": 5.3485204696655275,
      "epoch": 4.364379827351204,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003575650709168846,
      "loss": 4.9725,
      "mean_token_accuracy": 0.19541917145252227,
      "num_tokens": 9949541.0,
      "step": 4805
    },
    {
      "entropy": 5.297571897506714,
      "epoch": 4.368923216719673,
      "grad_norm": 1.25,
      "learning_rate": 0.00035723619582009715,
      "loss": 4.9256,
      "mean_token_accuracy": 0.19381688833236693,
      "num_tokens": 9959894.0,
      "step": 4810
    },
    {
      "entropy": 5.318052530288696,
      "epoch": 4.373466606088142,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035690711781367137,
      "loss": 4.8946,
      "mean_token_accuracy": 0.19770621955394746,
      "num_tokens": 9970291.0,
      "step": 4815
    },
    {
      "entropy": 5.320247316360474,
      "epoch": 4.378009995456611,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003565778377095744,
      "loss": 4.969,
      "mean_token_accuracy": 0.1938779056072235,
      "num_tokens": 9981581.0,
      "step": 4820
    },
    {
      "entropy": 5.249984502792358,
      "epoch": 4.38255338482508,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003562483563202724,
      "loss": 4.822,
      "mean_token_accuracy": 0.198915758728981,
      "num_tokens": 9991472.0,
      "step": 4825
    },
    {
      "entropy": 5.258716011047364,
      "epoch": 4.387096774193548,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00035591867445872784,
      "loss": 4.898,
      "mean_token_accuracy": 0.19729772508144378,
      "num_tokens": 10002682.0,
      "step": 4830
    },
    {
      "entropy": 5.277369356155395,
      "epoch": 4.391640163562017,
      "grad_norm": 1.234375,
      "learning_rate": 0.000355588792938398,
      "loss": 5.0024,
      "mean_token_accuracy": 0.18976609408855438,
      "num_tokens": 10013856.0,
      "step": 4835
    },
    {
      "entropy": 5.282770872116089,
      "epoch": 4.396183552930486,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003552587125732326,
      "loss": 4.9416,
      "mean_token_accuracy": 0.1907618135213852,
      "num_tokens": 10024750.0,
      "step": 4840
    },
    {
      "entropy": 5.37528247833252,
      "epoch": 4.400726942298955,
      "grad_norm": 1.453125,
      "learning_rate": 0.00035492843417767224,
      "loss": 4.9389,
      "mean_token_accuracy": 0.19605205953121185,
      "num_tokens": 10034891.0,
      "step": 4845
    },
    {
      "entropy": 5.320839309692383,
      "epoch": 4.405270331667424,
      "grad_norm": 1.3125,
      "learning_rate": 0.000354597958566646,
      "loss": 4.8821,
      "mean_token_accuracy": 0.2003989264369011,
      "num_tokens": 10044860.0,
      "step": 4850
    },
    {
      "entropy": 5.19720492362976,
      "epoch": 4.409813721035893,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035426728655556954,
      "loss": 4.9008,
      "mean_token_accuracy": 0.20077485591173172,
      "num_tokens": 10055959.0,
      "step": 4855
    },
    {
      "entropy": 5.303377485275268,
      "epoch": 4.414357110404362,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003539364189603433,
      "loss": 5.0087,
      "mean_token_accuracy": 0.1875552386045456,
      "num_tokens": 10067885.0,
      "step": 4860
    },
    {
      "entropy": 5.397436141967773,
      "epoch": 4.418900499772831,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00035360535659735006,
      "loss": 4.9935,
      "mean_token_accuracy": 0.19766906946897506,
      "num_tokens": 10078328.0,
      "step": 4865
    },
    {
      "entropy": 5.380247926712036,
      "epoch": 4.423443889141299,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003532741002834534,
      "loss": 4.8597,
      "mean_token_accuracy": 0.19596353322267532,
      "num_tokens": 10087880.0,
      "step": 4870
    },
    {
      "entropy": 5.245528507232666,
      "epoch": 4.427987278509768,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00035294265083599526,
      "loss": 4.9397,
      "mean_token_accuracy": 0.1911029651761055,
      "num_tokens": 10098432.0,
      "step": 4875
    },
    {
      "entropy": 5.2377357006073,
      "epoch": 4.432530667878237,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003526110090727942,
      "loss": 4.9127,
      "mean_token_accuracy": 0.20282569378614426,
      "num_tokens": 10108169.0,
      "step": 4880
    },
    {
      "entropy": 5.318375015258789,
      "epoch": 4.437074057246706,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003522791758121434,
      "loss": 4.9449,
      "mean_token_accuracy": 0.19013069719076156,
      "num_tokens": 10118580.0,
      "step": 4885
    },
    {
      "entropy": 5.247455072402954,
      "epoch": 4.441617446615175,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035194715187280835,
      "loss": 4.8073,
      "mean_token_accuracy": 0.20721604824066162,
      "num_tokens": 10128259.0,
      "step": 4890
    },
    {
      "entropy": 5.302587652206421,
      "epoch": 4.446160835983644,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003516149380740252,
      "loss": 4.926,
      "mean_token_accuracy": 0.1949806809425354,
      "num_tokens": 10137739.0,
      "step": 4895
    },
    {
      "entropy": 5.321853733062744,
      "epoch": 4.450704225352113,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003512825352354983,
      "loss": 4.952,
      "mean_token_accuracy": 0.19414915144443512,
      "num_tokens": 10148886.0,
      "step": 4900
    },
    {
      "entropy": 5.289352464675903,
      "epoch": 4.455247614720582,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00035094994417739884,
      "loss": 4.8845,
      "mean_token_accuracy": 0.19574244916439057,
      "num_tokens": 10158680.0,
      "step": 4905
    },
    {
      "entropy": 5.241608905792236,
      "epoch": 4.45979100408905,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000350617165720362,
      "loss": 4.8684,
      "mean_token_accuracy": 0.19666503518819808,
      "num_tokens": 10168893.0,
      "step": 4910
    },
    {
      "entropy": 5.1963512897491455,
      "epoch": 4.464334393457519,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035028420068548567,
      "loss": 4.8213,
      "mean_token_accuracy": 0.20612047165632247,
      "num_tokens": 10178781.0,
      "step": 4915
    },
    {
      "entropy": 5.274204301834106,
      "epoch": 4.468877782825988,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003499510498943279,
      "loss": 4.9404,
      "mean_token_accuracy": 0.191862653195858,
      "num_tokens": 10189233.0,
      "step": 4920
    },
    {
      "entropy": 5.293498182296753,
      "epoch": 4.473421172194457,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00034961771416890526,
      "loss": 4.8755,
      "mean_token_accuracy": 0.19937924146652222,
      "num_tokens": 10198374.0,
      "step": 4925
    },
    {
      "entropy": 5.388074064254761,
      "epoch": 4.477964561562926,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003492841943316903,
      "loss": 4.9799,
      "mean_token_accuracy": 0.19281488060951232,
      "num_tokens": 10208924.0,
      "step": 4930
    },
    {
      "entropy": 5.322542905807495,
      "epoch": 4.482507950931395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003489504912056103,
      "loss": 5.0524,
      "mean_token_accuracy": 0.18723482340574266,
      "num_tokens": 10219801.0,
      "step": 4935
    },
    {
      "entropy": 5.24891881942749,
      "epoch": 4.487051340299864,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003486166056140445,
      "loss": 4.8771,
      "mean_token_accuracy": 0.19545907378196717,
      "num_tokens": 10230233.0,
      "step": 4940
    },
    {
      "entropy": 5.3205338478088375,
      "epoch": 4.491594729668333,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034828253838082223,
      "loss": 4.9922,
      "mean_token_accuracy": 0.18871079683303832,
      "num_tokens": 10241448.0,
      "step": 4945
    },
    {
      "entropy": 5.327987813949585,
      "epoch": 4.496138119036801,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034794829033022134,
      "loss": 4.8628,
      "mean_token_accuracy": 0.1994113579392433,
      "num_tokens": 10250619.0,
      "step": 4950
    },
    {
      "entropy": 5.3413056373596195,
      "epoch": 4.50068150840527,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003476138622869657,
      "loss": 4.9825,
      "mean_token_accuracy": 0.192352232336998,
      "num_tokens": 10260246.0,
      "step": 4955
    },
    {
      "entropy": 5.222144222259521,
      "epoch": 4.505224897773739,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003472792550762232,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2000759169459343,
      "num_tokens": 10269626.0,
      "step": 4960
    },
    {
      "entropy": 5.299955987930298,
      "epoch": 4.509768287142208,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003469444695236038,
      "loss": 4.9117,
      "mean_token_accuracy": 0.19362198561429977,
      "num_tokens": 10279954.0,
      "step": 4965
    },
    {
      "entropy": 5.2166952133178714,
      "epoch": 4.514311676510677,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034660950645515765,
      "loss": 4.8808,
      "mean_token_accuracy": 0.1961284801363945,
      "num_tokens": 10290099.0,
      "step": 4970
    },
    {
      "entropy": 5.320992183685303,
      "epoch": 4.518855065879146,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003462743666973728,
      "loss": 4.8706,
      "mean_token_accuracy": 0.20428882241249086,
      "num_tokens": 10299469.0,
      "step": 4975
    },
    {
      "entropy": 5.308127975463867,
      "epoch": 4.523398455247615,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003459390510771734,
      "loss": 4.8529,
      "mean_token_accuracy": 0.2034368634223938,
      "num_tokens": 10310058.0,
      "step": 4980
    },
    {
      "entropy": 5.16323881149292,
      "epoch": 4.527941844616084,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003456035604219172,
      "loss": 4.8653,
      "mean_token_accuracy": 0.19996941536664964,
      "num_tokens": 10320960.0,
      "step": 4985
    },
    {
      "entropy": 5.259746170043945,
      "epoch": 4.532485233984552,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003452678955593942,
      "loss": 4.8993,
      "mean_token_accuracy": 0.19437046349048615,
      "num_tokens": 10331931.0,
      "step": 4990
    },
    {
      "entropy": 5.324531316757202,
      "epoch": 4.537028623353021,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000344932057317824,
      "loss": 4.9463,
      "mean_token_accuracy": 0.19267598837614058,
      "num_tokens": 10341964.0,
      "step": 4995
    },
    {
      "entropy": 5.327100038528442,
      "epoch": 4.54157201272149,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003445960465258542,
      "loss": 5.0289,
      "mean_token_accuracy": 0.18316574841737748,
      "num_tokens": 10352456.0,
      "step": 5000
    },
    {
      "epoch": 4.54157201272149,
      "eval_entropy": 5.208433178968208,
      "eval_loss": 5.746901035308838,
      "eval_mean_token_accuracy": 0.16671415553536525,
      "eval_num_tokens": 10352456.0,
      "eval_runtime": 2.6021,
      "eval_samples_per_second": 1320.064,
      "eval_steps_per_second": 165.248,
      "step": 5000
    },
    {
      "entropy": 5.2953649997711185,
      "epoch": 4.546115402089959,
      "grad_norm": 1.375,
      "learning_rate": 0.0003442598640125578,
      "loss": 4.916,
      "mean_token_accuracy": 0.1953711301088333,
      "num_tokens": 10363672.0,
      "step": 5005
    },
    {
      "entropy": 5.299848985671997,
      "epoch": 4.550658791458428,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000343923510607432,
      "loss": 5.005,
      "mean_token_accuracy": 0.19385025650262833,
      "num_tokens": 10374602.0,
      "step": 5010
    },
    {
      "entropy": 5.229483652114868,
      "epoch": 4.555202180826897,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003435869871403952,
      "loss": 4.8489,
      "mean_token_accuracy": 0.1997712090611458,
      "num_tokens": 10384374.0,
      "step": 5015
    },
    {
      "entropy": 5.330821800231933,
      "epoch": 4.559745570195366,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034325029444178565,
      "loss": 5.0139,
      "mean_token_accuracy": 0.18740748167037963,
      "num_tokens": 10395949.0,
      "step": 5020
    },
    {
      "entropy": 5.3294566631317135,
      "epoch": 4.564288959563835,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00034291343334235915,
      "loss": 4.937,
      "mean_token_accuracy": 0.19735765159130098,
      "num_tokens": 10407162.0,
      "step": 5025
    },
    {
      "entropy": 5.252464389801025,
      "epoch": 4.568832348932304,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003425764046732869,
      "loss": 4.8769,
      "mean_token_accuracy": 0.2007158026099205,
      "num_tokens": 10418191.0,
      "step": 5030
    },
    {
      "entropy": 5.315598392486573,
      "epoch": 4.573375738300772,
      "grad_norm": 1.578125,
      "learning_rate": 0.00034223920926615376,
      "loss": 4.933,
      "mean_token_accuracy": 0.19833574444055557,
      "num_tokens": 10427899.0,
      "step": 5035
    },
    {
      "entropy": 5.231576633453369,
      "epoch": 4.577919127669241,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034190184795295577,
      "loss": 4.8676,
      "mean_token_accuracy": 0.2007240816950798,
      "num_tokens": 10437882.0,
      "step": 5040
    },
    {
      "entropy": 5.220064640045166,
      "epoch": 4.58246251703771,
      "grad_norm": 1.5,
      "learning_rate": 0.00034156432156609844,
      "loss": 4.8952,
      "mean_token_accuracy": 0.19723475873470306,
      "num_tokens": 10448111.0,
      "step": 5045
    },
    {
      "entropy": 5.29509596824646,
      "epoch": 4.587005906406179,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00034122663093839466,
      "loss": 4.8851,
      "mean_token_accuracy": 0.197732675075531,
      "num_tokens": 10457522.0,
      "step": 5050
    },
    {
      "entropy": 5.267886924743652,
      "epoch": 4.591549295774648,
      "grad_norm": 1.609375,
      "learning_rate": 0.0003408887769030624,
      "loss": 4.7967,
      "mean_token_accuracy": 0.20774549543857573,
      "num_tokens": 10467229.0,
      "step": 5055
    },
    {
      "entropy": 5.272608518600464,
      "epoch": 4.596092685143117,
      "grad_norm": 1.390625,
      "learning_rate": 0.000340550760293723,
      "loss": 4.9073,
      "mean_token_accuracy": 0.19818672090768813,
      "num_tokens": 10477504.0,
      "step": 5060
    },
    {
      "entropy": 5.295639324188232,
      "epoch": 4.600636074511586,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034021258194439876,
      "loss": 4.8591,
      "mean_token_accuracy": 0.20400944650173186,
      "num_tokens": 10487238.0,
      "step": 5065
    },
    {
      "entropy": 5.247489023208618,
      "epoch": 4.605179463880054,
      "grad_norm": 1.515625,
      "learning_rate": 0.00033987424268951117,
      "loss": 4.941,
      "mean_token_accuracy": 0.18779279738664628,
      "num_tokens": 10497061.0,
      "step": 5070
    },
    {
      "entropy": 5.266718339920044,
      "epoch": 4.609722853248523,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033953574336387867,
      "loss": 4.9135,
      "mean_token_accuracy": 0.20587963610887527,
      "num_tokens": 10507763.0,
      "step": 5075
    },
    {
      "entropy": 5.259558391571045,
      "epoch": 4.614266242616992,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003391970848027148,
      "loss": 4.896,
      "mean_token_accuracy": 0.20419807583093644,
      "num_tokens": 10518466.0,
      "step": 5080
    },
    {
      "entropy": 5.269581747055054,
      "epoch": 4.618809631985461,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003388582678416257,
      "loss": 4.9841,
      "mean_token_accuracy": 0.1963512822985649,
      "num_tokens": 10528125.0,
      "step": 5085
    },
    {
      "entropy": 5.3088898181915285,
      "epoch": 4.62335302135393,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033851929331660876,
      "loss": 4.9046,
      "mean_token_accuracy": 0.19847527295351028,
      "num_tokens": 10537684.0,
      "step": 5090
    },
    {
      "entropy": 5.272795677185059,
      "epoch": 4.627896410722399,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033818016206404983,
      "loss": 4.986,
      "mean_token_accuracy": 0.19307581931352616,
      "num_tokens": 10548442.0,
      "step": 5095
    },
    {
      "entropy": 5.34151701927185,
      "epoch": 4.632439800090868,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003378408749207215,
      "loss": 4.9276,
      "mean_token_accuracy": 0.20033805519342424,
      "num_tokens": 10558885.0,
      "step": 5100
    },
    {
      "entropy": 5.2312287330627445,
      "epoch": 4.636983189459337,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003375014327237812,
      "loss": 4.8635,
      "mean_token_accuracy": 0.20271886736154557,
      "num_tokens": 10570481.0,
      "step": 5105
    },
    {
      "entropy": 5.297976732254028,
      "epoch": 4.641526578827806,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003371618363107686,
      "loss": 4.9411,
      "mean_token_accuracy": 0.19857709109783173,
      "num_tokens": 10580042.0,
      "step": 5110
    },
    {
      "entropy": 5.228000783920288,
      "epoch": 4.646069968196274,
      "grad_norm": 1.375,
      "learning_rate": 0.00033682208651960444,
      "loss": 4.921,
      "mean_token_accuracy": 0.19208317548036574,
      "num_tokens": 10590588.0,
      "step": 5115
    },
    {
      "entropy": 5.2662269115448,
      "epoch": 4.650613357564743,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033648218418858716,
      "loss": 4.9046,
      "mean_token_accuracy": 0.19068453311920167,
      "num_tokens": 10600952.0,
      "step": 5120
    },
    {
      "entropy": 5.315034627914429,
      "epoch": 4.655156746933212,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003361421301563922,
      "loss": 4.9317,
      "mean_token_accuracy": 0.20058779269456864,
      "num_tokens": 10610447.0,
      "step": 5125
    },
    {
      "entropy": 5.2689074039459225,
      "epoch": 4.659700136301681,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003358019252620691,
      "loss": 4.9746,
      "mean_token_accuracy": 0.1834099128842354,
      "num_tokens": 10621314.0,
      "step": 5130
    },
    {
      "entropy": 5.343396949768066,
      "epoch": 4.66424352567015,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033546157034503954,
      "loss": 4.983,
      "mean_token_accuracy": 0.19894181191921234,
      "num_tokens": 10631879.0,
      "step": 5135
    },
    {
      "entropy": 5.333950328826904,
      "epoch": 4.668786915038619,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003351210662450955,
      "loss": 4.8881,
      "mean_token_accuracy": 0.20211246609687805,
      "num_tokens": 10642233.0,
      "step": 5140
    },
    {
      "entropy": 5.3078293800354,
      "epoch": 4.673330304407088,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003347804138023968,
      "loss": 4.9596,
      "mean_token_accuracy": 0.19739264845848084,
      "num_tokens": 10652869.0,
      "step": 5145
    },
    {
      "entropy": 5.253484439849854,
      "epoch": 4.677873693775557,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003344396138574698,
      "loss": 4.9031,
      "mean_token_accuracy": 0.19011993557214737,
      "num_tokens": 10663368.0,
      "step": 5150
    },
    {
      "entropy": 5.380659484863282,
      "epoch": 4.682417083144026,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003340986672512042,
      "loss": 5.0285,
      "mean_token_accuracy": 0.1853848859667778,
      "num_tokens": 10674574.0,
      "step": 5155
    },
    {
      "entropy": 5.429159688949585,
      "epoch": 4.686960472512494,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0003337575748248521,
      "loss": 5.1086,
      "mean_token_accuracy": 0.18147067129611968,
      "num_tokens": 10685935.0,
      "step": 5160
    },
    {
      "entropy": 5.2758036136627195,
      "epoch": 4.691503861880963,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033341633742002503,
      "loss": 4.9717,
      "mean_token_accuracy": 0.18901186287403107,
      "num_tokens": 10696054.0,
      "step": 5165
    },
    {
      "entropy": 5.28107762336731,
      "epoch": 4.696047251249432,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003330749558786924,
      "loss": 4.9659,
      "mean_token_accuracy": 0.1900314673781395,
      "num_tokens": 10706334.0,
      "step": 5170
    },
    {
      "entropy": 5.3396186351776125,
      "epoch": 4.700590640617901,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003327334310431791,
      "loss": 4.9113,
      "mean_token_accuracy": 0.19568941593170167,
      "num_tokens": 10716953.0,
      "step": 5175
    },
    {
      "entropy": 5.307544279098511,
      "epoch": 4.70513402998637,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003323917637561638,
      "loss": 4.9092,
      "mean_token_accuracy": 0.19910464584827423,
      "num_tokens": 10726633.0,
      "step": 5180
    },
    {
      "entropy": 5.242602729797364,
      "epoch": 4.709677419354839,
      "grad_norm": 1.375,
      "learning_rate": 0.00033204995486067664,
      "loss": 4.8535,
      "mean_token_accuracy": 0.20304093807935714,
      "num_tokens": 10735862.0,
      "step": 5185
    },
    {
      "entropy": 5.259289932250977,
      "epoch": 4.714220808723308,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033170800520009703,
      "loss": 4.9337,
      "mean_token_accuracy": 0.19279786944389343,
      "num_tokens": 10746505.0,
      "step": 5190
    },
    {
      "entropy": 5.271475744247437,
      "epoch": 4.718764198091776,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003313659156181518,
      "loss": 4.8589,
      "mean_token_accuracy": 0.20393709242343902,
      "num_tokens": 10756513.0,
      "step": 5195
    },
    {
      "entropy": 5.251306867599487,
      "epoch": 4.723307587460245,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033102368695891283,
      "loss": 4.9125,
      "mean_token_accuracy": 0.19557353109121323,
      "num_tokens": 10766855.0,
      "step": 5200
    },
    {
      "entropy": 5.274575519561767,
      "epoch": 4.727850976828714,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033068132006679554,
      "loss": 4.8652,
      "mean_token_accuracy": 0.19894532710313798,
      "num_tokens": 10776213.0,
      "step": 5205
    },
    {
      "entropy": 5.324007892608643,
      "epoch": 4.732394366197183,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033033881578655605,
      "loss": 4.9955,
      "mean_token_accuracy": 0.1947008639574051,
      "num_tokens": 10786476.0,
      "step": 5210
    },
    {
      "entropy": 5.247025442123413,
      "epoch": 4.736937755565652,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003299961749632898,
      "loss": 4.8149,
      "mean_token_accuracy": 0.19929846674203872,
      "num_tokens": 10796274.0,
      "step": 5215
    },
    {
      "entropy": 5.226760673522949,
      "epoch": 4.741481144934121,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032965339844242864,
      "loss": 4.9672,
      "mean_token_accuracy": 0.19686986207962037,
      "num_tokens": 10806337.0,
      "step": 5220
    },
    {
      "entropy": 5.218794298171997,
      "epoch": 4.74602453430259,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003293104870697399,
      "loss": 4.8544,
      "mean_token_accuracy": 0.19678355753421783,
      "num_tokens": 10816403.0,
      "step": 5225
    },
    {
      "entropy": 5.3013077735900875,
      "epoch": 4.750567923671059,
      "grad_norm": 1.25,
      "learning_rate": 0.00032896744169132305,
      "loss": 5.0191,
      "mean_token_accuracy": 0.1857695996761322,
      "num_tokens": 10827830.0,
      "step": 5230
    },
    {
      "entropy": 5.335835933685303,
      "epoch": 4.755111313039528,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003286242631536087,
      "loss": 4.9518,
      "mean_token_accuracy": 0.19560089260339736,
      "num_tokens": 10839057.0,
      "step": 5235
    },
    {
      "entropy": 5.303196620941162,
      "epoch": 4.759654702407996,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032828095230335555,
      "loss": 5.0182,
      "mean_token_accuracy": 0.18970488458871843,
      "num_tokens": 10849814.0,
      "step": 5240
    },
    {
      "entropy": 5.289982604980469,
      "epoch": 4.764198091776465,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003279375099876491,
      "loss": 4.9331,
      "mean_token_accuracy": 0.19777359515428544,
      "num_tokens": 10860848.0,
      "step": 5245
    },
    {
      "entropy": 5.299950695037841,
      "epoch": 4.768741481144934,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003275939370538992,
      "loss": 4.9165,
      "mean_token_accuracy": 0.19502384066581727,
      "num_tokens": 10870349.0,
      "step": 5250
    },
    {
      "entropy": 5.30754132270813,
      "epoch": 4.773284870513403,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032725023434983767,
      "loss": 4.9523,
      "mean_token_accuracy": 0.19564595222473144,
      "num_tokens": 10880519.0,
      "step": 5255
    },
    {
      "entropy": 5.318440198898315,
      "epoch": 4.777828259881872,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032690640272351715,
      "loss": 4.9013,
      "mean_token_accuracy": 0.19555963277816774,
      "num_tokens": 10890254.0,
      "step": 5260
    },
    {
      "entropy": 5.243547105789185,
      "epoch": 4.782371649250341,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003265624430233076,
      "loss": 4.9599,
      "mean_token_accuracy": 0.18986516445875168,
      "num_tokens": 10901127.0,
      "step": 5265
    },
    {
      "entropy": 5.260863447189331,
      "epoch": 4.78691503861881,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003262183560978956,
      "loss": 4.8858,
      "mean_token_accuracy": 0.20389703661203384,
      "num_tokens": 10911572.0,
      "step": 5270
    },
    {
      "entropy": 5.23509464263916,
      "epoch": 4.791458427987278,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003258741427962814,
      "loss": 4.8511,
      "mean_token_accuracy": 0.20450012385845184,
      "num_tokens": 10922518.0,
      "step": 5275
    },
    {
      "entropy": 5.329229688644409,
      "epoch": 4.796001817355747,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032552980396777714,
      "loss": 4.9506,
      "mean_token_accuracy": 0.19240462481975557,
      "num_tokens": 10932818.0,
      "step": 5280
    },
    {
      "entropy": 5.334235525131225,
      "epoch": 4.800545206724216,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032518534046200464,
      "loss": 4.9769,
      "mean_token_accuracy": 0.19249373823404312,
      "num_tokens": 10943311.0,
      "step": 5285
    },
    {
      "entropy": 5.350389814376831,
      "epoch": 4.805088596092685,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003248407531288932,
      "loss": 4.9613,
      "mean_token_accuracy": 0.18849780261516572,
      "num_tokens": 10954157.0,
      "step": 5290
    },
    {
      "entropy": 5.304606866836548,
      "epoch": 4.809631985461154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000324496042818678,
      "loss": 4.9399,
      "mean_token_accuracy": 0.19612205475568772,
      "num_tokens": 10964745.0,
      "step": 5295
    },
    {
      "entropy": 5.280574560165405,
      "epoch": 4.814175374829623,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032415121038189746,
      "loss": 4.9197,
      "mean_token_accuracy": 0.20131670832633972,
      "num_tokens": 10975123.0,
      "step": 5300
    },
    {
      "entropy": 5.306539297103882,
      "epoch": 4.818718764198092,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032380625666939125,
      "loss": 4.9223,
      "mean_token_accuracy": 0.20078415274620057,
      "num_tokens": 10985189.0,
      "step": 5305
    },
    {
      "entropy": 5.230289077758789,
      "epoch": 4.823262153566561,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032346118253229834,
      "loss": 4.8703,
      "mean_token_accuracy": 0.19661126732826234,
      "num_tokens": 10994990.0,
      "step": 5310
    },
    {
      "entropy": 5.255189609527588,
      "epoch": 4.82780554293503,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032311598882205496,
      "loss": 4.8525,
      "mean_token_accuracy": 0.207609324157238,
      "num_tokens": 11004989.0,
      "step": 5315
    },
    {
      "entropy": 5.2993408203125,
      "epoch": 4.832348932303498,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032277067639039216,
      "loss": 4.9146,
      "mean_token_accuracy": 0.1950959876179695,
      "num_tokens": 11015225.0,
      "step": 5320
    },
    {
      "entropy": 5.320948076248169,
      "epoch": 4.836892321671967,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032242524608933423,
      "loss": 4.9743,
      "mean_token_accuracy": 0.1985565632581711,
      "num_tokens": 11026308.0,
      "step": 5325
    },
    {
      "entropy": 5.269458818435669,
      "epoch": 4.841435711040436,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032207969877119587,
      "loss": 4.9144,
      "mean_token_accuracy": 0.1904585540294647,
      "num_tokens": 11036313.0,
      "step": 5330
    },
    {
      "entropy": 5.2449976921081545,
      "epoch": 4.845979100408905,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032173403528858093,
      "loss": 4.9185,
      "mean_token_accuracy": 0.20063968896865844,
      "num_tokens": 11046444.0,
      "step": 5335
    },
    {
      "entropy": 5.3694836616516115,
      "epoch": 4.850522489777374,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003213882564943796,
      "loss": 4.9132,
      "mean_token_accuracy": 0.20081750750541688,
      "num_tokens": 11056514.0,
      "step": 5340
    },
    {
      "entropy": 5.289599418640137,
      "epoch": 4.855065879145843,
      "grad_norm": 1.375,
      "learning_rate": 0.0003210423632417666,
      "loss": 4.9089,
      "mean_token_accuracy": 0.19334154278039933,
      "num_tokens": 11066807.0,
      "step": 5345
    },
    {
      "entropy": 5.200515556335449,
      "epoch": 4.859609268514312,
      "grad_norm": 1.375,
      "learning_rate": 0.0003206963563841993,
      "loss": 4.9421,
      "mean_token_accuracy": 0.19071346670389175,
      "num_tokens": 11076141.0,
      "step": 5350
    },
    {
      "entropy": 5.248235511779785,
      "epoch": 4.86415265788278,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032035023677541515,
      "loss": 4.8115,
      "mean_token_accuracy": 0.20269840359687805,
      "num_tokens": 11085684.0,
      "step": 5355
    },
    {
      "entropy": 5.231860542297364,
      "epoch": 4.868696047251249,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003200040052694299,
      "loss": 4.8249,
      "mean_token_accuracy": 0.20005688369274138,
      "num_tokens": 11095401.0,
      "step": 5360
    },
    {
      "entropy": 5.305408477783203,
      "epoch": 4.873239436619718,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031965766272053534,
      "loss": 4.9943,
      "mean_token_accuracy": 0.19335354268550872,
      "num_tokens": 11105507.0,
      "step": 5365
    },
    {
      "entropy": 5.288741254806519,
      "epoch": 4.877782825988187,
      "grad_norm": 1.609375,
      "learning_rate": 0.00031931120998329737,
      "loss": 4.897,
      "mean_token_accuracy": 0.2048968568444252,
      "num_tokens": 11114809.0,
      "step": 5370
    },
    {
      "entropy": 5.277415323257446,
      "epoch": 4.882326215356656,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003189646479125536,
      "loss": 5.0265,
      "mean_token_accuracy": 0.19402653574943543,
      "num_tokens": 11124870.0,
      "step": 5375
    },
    {
      "entropy": 5.341051721572876,
      "epoch": 4.886869604725125,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003186179773634115,
      "loss": 4.9602,
      "mean_token_accuracy": 0.19005123674869537,
      "num_tokens": 11135925.0,
      "step": 5380
    },
    {
      "entropy": 5.315563106536866,
      "epoch": 4.891412994093594,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031827119919124614,
      "loss": 4.8998,
      "mean_token_accuracy": 0.19761408865451813,
      "num_tokens": 11146494.0,
      "step": 5385
    },
    {
      "entropy": 5.2539393424987795,
      "epoch": 4.895956383462063,
      "grad_norm": 1.65625,
      "learning_rate": 0.00031792431425169826,
      "loss": 4.8927,
      "mean_token_accuracy": 0.19896164834499358,
      "num_tokens": 11155817.0,
      "step": 5390
    },
    {
      "entropy": 5.30720739364624,
      "epoch": 4.900499772830532,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031757732340067195,
      "loss": 5.0344,
      "mean_token_accuracy": 0.18819668889045715,
      "num_tokens": 11167129.0,
      "step": 5395
    },
    {
      "entropy": 5.341184329986572,
      "epoch": 4.905043162199,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031723022749433267,
      "loss": 4.9844,
      "mean_token_accuracy": 0.19367574602365495,
      "num_tokens": 11177115.0,
      "step": 5400
    },
    {
      "entropy": 5.325886392593384,
      "epoch": 4.909586551567469,
      "grad_norm": 1.3125,
      "learning_rate": 0.000316883027389105,
      "loss": 4.9532,
      "mean_token_accuracy": 0.19046524167060852,
      "num_tokens": 11188551.0,
      "step": 5405
    },
    {
      "entropy": 5.351939868927002,
      "epoch": 4.914129940935938,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003165357239416706,
      "loss": 4.996,
      "mean_token_accuracy": 0.1874224990606308,
      "num_tokens": 11199560.0,
      "step": 5410
    },
    {
      "entropy": 5.283231401443482,
      "epoch": 4.918673330304407,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003161883180089665,
      "loss": 4.898,
      "mean_token_accuracy": 0.20079217702150345,
      "num_tokens": 11210265.0,
      "step": 5415
    },
    {
      "entropy": 5.312991714477539,
      "epoch": 4.923216719672876,
      "grad_norm": 1.25,
      "learning_rate": 0.0003158408104481821,
      "loss": 5.0294,
      "mean_token_accuracy": 0.19031682908535003,
      "num_tokens": 11221649.0,
      "step": 5420
    },
    {
      "entropy": 5.223658752441406,
      "epoch": 4.927760109041345,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031549320211675776,
      "loss": 4.8625,
      "mean_token_accuracy": 0.2017463505268097,
      "num_tokens": 11232417.0,
      "step": 5425
    },
    {
      "entropy": 5.238604545593262,
      "epoch": 4.932303498409814,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031514549387238246,
      "loss": 4.9754,
      "mean_token_accuracy": 0.1905214950442314,
      "num_tokens": 11243022.0,
      "step": 5430
    },
    {
      "entropy": 5.299124336242675,
      "epoch": 4.936846887778283,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0003147976865729918,
      "loss": 4.9248,
      "mean_token_accuracy": 0.19438218474388122,
      "num_tokens": 11253663.0,
      "step": 5435
    },
    {
      "entropy": 5.309145927429199,
      "epoch": 4.941390277146752,
      "grad_norm": 1.375,
      "learning_rate": 0.00031444978107676567,
      "loss": 4.9832,
      "mean_token_accuracy": 0.19447924941778183,
      "num_tokens": 11263482.0,
      "step": 5440
    },
    {
      "entropy": 5.283971166610717,
      "epoch": 4.94593366651522,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031410177824212635,
      "loss": 4.8425,
      "mean_token_accuracy": 0.20115117430686952,
      "num_tokens": 11273956.0,
      "step": 5445
    },
    {
      "entropy": 5.23683533668518,
      "epoch": 4.950477055883689,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003137536789277362,
      "loss": 4.9776,
      "mean_token_accuracy": 0.19542041122913362,
      "num_tokens": 11283590.0,
      "step": 5450
    },
    {
      "entropy": 5.26896390914917,
      "epoch": 4.955020445252158,
      "grad_norm": 1.640625,
      "learning_rate": 0.0003134054839924957,
      "loss": 4.912,
      "mean_token_accuracy": 0.18992525339126587,
      "num_tokens": 11294724.0,
      "step": 5455
    },
    {
      "entropy": 5.322569751739502,
      "epoch": 4.959563834620627,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031305719429554124,
      "loss": 4.8792,
      "mean_token_accuracy": 0.20103008002042771,
      "num_tokens": 11305995.0,
      "step": 5460
    },
    {
      "entropy": 5.2900594711303714,
      "epoch": 4.964107223989096,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003127088106962429,
      "loss": 4.8666,
      "mean_token_accuracy": 0.19979364573955535,
      "num_tokens": 11315373.0,
      "step": 5465
    },
    {
      "entropy": 5.209174156188965,
      "epoch": 4.968650613357565,
      "grad_norm": 1.375,
      "learning_rate": 0.00031236033405420284,
      "loss": 4.8917,
      "mean_token_accuracy": 0.196859709918499,
      "num_tokens": 11325662.0,
      "step": 5470
    },
    {
      "entropy": 5.363023805618286,
      "epoch": 4.973194002726034,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031201176522925236,
      "loss": 5.0692,
      "mean_token_accuracy": 0.1822429046034813,
      "num_tokens": 11335689.0,
      "step": 5475
    },
    {
      "entropy": 5.330723905563355,
      "epoch": 4.977737392094502,
      "grad_norm": 2.0625,
      "learning_rate": 0.0003116631050814503,
      "loss": 4.9072,
      "mean_token_accuracy": 0.20351831316947938,
      "num_tokens": 11345501.0,
      "step": 5480
    },
    {
      "entropy": 5.239264678955078,
      "epoch": 4.982280781462971,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031131435447108116,
      "loss": 4.8538,
      "mean_token_accuracy": 0.2034408926963806,
      "num_tokens": 11356146.0,
      "step": 5485
    },
    {
      "entropy": 5.260608577728272,
      "epoch": 4.98682417083144,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003109655142586521,
      "loss": 4.8929,
      "mean_token_accuracy": 0.20854515582323074,
      "num_tokens": 11366507.0,
      "step": 5490
    },
    {
      "entropy": 5.3064474105834964,
      "epoch": 4.991367560199909,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031061658530489195,
      "loss": 4.9763,
      "mean_token_accuracy": 0.1875535100698471,
      "num_tokens": 11376479.0,
      "step": 5495
    },
    {
      "entropy": 5.3290447235107425,
      "epoch": 4.995910949568378,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031026756847074804,
      "loss": 4.9822,
      "mean_token_accuracy": 0.1872828096151352,
      "num_tokens": 11387863.0,
      "step": 5500
    },
    {
      "epoch": 4.995910949568378,
      "eval_entropy": 5.255342186328977,
      "eval_loss": 5.705862045288086,
      "eval_mean_token_accuracy": 0.16985305537318074,
      "eval_num_tokens": 11387863.0,
      "eval_runtime": 2.5995,
      "eval_samples_per_second": 1321.391,
      "eval_steps_per_second": 165.414,
      "step": 5500
    },
    {
      "entropy": 5.29050514433119,
      "epoch": 5.0,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0003099184646173849,
      "loss": 4.8589,
      "mean_token_accuracy": 0.19876107573509216,
      "num_tokens": 11396985.0,
      "step": 5505
    },
    {
      "entropy": 5.210515260696411,
      "epoch": 5.004543389368469,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030956927460618126,
      "loss": 4.7166,
      "mean_token_accuracy": 0.20995302945375444,
      "num_tokens": 11406519.0,
      "step": 5510
    },
    {
      "entropy": 5.303342199325561,
      "epoch": 5.009086778736938,
      "grad_norm": 1.25,
      "learning_rate": 0.0003092199992987291,
      "loss": 4.8299,
      "mean_token_accuracy": 0.20484553128480912,
      "num_tokens": 11416746.0,
      "step": 5515
    },
    {
      "entropy": 5.299741649627686,
      "epoch": 5.013630168105407,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030887063955683044,
      "loss": 4.7004,
      "mean_token_accuracy": 0.20856853723526,
      "num_tokens": 11427218.0,
      "step": 5520
    },
    {
      "entropy": 5.226451969146728,
      "epoch": 5.018173557473876,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00030852119624249553,
      "loss": 4.6776,
      "mean_token_accuracy": 0.20547396391630174,
      "num_tokens": 11437623.0,
      "step": 5525
    },
    {
      "entropy": 5.263509082794189,
      "epoch": 5.022716946842344,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003081716702179412,
      "loss": 4.6169,
      "mean_token_accuracy": 0.21143995225429535,
      "num_tokens": 11446438.0,
      "step": 5530
    },
    {
      "entropy": 5.2789074897766115,
      "epoch": 5.027260336210813,
      "grad_norm": 1.390625,
      "learning_rate": 0.000307822062345588,
      "loss": 4.7517,
      "mean_token_accuracy": 0.20460224598646165,
      "num_tokens": 11457401.0,
      "step": 5535
    },
    {
      "entropy": 5.241945409774781,
      "epoch": 5.031803725579282,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003074723734880588,
      "loss": 4.7065,
      "mean_token_accuracy": 0.2068637192249298,
      "num_tokens": 11468295.0,
      "step": 5540
    },
    {
      "entropy": 5.29494080543518,
      "epoch": 5.036347114947751,
      "grad_norm": 1.296875,
      "learning_rate": 0.000307122604508176,
      "loss": 4.6641,
      "mean_token_accuracy": 0.2123077929019928,
      "num_tokens": 11478687.0,
      "step": 5545
    },
    {
      "entropy": 5.2427366256713865,
      "epoch": 5.04089050431622,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003067727562689598,
      "loss": 4.719,
      "mean_token_accuracy": 0.20857467502355576,
      "num_tokens": 11488121.0,
      "step": 5550
    },
    {
      "entropy": 5.205337047576904,
      "epoch": 5.045433893684689,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003064228296336259,
      "loss": 4.7415,
      "mean_token_accuracy": 0.20298799574375154,
      "num_tokens": 11498129.0,
      "step": 5555
    },
    {
      "entropy": 5.264484167098999,
      "epoch": 5.049977283053158,
      "grad_norm": 1.25,
      "learning_rate": 0.0003060728254655836,
      "loss": 4.7053,
      "mean_token_accuracy": 0.2107127785682678,
      "num_tokens": 11508870.0,
      "step": 5560
    },
    {
      "entropy": 5.30712742805481,
      "epoch": 5.054520672421627,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030572274462843327,
      "loss": 4.7763,
      "mean_token_accuracy": 0.20709846764802933,
      "num_tokens": 11518333.0,
      "step": 5565
    },
    {
      "entropy": 5.247055435180664,
      "epoch": 5.059064061790095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030537258798596465,
      "loss": 4.7109,
      "mean_token_accuracy": 0.20619940310716628,
      "num_tokens": 11529106.0,
      "step": 5570
    },
    {
      "entropy": 5.279263877868653,
      "epoch": 5.063607451158564,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003050223564021543,
      "loss": 4.8321,
      "mean_token_accuracy": 0.19837115556001664,
      "num_tokens": 11540248.0,
      "step": 5575
    },
    {
      "entropy": 5.212498760223388,
      "epoch": 5.068150840527033,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003046720507411641,
      "loss": 4.6638,
      "mean_token_accuracy": 0.210236494243145,
      "num_tokens": 11550291.0,
      "step": 5580
    },
    {
      "entropy": 5.272657871246338,
      "epoch": 5.072694229895502,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003043216718673383,
      "loss": 4.7383,
      "mean_token_accuracy": 0.20785504579544067,
      "num_tokens": 11560915.0,
      "step": 5585
    },
    {
      "entropy": 5.2767353534698485,
      "epoch": 5.077237619263971,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030397122064520186,
      "loss": 4.7355,
      "mean_token_accuracy": 0.20762938559055327,
      "num_tokens": 11571169.0,
      "step": 5590
    },
    {
      "entropy": 5.336246585845947,
      "epoch": 5.08178100863244,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030362069793945856,
      "loss": 4.8107,
      "mean_token_accuracy": 0.20400766432285308,
      "num_tokens": 11582784.0,
      "step": 5595
    },
    {
      "entropy": 5.342488098144531,
      "epoch": 5.086324398000909,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003032701046149881,
      "loss": 4.7849,
      "mean_token_accuracy": 0.20592684298753738,
      "num_tokens": 11593160.0,
      "step": 5600
    },
    {
      "entropy": 5.17545518875122,
      "epoch": 5.090867787369378,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003029194415368448,
      "loss": 4.6785,
      "mean_token_accuracy": 0.2105574905872345,
      "num_tokens": 11603277.0,
      "step": 5605
    },
    {
      "entropy": 5.247653770446777,
      "epoch": 5.095411176737846,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0003025687095702548,
      "loss": 4.744,
      "mean_token_accuracy": 0.21025571078062058,
      "num_tokens": 11612847.0,
      "step": 5610
    },
    {
      "entropy": 5.226384544372559,
      "epoch": 5.099954566106315,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003022179095806146,
      "loss": 4.7317,
      "mean_token_accuracy": 0.20709625333547593,
      "num_tokens": 11623090.0,
      "step": 5615
    },
    {
      "entropy": 5.263025093078613,
      "epoch": 5.104497955474784,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000301867042433488,
      "loss": 4.6553,
      "mean_token_accuracy": 0.22344305515289306,
      "num_tokens": 11634430.0,
      "step": 5620
    },
    {
      "entropy": 5.259416770935059,
      "epoch": 5.109041344843253,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030151610899460505,
      "loss": 4.7293,
      "mean_token_accuracy": 0.20855703204870224,
      "num_tokens": 11645425.0,
      "step": 5625
    },
    {
      "entropy": 5.230159330368042,
      "epoch": 5.113584734211722,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030116511012985916,
      "loss": 4.7656,
      "mean_token_accuracy": 0.20232328325510024,
      "num_tokens": 11655446.0,
      "step": 5630
    },
    {
      "entropy": 5.236868238449096,
      "epoch": 5.118128123580191,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003008140467053048,
      "loss": 4.7555,
      "mean_token_accuracy": 0.20313424617052078,
      "num_tokens": 11665383.0,
      "step": 5635
    },
    {
      "entropy": 5.234770107269287,
      "epoch": 5.12267151294866,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003004629195871565,
      "loss": 4.7024,
      "mean_token_accuracy": 0.20794471502304077,
      "num_tokens": 11676366.0,
      "step": 5640
    },
    {
      "entropy": 5.232264423370362,
      "epoch": 5.127214902317129,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003001117296417852,
      "loss": 4.7357,
      "mean_token_accuracy": 0.20590222626924515,
      "num_tokens": 11686537.0,
      "step": 5645
    },
    {
      "entropy": 5.290781402587891,
      "epoch": 5.131758291685597,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002997604777357173,
      "loss": 4.8566,
      "mean_token_accuracy": 0.20377578288316728,
      "num_tokens": 11698142.0,
      "step": 5650
    },
    {
      "entropy": 5.2289149284362795,
      "epoch": 5.136301681054066,
      "grad_norm": 1.328125,
      "learning_rate": 0.000299409164735632,
      "loss": 4.7303,
      "mean_token_accuracy": 0.2046062707901001,
      "num_tokens": 11709457.0,
      "step": 5655
    },
    {
      "entropy": 5.2671058177948,
      "epoch": 5.140845070422535,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002990577915083591,
      "loss": 4.683,
      "mean_token_accuracy": 0.21547812074422837,
      "num_tokens": 11719077.0,
      "step": 5660
    },
    {
      "entropy": 5.246507596969605,
      "epoch": 5.145388459791004,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029870635892087716,
      "loss": 4.7169,
      "mean_token_accuracy": 0.2061765491962433,
      "num_tokens": 11728823.0,
      "step": 5665
    },
    {
      "entropy": 5.3079712867736815,
      "epoch": 5.149931849159473,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002983548678403112,
      "loss": 4.7335,
      "mean_token_accuracy": 0.21144867986440657,
      "num_tokens": 11739574.0,
      "step": 5670
    },
    {
      "entropy": 5.2431378841400145,
      "epoch": 5.154475238527942,
      "grad_norm": 1.25,
      "learning_rate": 0.0002980033191339305,
      "loss": 4.7077,
      "mean_token_accuracy": 0.21345119476318358,
      "num_tokens": 11749499.0,
      "step": 5675
    },
    {
      "entropy": 5.151065731048584,
      "epoch": 5.159018627896411,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002976517136691465,
      "loss": 4.6136,
      "mean_token_accuracy": 0.21556092351675032,
      "num_tokens": 11759438.0,
      "step": 5680
    },
    {
      "entropy": 5.176685047149658,
      "epoch": 5.16356201726488,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029730005231351074,
      "loss": 4.6447,
      "mean_token_accuracy": 0.21828938722610475,
      "num_tokens": 11769432.0,
      "step": 5685
    },
    {
      "entropy": 5.272683191299438,
      "epoch": 5.168105406633348,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002969483359347127,
      "loss": 4.7167,
      "mean_token_accuracy": 0.209237203001976,
      "num_tokens": 11780886.0,
      "step": 5690
    },
    {
      "entropy": 5.267275667190551,
      "epoch": 5.172648796001817,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002965965654005776,
      "loss": 4.7934,
      "mean_token_accuracy": 0.21198458671569825,
      "num_tokens": 11791824.0,
      "step": 5695
    },
    {
      "entropy": 5.183382272720337,
      "epoch": 5.177192185370286,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002962447415790642,
      "loss": 4.715,
      "mean_token_accuracy": 0.20990239530801774,
      "num_tokens": 11801941.0,
      "step": 5700
    },
    {
      "entropy": 5.237336206436157,
      "epoch": 5.181735574738755,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029589286533826276,
      "loss": 4.7777,
      "mean_token_accuracy": 0.2048346519470215,
      "num_tokens": 11812258.0,
      "step": 5705
    },
    {
      "entropy": 5.3270354747772215,
      "epoch": 5.186278964107224,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002955409375463931,
      "loss": 4.8513,
      "mean_token_accuracy": 0.19952120631933212,
      "num_tokens": 11823046.0,
      "step": 5710
    },
    {
      "entropy": 5.333817338943481,
      "epoch": 5.190822353475693,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002951889590718019,
      "loss": 4.7937,
      "mean_token_accuracy": 0.2048426851630211,
      "num_tokens": 11834080.0,
      "step": 5715
    },
    {
      "entropy": 5.204223012924194,
      "epoch": 5.195365742844162,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002948369307829612,
      "loss": 4.6797,
      "mean_token_accuracy": 0.2157963141798973,
      "num_tokens": 11843848.0,
      "step": 5720
    },
    {
      "entropy": 5.1814733982086185,
      "epoch": 5.199909132212631,
      "grad_norm": 1.703125,
      "learning_rate": 0.0002944848535484656,
      "loss": 4.6458,
      "mean_token_accuracy": 0.21395381689071655,
      "num_tokens": 11854429.0,
      "step": 5725
    },
    {
      "entropy": 5.250406074523926,
      "epoch": 5.2044525215811,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029413272823703085,
      "loss": 4.7997,
      "mean_token_accuracy": 0.19708529710769654,
      "num_tokens": 11865035.0,
      "step": 5730
    },
    {
      "entropy": 5.347593688964844,
      "epoch": 5.208995910949568,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002937805557174911,
      "loss": 4.7159,
      "mean_token_accuracy": 0.2102460667490959,
      "num_tokens": 11875829.0,
      "step": 5735
    },
    {
      "entropy": 5.2683430194854735,
      "epoch": 5.213539300318037,
      "grad_norm": 1.375,
      "learning_rate": 0.00029342833685879713,
      "loss": 4.774,
      "mean_token_accuracy": 0.20188582092523574,
      "num_tokens": 11886397.0,
      "step": 5740
    },
    {
      "entropy": 5.275053930282593,
      "epoch": 5.218082689686506,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002930760725300137,
      "loss": 4.7776,
      "mean_token_accuracy": 0.20780951976776124,
      "num_tokens": 11896619.0,
      "step": 5745
    },
    {
      "entropy": 5.284980154037475,
      "epoch": 5.222626079054975,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00029272376360031833,
      "loss": 4.7948,
      "mean_token_accuracy": 0.2038257509469986,
      "num_tokens": 11906825.0,
      "step": 5750
    },
    {
      "entropy": 5.253081464767456,
      "epoch": 5.227169468423444,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029237141093899804,
      "loss": 4.7354,
      "mean_token_accuracy": 0.20714059174060823,
      "num_tokens": 11917051.0,
      "step": 5755
    },
    {
      "entropy": 5.197197294235229,
      "epoch": 5.231712857791913,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029201901541544824,
      "loss": 4.7326,
      "mean_token_accuracy": 0.2095335379242897,
      "num_tokens": 11927830.0,
      "step": 5760
    },
    {
      "entropy": 5.221789169311523,
      "epoch": 5.236256247160382,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002916665778991696,
      "loss": 4.7048,
      "mean_token_accuracy": 0.20916128903627396,
      "num_tokens": 11937541.0,
      "step": 5765
    },
    {
      "entropy": 5.249070262908935,
      "epoch": 5.240799636528851,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002913140992597668,
      "loss": 4.7529,
      "mean_token_accuracy": 0.20745014250278473,
      "num_tokens": 11947915.0,
      "step": 5770
    },
    {
      "entropy": 5.274767732620239,
      "epoch": 5.245343025897319,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002909615803669458,
      "loss": 4.6615,
      "mean_token_accuracy": 0.21216213703155518,
      "num_tokens": 11958340.0,
      "step": 5775
    },
    {
      "entropy": 5.210032558441162,
      "epoch": 5.249886415265788,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002906090220905119,
      "loss": 4.711,
      "mean_token_accuracy": 0.20219899266958236,
      "num_tokens": 11968661.0,
      "step": 5780
    },
    {
      "entropy": 5.203604078292846,
      "epoch": 5.254429804634257,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002902564253003677,
      "loss": 4.7147,
      "mean_token_accuracy": 0.210042867064476,
      "num_tokens": 11978282.0,
      "step": 5785
    },
    {
      "entropy": 5.197122287750244,
      "epoch": 5.258973194002726,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002899037908665106,
      "loss": 4.6722,
      "mean_token_accuracy": 0.21617419719696046,
      "num_tokens": 11987670.0,
      "step": 5790
    },
    {
      "entropy": 5.231730604171753,
      "epoch": 5.263516583371195,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002895511196590312,
      "loss": 4.784,
      "mean_token_accuracy": 0.20394459068775178,
      "num_tokens": 11999396.0,
      "step": 5795
    },
    {
      "entropy": 5.211545085906982,
      "epoch": 5.268059972739664,
      "grad_norm": 1.25,
      "learning_rate": 0.0002891984125481104,
      "loss": 4.7077,
      "mean_token_accuracy": 0.2115378797054291,
      "num_tokens": 12009894.0,
      "step": 5800
    },
    {
      "entropy": 5.245060062408447,
      "epoch": 5.272603362108133,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028884567040401814,
      "loss": 4.7006,
      "mean_token_accuracy": 0.21144134998321534,
      "num_tokens": 12020132.0,
      "step": 5805
    },
    {
      "entropy": 5.197646522521973,
      "epoch": 5.277146751476602,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002884928940971105,
      "loss": 4.7078,
      "mean_token_accuracy": 0.20651437491178512,
      "num_tokens": 12030844.0,
      "step": 5810
    },
    {
      "entropy": 5.181122207641602,
      "epoch": 5.28169014084507,
      "grad_norm": 1.40625,
      "learning_rate": 0.000288140084497828,
      "loss": 4.6928,
      "mean_token_accuracy": 0.21216001808643342,
      "num_tokens": 12040917.0,
      "step": 5815
    },
    {
      "entropy": 5.298718357086182,
      "epoch": 5.286233530213539,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002877872424766932,
      "loss": 4.7749,
      "mean_token_accuracy": 0.2075834631919861,
      "num_tokens": 12051443.0,
      "step": 5820
    },
    {
      "entropy": 5.264137125015258,
      "epoch": 5.290776919582008,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002874343689043086,
      "loss": 4.7017,
      "mean_token_accuracy": 0.21420291662216187,
      "num_tokens": 12061383.0,
      "step": 5825
    },
    {
      "entropy": 5.193834257125855,
      "epoch": 5.295320308950477,
      "grad_norm": 1.359375,
      "learning_rate": 0.000287081464651355,
      "loss": 4.6547,
      "mean_token_accuracy": 0.21517587900161744,
      "num_tokens": 12070251.0,
      "step": 5830
    },
    {
      "entropy": 5.170629596710205,
      "epoch": 5.299863698318946,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002867285305885882,
      "loss": 4.7101,
      "mean_token_accuracy": 0.21465217620134353,
      "num_tokens": 12080828.0,
      "step": 5835
    },
    {
      "entropy": 5.347774028778076,
      "epoch": 5.304407087687415,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028637556758683816,
      "loss": 4.7954,
      "mean_token_accuracy": 0.2047956496477127,
      "num_tokens": 12091695.0,
      "step": 5840
    },
    {
      "entropy": 5.193509435653686,
      "epoch": 5.308950477055884,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002860225765170058,
      "loss": 4.6753,
      "mean_token_accuracy": 0.219778910279274,
      "num_tokens": 12103058.0,
      "step": 5845
    },
    {
      "entropy": 5.226342964172363,
      "epoch": 5.313493866424353,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028566955825006174,
      "loss": 4.7889,
      "mean_token_accuracy": 0.20345086008310317,
      "num_tokens": 12113026.0,
      "step": 5850
    },
    {
      "entropy": 5.192390108108521,
      "epoch": 5.318037255792821,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00028531651365704325,
      "loss": 4.6831,
      "mean_token_accuracy": 0.21562511771917342,
      "num_tokens": 12123055.0,
      "step": 5855
    },
    {
      "entropy": 5.234465789794922,
      "epoch": 5.32258064516129,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028496344360905287,
      "loss": 4.7128,
      "mean_token_accuracy": 0.21003338247537612,
      "num_tokens": 12133538.0,
      "step": 5860
    },
    {
      "entropy": 5.187232446670532,
      "epoch": 5.327124034529759,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028461034897725587,
      "loss": 4.7067,
      "mean_token_accuracy": 0.21297314912080764,
      "num_tokens": 12143840.0,
      "step": 5865
    },
    {
      "entropy": 5.202144908905029,
      "epoch": 5.331667423898228,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028425723063287803,
      "loss": 4.8104,
      "mean_token_accuracy": 0.1951537549495697,
      "num_tokens": 12154822.0,
      "step": 5870
    },
    {
      "entropy": 5.226953601837158,
      "epoch": 5.336210813266697,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028390408944720385,
      "loss": 4.6986,
      "mean_token_accuracy": 0.21204953044652938,
      "num_tokens": 12163731.0,
      "step": 5875
    },
    {
      "entropy": 5.27759199142456,
      "epoch": 5.340754202635166,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002835509262915741,
      "loss": 4.7419,
      "mean_token_accuracy": 0.2077958270907402,
      "num_tokens": 12173574.0,
      "step": 5880
    },
    {
      "entropy": 5.226201295852661,
      "epoch": 5.345297592003635,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028319774203738375,
      "loss": 4.6992,
      "mean_token_accuracy": 0.21021638959646224,
      "num_tokens": 12183456.0,
      "step": 5885
    },
    {
      "entropy": 5.2069299697875975,
      "epoch": 5.349840981372104,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002828445375560798,
      "loss": 4.6768,
      "mean_token_accuracy": 0.2161361336708069,
      "num_tokens": 12192821.0,
      "step": 5890
    },
    {
      "entropy": 5.203872156143189,
      "epoch": 5.354384370740572,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028249131371915927,
      "loss": 4.6517,
      "mean_token_accuracy": 0.21896543800830842,
      "num_tokens": 12202904.0,
      "step": 5895
    },
    {
      "entropy": 5.25031886100769,
      "epoch": 5.358927760109041,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002821380713981668,
      "loss": 4.7387,
      "mean_token_accuracy": 0.20521695613861085,
      "num_tokens": 12213579.0,
      "step": 5900
    },
    {
      "entropy": 5.207906818389892,
      "epoch": 5.36347114947751,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028178481146469273,
      "loss": 4.7806,
      "mean_token_accuracy": 0.21011149734258652,
      "num_tokens": 12223668.0,
      "step": 5905
    },
    {
      "entropy": 5.276265096664429,
      "epoch": 5.368014538845979,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002814315347903709,
      "loss": 4.831,
      "mean_token_accuracy": 0.20721120685338973,
      "num_tokens": 12234869.0,
      "step": 5910
    },
    {
      "entropy": 5.2043966293334964,
      "epoch": 5.372557928214448,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002810782422468763,
      "loss": 4.6646,
      "mean_token_accuracy": 0.2071623384952545,
      "num_tokens": 12246154.0,
      "step": 5915
    },
    {
      "entropy": 5.237062454223633,
      "epoch": 5.377101317582917,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028072493470592317,
      "loss": 4.7463,
      "mean_token_accuracy": 0.21256787925958634,
      "num_tokens": 12256722.0,
      "step": 5920
    },
    {
      "entropy": 5.2306334495544435,
      "epoch": 5.381644706951386,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028037161303926265,
      "loss": 4.7006,
      "mean_token_accuracy": 0.21246959865093232,
      "num_tokens": 12266491.0,
      "step": 5925
    },
    {
      "entropy": 5.242271280288696,
      "epoch": 5.386188096319855,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000280018278118681,
      "loss": 4.7004,
      "mean_token_accuracy": 0.20876377671957017,
      "num_tokens": 12276946.0,
      "step": 5930
    },
    {
      "entropy": 5.2151103019714355,
      "epoch": 5.390731485688324,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000279664930815997,
      "loss": 4.7038,
      "mean_token_accuracy": 0.21210172325372695,
      "num_tokens": 12286411.0,
      "step": 5935
    },
    {
      "entropy": 5.2273202419281,
      "epoch": 5.395274875056792,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002793115720030599,
      "loss": 4.7799,
      "mean_token_accuracy": 0.20446934849023818,
      "num_tokens": 12296215.0,
      "step": 5940
    },
    {
      "entropy": 5.278167629241944,
      "epoch": 5.399818264425261,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002789582025517474,
      "loss": 4.7185,
      "mean_token_accuracy": 0.21359268724918365,
      "num_tokens": 12306246.0,
      "step": 5945
    },
    {
      "entropy": 5.214738893508911,
      "epoch": 5.40436165379373,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002786048233339637,
      "loss": 4.7347,
      "mean_token_accuracy": 0.20940390080213547,
      "num_tokens": 12316824.0,
      "step": 5950
    },
    {
      "entropy": 5.179840087890625,
      "epoch": 5.408905043162199,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027825143522163675,
      "loss": 4.7197,
      "mean_token_accuracy": 0.2143395334482193,
      "num_tokens": 12326729.0,
      "step": 5955
    },
    {
      "entropy": 5.186351156234741,
      "epoch": 5.413448432530668,
      "grad_norm": 1.5,
      "learning_rate": 0.0002778980390867165,
      "loss": 4.6249,
      "mean_token_accuracy": 0.2213203191757202,
      "num_tokens": 12335990.0,
      "step": 5960
    },
    {
      "entropy": 5.186501932144165,
      "epoch": 5.417991821899137,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027754463580117306,
      "loss": 4.682,
      "mean_token_accuracy": 0.21205735206604004,
      "num_tokens": 12345366.0,
      "step": 5965
    },
    {
      "entropy": 5.18764100074768,
      "epoch": 5.422535211267606,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027719122623699353,
      "loss": 4.7843,
      "mean_token_accuracy": 0.20766123980283738,
      "num_tokens": 12355668.0,
      "step": 5970
    },
    {
      "entropy": 5.236432456970215,
      "epoch": 5.427078600636074,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002768378112661813,
      "loss": 4.7681,
      "mean_token_accuracy": 0.20863338708877563,
      "num_tokens": 12365885.0,
      "step": 5975
    },
    {
      "entropy": 5.23615255355835,
      "epoch": 5.431621990004543,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002764843917607523,
      "loss": 4.7041,
      "mean_token_accuracy": 0.21391793638467788,
      "num_tokens": 12376795.0,
      "step": 5980
    },
    {
      "entropy": 5.2323531150817875,
      "epoch": 5.436165379373012,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002761309685927343,
      "loss": 4.7183,
      "mean_token_accuracy": 0.214588962495327,
      "num_tokens": 12386726.0,
      "step": 5985
    },
    {
      "entropy": 5.255731201171875,
      "epoch": 5.440708768741481,
      "grad_norm": 1.5,
      "learning_rate": 0.0002757775426341636,
      "loss": 4.7181,
      "mean_token_accuracy": 0.21335975974798202,
      "num_tokens": 12397264.0,
      "step": 5990
    },
    {
      "entropy": 5.261594820022583,
      "epoch": 5.44525215810995,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027542411475708385,
      "loss": 4.7655,
      "mean_token_accuracy": 0.21408816874027253,
      "num_tokens": 12407744.0,
      "step": 5995
    },
    {
      "entropy": 5.236018800735474,
      "epoch": 5.449795547478419,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027507068583354304,
      "loss": 4.7143,
      "mean_token_accuracy": 0.20737908482551576,
      "num_tokens": 12418009.0,
      "step": 6000
    },
    {
      "epoch": 5.449795547478419,
      "eval_entropy": 5.2022674116977425,
      "eval_loss": 5.710021495819092,
      "eval_mean_token_accuracy": 0.16998702256485473,
      "eval_num_tokens": 12418009.0,
      "eval_runtime": 2.7834,
      "eval_samples_per_second": 1234.112,
      "eval_steps_per_second": 154.489,
      "step": 6000
    },
    {
      "entropy": 5.307690238952636,
      "epoch": 5.454338936846888,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027471725673559205,
      "loss": 4.8114,
      "mean_token_accuracy": 0.2028793841600418,
      "num_tokens": 12427436.0,
      "step": 6005
    },
    {
      "entropy": 5.254571628570557,
      "epoch": 5.458882326215357,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027436382833528184,
      "loss": 4.7746,
      "mean_token_accuracy": 0.20295972228050232,
      "num_tokens": 12438110.0,
      "step": 6010
    },
    {
      "entropy": 5.228541040420533,
      "epoch": 5.463425715583826,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002740104015046621,
      "loss": 4.7701,
      "mean_token_accuracy": 0.21050762981176377,
      "num_tokens": 12448428.0,
      "step": 6015
    },
    {
      "entropy": 5.31198673248291,
      "epoch": 5.467969104952294,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027365697711577815,
      "loss": 4.7598,
      "mean_token_accuracy": 0.21337515860795975,
      "num_tokens": 12459339.0,
      "step": 6020
    },
    {
      "entropy": 5.231401586532593,
      "epoch": 5.472512494320763,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002733035560406697,
      "loss": 4.7619,
      "mean_token_accuracy": 0.2061229318380356,
      "num_tokens": 12469723.0,
      "step": 6025
    },
    {
      "entropy": 5.150815153121949,
      "epoch": 5.477055883689232,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027295013915136805,
      "loss": 4.6562,
      "mean_token_accuracy": 0.21540455818176268,
      "num_tokens": 12480191.0,
      "step": 6030
    },
    {
      "entropy": 5.143084859848022,
      "epoch": 5.481599273057701,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00027259672731989436,
      "loss": 4.6384,
      "mean_token_accuracy": 0.21573119908571242,
      "num_tokens": 12489881.0,
      "step": 6035
    },
    {
      "entropy": 5.157823944091797,
      "epoch": 5.48614266242617,
      "grad_norm": 1.453125,
      "learning_rate": 0.000272243321418257,
      "loss": 4.7159,
      "mean_token_accuracy": 0.21166936606168746,
      "num_tokens": 12499410.0,
      "step": 6040
    },
    {
      "entropy": 5.227223253250122,
      "epoch": 5.490686051794639,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002718899223184501,
      "loss": 4.7128,
      "mean_token_accuracy": 0.21740603893995286,
      "num_tokens": 12510079.0,
      "step": 6045
    },
    {
      "entropy": 5.205009698867798,
      "epoch": 5.495229441163108,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027153653089245067,
      "loss": 4.7178,
      "mean_token_accuracy": 0.21529059559106828,
      "num_tokens": 12520093.0,
      "step": 6050
    },
    {
      "entropy": 5.210605430603027,
      "epoch": 5.499772830531577,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000271183148012217,
      "loss": 4.7588,
      "mean_token_accuracy": 0.20947457998991012,
      "num_tokens": 12530767.0,
      "step": 6055
    },
    {
      "entropy": 5.214390230178833,
      "epoch": 5.504316219900045,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002708297745496861,
      "loss": 4.6957,
      "mean_token_accuracy": 0.20930741727352142,
      "num_tokens": 12541218.0,
      "step": 6060
    },
    {
      "entropy": 5.164271068572998,
      "epoch": 5.508859609268514,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027047641137677213,
      "loss": 4.6517,
      "mean_token_accuracy": 0.21656168699264527,
      "num_tokens": 12551112.0,
      "step": 6065
    },
    {
      "entropy": 5.1992597579956055,
      "epoch": 5.513402998636983,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002701230593653632,
      "loss": 4.7639,
      "mean_token_accuracy": 0.20667514353990554,
      "num_tokens": 12561225.0,
      "step": 6070
    },
    {
      "entropy": 5.2384007453918455,
      "epoch": 5.517946388005452,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002697697193873207,
      "loss": 4.6713,
      "mean_token_accuracy": 0.21966200768947602,
      "num_tokens": 12571508.0,
      "step": 6075
    },
    {
      "entropy": 5.215926551818848,
      "epoch": 5.522489777373921,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026941639231447547,
      "loss": 4.7682,
      "mean_token_accuracy": 0.20917714685201644,
      "num_tokens": 12582005.0,
      "step": 6080
    },
    {
      "entropy": 5.169674062728882,
      "epoch": 5.52703316674239,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002690630790186273,
      "loss": 4.7349,
      "mean_token_accuracy": 0.21867798864841462,
      "num_tokens": 12592243.0,
      "step": 6085
    },
    {
      "entropy": 5.198161220550537,
      "epoch": 5.531576556110859,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026870978037154137,
      "loss": 4.741,
      "mean_token_accuracy": 0.20532712042331697,
      "num_tokens": 12603141.0,
      "step": 6090
    },
    {
      "entropy": 5.277499151229859,
      "epoch": 5.536119945479328,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002683564972449472,
      "loss": 4.8244,
      "mean_token_accuracy": 0.20317636728286742,
      "num_tokens": 12614811.0,
      "step": 6095
    },
    {
      "entropy": 5.277568578720093,
      "epoch": 5.540663334847796,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002680032305105355,
      "loss": 4.8041,
      "mean_token_accuracy": 0.20516732931137086,
      "num_tokens": 12624957.0,
      "step": 6100
    },
    {
      "entropy": 5.252806806564331,
      "epoch": 5.545206724216265,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002676499810399572,
      "loss": 4.7837,
      "mean_token_accuracy": 0.19973315298557281,
      "num_tokens": 12635898.0,
      "step": 6105
    },
    {
      "entropy": 5.19001030921936,
      "epoch": 5.549750113584734,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026729674970481985,
      "loss": 4.7641,
      "mean_token_accuracy": 0.20911154597997667,
      "num_tokens": 12646176.0,
      "step": 6110
    },
    {
      "entropy": 5.214461040496826,
      "epoch": 5.554293502953203,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00026694353737668705,
      "loss": 4.698,
      "mean_token_accuracy": 0.21505919992923736,
      "num_tokens": 12656226.0,
      "step": 6115
    },
    {
      "entropy": 5.23705472946167,
      "epoch": 5.558836892321672,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026659034492707473,
      "loss": 4.7116,
      "mean_token_accuracy": 0.2098938688635826,
      "num_tokens": 12666768.0,
      "step": 6120
    },
    {
      "entropy": 5.242269563674927,
      "epoch": 5.563380281690141,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002662371732274504,
      "loss": 4.7359,
      "mean_token_accuracy": 0.20932943671941756,
      "num_tokens": 12676367.0,
      "step": 6125
    },
    {
      "entropy": 5.237624454498291,
      "epoch": 5.56792367105861,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026588402314923,
      "loss": 4.766,
      "mean_token_accuracy": 0.20764026641845704,
      "num_tokens": 12687906.0,
      "step": 6130
    },
    {
      "entropy": 5.197819662094116,
      "epoch": 5.572467060427079,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002655308955637763,
      "loss": 4.7504,
      "mean_token_accuracy": 0.21584542244672775,
      "num_tokens": 12698236.0,
      "step": 6135
    },
    {
      "entropy": 5.185625696182251,
      "epoch": 5.577010449795548,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002651777913423965,
      "loss": 4.6684,
      "mean_token_accuracy": 0.21818223148584365,
      "num_tokens": 12707778.0,
      "step": 6140
    },
    {
      "entropy": 5.204755783081055,
      "epoch": 5.581553839164016,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002648247113563402,
      "loss": 4.7144,
      "mean_token_accuracy": 0.2089651718735695,
      "num_tokens": 12717863.0,
      "step": 6145
    },
    {
      "entropy": 5.17846646308899,
      "epoch": 5.586097228532485,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026447165647679703,
      "loss": 4.7375,
      "mean_token_accuracy": 0.21253000497817992,
      "num_tokens": 12728061.0,
      "step": 6150
    },
    {
      "entropy": 5.23069658279419,
      "epoch": 5.590640617900954,
      "grad_norm": 1.328125,
      "learning_rate": 0.000264118627574895,
      "loss": 4.765,
      "mean_token_accuracy": 0.21085439771413803,
      "num_tokens": 12738069.0,
      "step": 6155
    },
    {
      "entropy": 5.190632915496826,
      "epoch": 5.595184007269423,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026376562552169776,
      "loss": 4.6994,
      "mean_token_accuracy": 0.21590688079595566,
      "num_tokens": 12749024.0,
      "step": 6160
    },
    {
      "entropy": 5.224492979049683,
      "epoch": 5.599727396637892,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026341265118820284,
      "loss": 4.7124,
      "mean_token_accuracy": 0.2098492205142975,
      "num_tokens": 12757861.0,
      "step": 6165
    },
    {
      "entropy": 5.233859491348267,
      "epoch": 5.604270786006361,
      "grad_norm": 1.375,
      "learning_rate": 0.00026305970544533903,
      "loss": 4.7756,
      "mean_token_accuracy": 0.20960801541805268,
      "num_tokens": 12768108.0,
      "step": 6170
    },
    {
      "entropy": 5.2454674243927,
      "epoch": 5.60881417537483,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026270678916396527,
      "loss": 4.8352,
      "mean_token_accuracy": 0.19826364070177077,
      "num_tokens": 12778471.0,
      "step": 6175
    },
    {
      "entropy": 5.301764106750488,
      "epoch": 5.613357564743298,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026235390321486705,
      "loss": 4.7856,
      "mean_token_accuracy": 0.20267389565706254,
      "num_tokens": 12789450.0,
      "step": 6180
    },
    {
      "entropy": 5.3542835235595705,
      "epoch": 5.617900954111767,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026200104846875574,
      "loss": 4.8308,
      "mean_token_accuracy": 0.19854771941900254,
      "num_tokens": 12800520.0,
      "step": 6185
    },
    {
      "entropy": 5.1957540035247805,
      "epoch": 5.622444343480236,
      "grad_norm": 1.390625,
      "learning_rate": 0.000261648225796265,
      "loss": 4.6684,
      "mean_token_accuracy": 0.21966566741466523,
      "num_tokens": 12810587.0,
      "step": 6190
    },
    {
      "entropy": 5.170246028900147,
      "epoch": 5.626987732848705,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002612954360679499,
      "loss": 4.68,
      "mean_token_accuracy": 0.2063571348786354,
      "num_tokens": 12820647.0,
      "step": 6195
    },
    {
      "entropy": 5.179163885116577,
      "epoch": 5.631531122217174,
      "grad_norm": 1.28125,
      "learning_rate": 0.000260942680154284,
      "loss": 4.7648,
      "mean_token_accuracy": 0.20375133454799652,
      "num_tokens": 12830863.0,
      "step": 6200
    },
    {
      "entropy": 5.282953405380249,
      "epoch": 5.636074511585643,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002605899589256574,
      "loss": 4.8036,
      "mean_token_accuracy": 0.20105793923139573,
      "num_tokens": 12842484.0,
      "step": 6205
    },
    {
      "entropy": 5.322309160232544,
      "epoch": 5.640617900954112,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026023727325237463,
      "loss": 4.7734,
      "mean_token_accuracy": 0.20259934961795806,
      "num_tokens": 12853009.0,
      "step": 6210
    },
    {
      "entropy": 5.193810272216797,
      "epoch": 5.645161290322581,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002598846240046527,
      "loss": 4.6696,
      "mean_token_accuracy": 0.22028847187757492,
      "num_tokens": 12862726.0,
      "step": 6215
    },
    {
      "entropy": 5.206301736831665,
      "epoch": 5.64970467969105,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002595320120526183,
      "loss": 4.7321,
      "mean_token_accuracy": 0.20320453941822053,
      "num_tokens": 12873389.0,
      "step": 6220
    },
    {
      "entropy": 5.138481187820434,
      "epoch": 5.654248069059518,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025917943826630666,
      "loss": 4.6714,
      "mean_token_accuracy": 0.21657271087169647,
      "num_tokens": 12883221.0,
      "step": 6225
    },
    {
      "entropy": 5.207355165481568,
      "epoch": 5.658791458427987,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002588269035156584,
      "loss": 4.7775,
      "mean_token_accuracy": 0.2075924351811409,
      "num_tokens": 12893761.0,
      "step": 6230
    },
    {
      "entropy": 5.220638465881348,
      "epoch": 5.663334847796456,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025847440867051796,
      "loss": 4.6864,
      "mean_token_accuracy": 0.21597739607095717,
      "num_tokens": 12903543.0,
      "step": 6235
    },
    {
      "entropy": 5.223698568344116,
      "epoch": 5.667878237164925,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002581219546006314,
      "loss": 4.7597,
      "mean_token_accuracy": 0.20733856409788132,
      "num_tokens": 12914330.0,
      "step": 6240
    },
    {
      "entropy": 5.259350824356079,
      "epoch": 5.672421626533394,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002577695421756441,
      "loss": 4.7909,
      "mean_token_accuracy": 0.1980230212211609,
      "num_tokens": 12925737.0,
      "step": 6245
    },
    {
      "entropy": 5.185084295272827,
      "epoch": 5.676965015901863,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002574171722650986,
      "loss": 4.6651,
      "mean_token_accuracy": 0.22046302407979965,
      "num_tokens": 12935808.0,
      "step": 6250
    },
    {
      "entropy": 5.166093540191651,
      "epoch": 5.681508405270332,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025706484573843273,
      "loss": 4.6698,
      "mean_token_accuracy": 0.22132380157709122,
      "num_tokens": 12946358.0,
      "step": 6255
    },
    {
      "entropy": 5.1884236335754395,
      "epoch": 5.6860517946388,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000256712563464977,
      "loss": 4.8036,
      "mean_token_accuracy": 0.2123783901333809,
      "num_tokens": 12956773.0,
      "step": 6260
    },
    {
      "entropy": 5.2091902732849125,
      "epoch": 5.690595184007269,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002563603263139532,
      "loss": 4.7397,
      "mean_token_accuracy": 0.20069270133972167,
      "num_tokens": 12967381.0,
      "step": 6265
    },
    {
      "entropy": 5.233079195022583,
      "epoch": 5.695138573375738,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002560081351544711,
      "loss": 4.7639,
      "mean_token_accuracy": 0.20792834013700484,
      "num_tokens": 12977844.0,
      "step": 6270
    },
    {
      "entropy": 5.25581202507019,
      "epoch": 5.699681962744207,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025565599085552776,
      "loss": 4.725,
      "mean_token_accuracy": 0.2103871539235115,
      "num_tokens": 12988180.0,
      "step": 6275
    },
    {
      "entropy": 5.199326038360596,
      "epoch": 5.704225352112676,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025530389428600394,
      "loss": 4.7534,
      "mean_token_accuracy": 0.20343973338603974,
      "num_tokens": 12998651.0,
      "step": 6280
    },
    {
      "entropy": 5.153060436248779,
      "epoch": 5.708768741481145,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002549518463146631,
      "loss": 4.6749,
      "mean_token_accuracy": 0.21319049149751662,
      "num_tokens": 13009494.0,
      "step": 6285
    },
    {
      "entropy": 5.265935707092285,
      "epoch": 5.713312130849614,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002545998478101485,
      "loss": 4.7132,
      "mean_token_accuracy": 0.20640659779310228,
      "num_tokens": 13019977.0,
      "step": 6290
    },
    {
      "entropy": 5.2016631126403805,
      "epoch": 5.717855520218083,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025424789964098154,
      "loss": 4.7427,
      "mean_token_accuracy": 0.21135631799697877,
      "num_tokens": 13030299.0,
      "step": 6295
    },
    {
      "entropy": 5.29795470237732,
      "epoch": 5.722398909586552,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002538960026755593,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2054819405078888,
      "num_tokens": 13040339.0,
      "step": 6300
    },
    {
      "entropy": 5.229258918762207,
      "epoch": 5.72694229895502,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025354415778215255,
      "loss": 4.6995,
      "mean_token_accuracy": 0.21058149188756942,
      "num_tokens": 13050187.0,
      "step": 6305
    },
    {
      "entropy": 5.205256986618042,
      "epoch": 5.731485688323489,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025319236582890364,
      "loss": 4.7886,
      "mean_token_accuracy": 0.20890921652317046,
      "num_tokens": 13061067.0,
      "step": 6310
    },
    {
      "entropy": 5.230925226211548,
      "epoch": 5.736029077691958,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025284062768382424,
      "loss": 4.7512,
      "mean_token_accuracy": 0.21103373616933824,
      "num_tokens": 13071525.0,
      "step": 6315
    },
    {
      "entropy": 5.279791069030762,
      "epoch": 5.740572467060427,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002524889442147932,
      "loss": 4.7496,
      "mean_token_accuracy": 0.2083263486623764,
      "num_tokens": 13081496.0,
      "step": 6320
    },
    {
      "entropy": 5.251438760757447,
      "epoch": 5.745115856428896,
      "grad_norm": 1.59375,
      "learning_rate": 0.00025213731628955454,
      "loss": 4.7945,
      "mean_token_accuracy": 0.20267647206783296,
      "num_tokens": 13091978.0,
      "step": 6325
    },
    {
      "entropy": 5.204975128173828,
      "epoch": 5.749659245797365,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025178574477571525,
      "loss": 4.7054,
      "mean_token_accuracy": 0.20962357223033906,
      "num_tokens": 13102540.0,
      "step": 6330
    },
    {
      "entropy": 5.322844791412353,
      "epoch": 5.754202635165834,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025143423054074305,
      "loss": 4.7653,
      "mean_token_accuracy": 0.2132736936211586,
      "num_tokens": 13112794.0,
      "step": 6335
    },
    {
      "entropy": 5.262812232971191,
      "epoch": 5.758746024534303,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002510827744519644,
      "loss": 4.8495,
      "mean_token_accuracy": 0.20389894098043443,
      "num_tokens": 13123141.0,
      "step": 6340
    },
    {
      "entropy": 5.240767765045166,
      "epoch": 5.763289413902772,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025073137737656226,
      "loss": 4.7058,
      "mean_token_accuracy": 0.2114260822534561,
      "num_tokens": 13133410.0,
      "step": 6345
    },
    {
      "entropy": 5.2652887344360355,
      "epoch": 5.76783280327124,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002503800401815739,
      "loss": 4.7581,
      "mean_token_accuracy": 0.20524149090051652,
      "num_tokens": 13143990.0,
      "step": 6350
    },
    {
      "entropy": 5.326006269454956,
      "epoch": 5.772376192639709,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025002876373388913,
      "loss": 4.7966,
      "mean_token_accuracy": 0.20357663184404373,
      "num_tokens": 13154664.0,
      "step": 6355
    },
    {
      "entropy": 5.229371786117554,
      "epoch": 5.776919582008178,
      "grad_norm": 2.046875,
      "learning_rate": 0.00024967754890024737,
      "loss": 4.7893,
      "mean_token_accuracy": 0.2025609791278839,
      "num_tokens": 13165292.0,
      "step": 6360
    },
    {
      "entropy": 5.228417778015137,
      "epoch": 5.781462971376647,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002493263965472366,
      "loss": 4.8265,
      "mean_token_accuracy": 0.20418507903814315,
      "num_tokens": 13177111.0,
      "step": 6365
    },
    {
      "entropy": 5.275480175018311,
      "epoch": 5.786006360745116,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002489753075412901,
      "loss": 4.7815,
      "mean_token_accuracy": 0.20181158035993577,
      "num_tokens": 13186988.0,
      "step": 6370
    },
    {
      "entropy": 5.234550046920776,
      "epoch": 5.790549750113585,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024862428274868533,
      "loss": 4.7712,
      "mean_token_accuracy": 0.21438572704792022,
      "num_tokens": 13197338.0,
      "step": 6375
    },
    {
      "entropy": 5.217247056961059,
      "epoch": 5.795093139482054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002482733230355408,
      "loss": 4.675,
      "mean_token_accuracy": 0.21664434522390366,
      "num_tokens": 13208202.0,
      "step": 6380
    },
    {
      "entropy": 5.2220031261444095,
      "epoch": 5.799636528850522,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024792242926781495,
      "loss": 4.7147,
      "mean_token_accuracy": 0.2156476378440857,
      "num_tokens": 13218895.0,
      "step": 6385
    },
    {
      "entropy": 5.191397857666016,
      "epoch": 5.804179918218991,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024757160231130313,
      "loss": 4.724,
      "mean_token_accuracy": 0.2113228127360344,
      "num_tokens": 13229060.0,
      "step": 6390
    },
    {
      "entropy": 5.196179962158203,
      "epoch": 5.80872330758746,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024722084303163607,
      "loss": 4.7458,
      "mean_token_accuracy": 0.2061508044600487,
      "num_tokens": 13239635.0,
      "step": 6395
    },
    {
      "entropy": 5.20751051902771,
      "epoch": 5.813266696955929,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002468701522942774,
      "loss": 4.7796,
      "mean_token_accuracy": 0.1990472838282585,
      "num_tokens": 13250783.0,
      "step": 6400
    },
    {
      "entropy": 5.201352643966675,
      "epoch": 5.817810086324398,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024651953096452165,
      "loss": 4.6707,
      "mean_token_accuracy": 0.21681001484394075,
      "num_tokens": 13260718.0,
      "step": 6405
    },
    {
      "entropy": 5.206371021270752,
      "epoch": 5.822353475692867,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00024616897990749196,
      "loss": 4.6928,
      "mean_token_accuracy": 0.21215641796588897,
      "num_tokens": 13271606.0,
      "step": 6410
    },
    {
      "entropy": 5.196653079986572,
      "epoch": 5.826896865061336,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002458184999881385,
      "loss": 4.7047,
      "mean_token_accuracy": 0.2114178091287613,
      "num_tokens": 13281327.0,
      "step": 6415
    },
    {
      "entropy": 5.259857940673828,
      "epoch": 5.831440254429805,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024546809207123534,
      "loss": 4.8001,
      "mean_token_accuracy": 0.20363649278879165,
      "num_tokens": 13292871.0,
      "step": 6420
    },
    {
      "entropy": 5.179827070236206,
      "epoch": 5.835983643798274,
      "grad_norm": 1.25,
      "learning_rate": 0.0002451177570213794,
      "loss": 4.709,
      "mean_token_accuracy": 0.21026542782783508,
      "num_tokens": 13303282.0,
      "step": 6425
    },
    {
      "entropy": 5.197968244552612,
      "epoch": 5.840527033166742,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024476749570298746,
      "loss": 4.7159,
      "mean_token_accuracy": 0.2122677430510521,
      "num_tokens": 13312987.0,
      "step": 6430
    },
    {
      "entropy": 5.21340765953064,
      "epoch": 5.845070422535211,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024441730898029456,
      "loss": 4.7787,
      "mean_token_accuracy": 0.2060747876763344,
      "num_tokens": 13324164.0,
      "step": 6435
    },
    {
      "entropy": 5.2442995548248295,
      "epoch": 5.84961381190368,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024406719771735165,
      "loss": 4.7778,
      "mean_token_accuracy": 0.20355363488197326,
      "num_tokens": 13336790.0,
      "step": 6440
    },
    {
      "entropy": 5.273229455947876,
      "epoch": 5.854157201272149,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002437171627780234,
      "loss": 4.7624,
      "mean_token_accuracy": 0.2053859457373619,
      "num_tokens": 13347107.0,
      "step": 6445
    },
    {
      "entropy": 5.18575644493103,
      "epoch": 5.858700590640618,
      "grad_norm": 1.375,
      "learning_rate": 0.00024336720502598624,
      "loss": 4.6618,
      "mean_token_accuracy": 0.22433497607707978,
      "num_tokens": 13356211.0,
      "step": 6450
    },
    {
      "entropy": 5.220148181915283,
      "epoch": 5.863243980009087,
      "grad_norm": 1.375,
      "learning_rate": 0.0002430173253247262,
      "loss": 4.6686,
      "mean_token_accuracy": 0.22676336616277695,
      "num_tokens": 13366357.0,
      "step": 6455
    },
    {
      "entropy": 5.255270910263062,
      "epoch": 5.867787369377556,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024266752453753655,
      "loss": 4.7795,
      "mean_token_accuracy": 0.20243070721626283,
      "num_tokens": 13375670.0,
      "step": 6460
    },
    {
      "entropy": 5.226347398757935,
      "epoch": 5.872330758746024,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024231780352751615,
      "loss": 4.8062,
      "mean_token_accuracy": 0.21104159504175185,
      "num_tokens": 13387054.0,
      "step": 6465
    },
    {
      "entropy": 5.199756908416748,
      "epoch": 5.876874148114493,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024196816315756665,
      "loss": 4.6455,
      "mean_token_accuracy": 0.2188382253050804,
      "num_tokens": 13396314.0,
      "step": 6470
    },
    {
      "entropy": 5.227719974517822,
      "epoch": 5.881417537482962,
      "grad_norm": 1.25,
      "learning_rate": 0.0002416186042903909,
      "loss": 4.7453,
      "mean_token_accuracy": 0.21217637807130812,
      "num_tokens": 13405595.0,
      "step": 6475
    },
    {
      "entropy": 5.28304615020752,
      "epoch": 5.885960926851431,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024126912778849074,
      "loss": 4.8789,
      "mean_token_accuracy": 0.19967182129621505,
      "num_tokens": 13417008.0,
      "step": 6480
    },
    {
      "entropy": 5.251292896270752,
      "epoch": 5.8905043162199,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024091973451416472,
      "loss": 4.7743,
      "mean_token_accuracy": 0.21336886435747146,
      "num_tokens": 13427650.0,
      "step": 6485
    },
    {
      "entropy": 5.226341438293457,
      "epoch": 5.895047705588369,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024057042532950597,
      "loss": 4.7724,
      "mean_token_accuracy": 0.21212844997644426,
      "num_tokens": 13437821.0,
      "step": 6490
    },
    {
      "entropy": 5.18856406211853,
      "epoch": 5.899591094956838,
      "grad_norm": 1.375,
      "learning_rate": 0.0002402212010964001,
      "loss": 4.732,
      "mean_token_accuracy": 0.20686429291963576,
      "num_tokens": 13447685.0,
      "step": 6495
    },
    {
      "entropy": 5.222682237625122,
      "epoch": 5.904134484325307,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023987206267652334,
      "loss": 4.7373,
      "mean_token_accuracy": 0.2066054090857506,
      "num_tokens": 13457929.0,
      "step": 6500
    },
    {
      "epoch": 5.904134484325307,
      "eval_entropy": 5.160699034846107,
      "eval_loss": 5.6763505935668945,
      "eval_mean_token_accuracy": 0.17259121559733567,
      "eval_num_tokens": 13457929.0,
      "eval_runtime": 2.5862,
      "eval_samples_per_second": 1328.208,
      "eval_steps_per_second": 166.268,
      "step": 6500
    },
    {
      "entropy": 5.219326639175415,
      "epoch": 5.908677873693776,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023952301093133993,
      "loss": 4.6946,
      "mean_token_accuracy": 0.21792764067649842,
      "num_tokens": 13467220.0,
      "step": 6505
    },
    {
      "entropy": 5.262956666946411,
      "epoch": 5.913221263062244,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023917404672210046,
      "loss": 4.772,
      "mean_token_accuracy": 0.20719584822654724,
      "num_tokens": 13478048.0,
      "step": 6510
    },
    {
      "entropy": 5.217322731018067,
      "epoch": 5.917764652430713,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002388251709098392,
      "loss": 4.7384,
      "mean_token_accuracy": 0.21209635436534882,
      "num_tokens": 13488212.0,
      "step": 6515
    },
    {
      "entropy": 5.182860088348389,
      "epoch": 5.922308041799182,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023847638435537283,
      "loss": 4.6955,
      "mean_token_accuracy": 0.21994332671165467,
      "num_tokens": 13498748.0,
      "step": 6520
    },
    {
      "entropy": 5.123570203781128,
      "epoch": 5.926851431167651,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023812768791929724,
      "loss": 4.6764,
      "mean_token_accuracy": 0.212064428627491,
      "num_tokens": 13509296.0,
      "step": 6525
    },
    {
      "entropy": 5.115686273574829,
      "epoch": 5.93139482053612,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023777908246198643,
      "loss": 4.7141,
      "mean_token_accuracy": 0.2105269730091095,
      "num_tokens": 13519804.0,
      "step": 6530
    },
    {
      "entropy": 5.267679738998413,
      "epoch": 5.935938209904589,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023743056884358954,
      "loss": 4.8118,
      "mean_token_accuracy": 0.20512096136808394,
      "num_tokens": 13531437.0,
      "step": 6535
    },
    {
      "entropy": 5.2762034893035885,
      "epoch": 5.940481599273058,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023708214792402942,
      "loss": 4.7946,
      "mean_token_accuracy": 0.2043566733598709,
      "num_tokens": 13541925.0,
      "step": 6540
    },
    {
      "entropy": 5.169518709182739,
      "epoch": 5.945024988641527,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023673382056299997,
      "loss": 4.7077,
      "mean_token_accuracy": 0.21209318488836287,
      "num_tokens": 13552843.0,
      "step": 6545
    },
    {
      "entropy": 5.182296705245972,
      "epoch": 5.949568378009996,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023638558761996436,
      "loss": 4.7182,
      "mean_token_accuracy": 0.2131866216659546,
      "num_tokens": 13563325.0,
      "step": 6550
    },
    {
      "entropy": 5.190829563140869,
      "epoch": 5.954111767378464,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023603744995415277,
      "loss": 4.7349,
      "mean_token_accuracy": 0.21483782082796096,
      "num_tokens": 13572919.0,
      "step": 6555
    },
    {
      "entropy": 5.206899738311767,
      "epoch": 5.958655156746933,
      "grad_norm": 1.25,
      "learning_rate": 0.00023568940842456026,
      "loss": 4.7505,
      "mean_token_accuracy": 0.20809393376111984,
      "num_tokens": 13582865.0,
      "step": 6560
    },
    {
      "entropy": 5.196602249145508,
      "epoch": 5.963198546115402,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002353414638899447,
      "loss": 4.6918,
      "mean_token_accuracy": 0.21488803923130034,
      "num_tokens": 13592865.0,
      "step": 6565
    },
    {
      "entropy": 5.176244735717773,
      "epoch": 5.967741935483871,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023499361720882468,
      "loss": 4.7281,
      "mean_token_accuracy": 0.20966595858335496,
      "num_tokens": 13603218.0,
      "step": 6570
    },
    {
      "entropy": 5.19638729095459,
      "epoch": 5.97228532485234,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023464586923947718,
      "loss": 4.7312,
      "mean_token_accuracy": 0.21076542884111404,
      "num_tokens": 13613854.0,
      "step": 6575
    },
    {
      "entropy": 5.181171321868897,
      "epoch": 5.976828714220809,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023429822083993596,
      "loss": 4.6673,
      "mean_token_accuracy": 0.21846974790096282,
      "num_tokens": 13624164.0,
      "step": 6580
    },
    {
      "entropy": 5.203751850128174,
      "epoch": 5.981372103589278,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023395067286798872,
      "loss": 4.7451,
      "mean_token_accuracy": 0.2098066806793213,
      "num_tokens": 13634538.0,
      "step": 6585
    },
    {
      "entropy": 5.222130489349365,
      "epoch": 5.985915492957746,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023360322618117563,
      "loss": 4.8004,
      "mean_token_accuracy": 0.19935833662748337,
      "num_tokens": 13645014.0,
      "step": 6590
    },
    {
      "entropy": 5.1950311183929445,
      "epoch": 5.990458882326215,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023325588163678674,
      "loss": 4.6811,
      "mean_token_accuracy": 0.21879498213529586,
      "num_tokens": 13655152.0,
      "step": 6595
    },
    {
      "entropy": 5.201615619659424,
      "epoch": 5.995002271694684,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023290864009186036,
      "loss": 4.717,
      "mean_token_accuracy": 0.20798347443342208,
      "num_tokens": 13665549.0,
      "step": 6600
    },
    {
      "entropy": 5.251978874206543,
      "epoch": 5.999545661063153,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023256150240318025,
      "loss": 4.7223,
      "mean_token_accuracy": 0.2171073392033577,
      "num_tokens": 13675580.0,
      "step": 6605
    },
    {
      "entropy": 5.212876849704319,
      "epoch": 6.003634711494775,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023221446942727438,
      "loss": 4.5549,
      "mean_token_accuracy": 0.22601422336366442,
      "num_tokens": 13683425.0,
      "step": 6610
    },
    {
      "entropy": 5.192490339279175,
      "epoch": 6.008178100863244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002318675420204119,
      "loss": 4.5489,
      "mean_token_accuracy": 0.22889133393764496,
      "num_tokens": 13693008.0,
      "step": 6615
    },
    {
      "entropy": 5.202480030059815,
      "epoch": 6.012721490231713,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023152072103860188,
      "loss": 4.5368,
      "mean_token_accuracy": 0.22495754659175873,
      "num_tokens": 13703396.0,
      "step": 6620
    },
    {
      "entropy": 5.161708879470825,
      "epoch": 6.017264879600182,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023117400733759053,
      "loss": 4.5675,
      "mean_token_accuracy": 0.2253779575228691,
      "num_tokens": 13713171.0,
      "step": 6625
    },
    {
      "entropy": 5.2064728260040285,
      "epoch": 6.021808268968651,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002308274017728595,
      "loss": 4.6205,
      "mean_token_accuracy": 0.21572703421115874,
      "num_tokens": 13723543.0,
      "step": 6630
    },
    {
      "entropy": 5.192204236984253,
      "epoch": 6.026351658337119,
      "grad_norm": 1.375,
      "learning_rate": 0.0002304809051996235,
      "loss": 4.4918,
      "mean_token_accuracy": 0.22541241645812987,
      "num_tokens": 13733693.0,
      "step": 6635
    },
    {
      "entropy": 5.1649693012237545,
      "epoch": 6.030895047705588,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002301345184728284,
      "loss": 4.5836,
      "mean_token_accuracy": 0.21210667341947556,
      "num_tokens": 13744614.0,
      "step": 6640
    },
    {
      "entropy": 5.170063018798828,
      "epoch": 6.035438437074057,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022978824244714903,
      "loss": 4.5182,
      "mean_token_accuracy": 0.2307887479662895,
      "num_tokens": 13755116.0,
      "step": 6645
    },
    {
      "entropy": 5.235706806182861,
      "epoch": 6.039981826442526,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022944207797698707,
      "loss": 4.5957,
      "mean_token_accuracy": 0.21507111340761184,
      "num_tokens": 13765492.0,
      "step": 6650
    },
    {
      "entropy": 5.149616289138794,
      "epoch": 6.044525215810995,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002290960259164689,
      "loss": 4.5614,
      "mean_token_accuracy": 0.226220765709877,
      "num_tokens": 13775661.0,
      "step": 6655
    },
    {
      "entropy": 5.11228346824646,
      "epoch": 6.049068605179464,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022875008711944355,
      "loss": 4.4821,
      "mean_token_accuracy": 0.2332501381635666,
      "num_tokens": 13785473.0,
      "step": 6660
    },
    {
      "entropy": 5.229855966567993,
      "epoch": 6.053611994547933,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022840426243948077,
      "loss": 4.5961,
      "mean_token_accuracy": 0.22616648226976394,
      "num_tokens": 13796011.0,
      "step": 6665
    },
    {
      "entropy": 5.217935371398926,
      "epoch": 6.058155383916402,
      "grad_norm": 1.671875,
      "learning_rate": 0.00022805855272986842,
      "loss": 4.5091,
      "mean_token_accuracy": 0.22796317487955092,
      "num_tokens": 13805904.0,
      "step": 6670
    },
    {
      "entropy": 5.188564777374268,
      "epoch": 6.062698773284871,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022771295884361088,
      "loss": 4.5225,
      "mean_token_accuracy": 0.23339703530073166,
      "num_tokens": 13816491.0,
      "step": 6675
    },
    {
      "entropy": 5.168851613998413,
      "epoch": 6.067242162653339,
      "grad_norm": 1.796875,
      "learning_rate": 0.0002273674816334268,
      "loss": 4.6494,
      "mean_token_accuracy": 0.2135434627532959,
      "num_tokens": 13826483.0,
      "step": 6680
    },
    {
      "entropy": 5.111876058578491,
      "epoch": 6.071785552021808,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022702212195174664,
      "loss": 4.4466,
      "mean_token_accuracy": 0.23258927166461946,
      "num_tokens": 13837307.0,
      "step": 6685
    },
    {
      "entropy": 5.217587089538574,
      "epoch": 6.076328941390277,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022667688065071146,
      "loss": 4.6079,
      "mean_token_accuracy": 0.2235248863697052,
      "num_tokens": 13847454.0,
      "step": 6690
    },
    {
      "entropy": 5.160775566101075,
      "epoch": 6.080872330758746,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022633175858216936,
      "loss": 4.4545,
      "mean_token_accuracy": 0.2278663009405136,
      "num_tokens": 13857680.0,
      "step": 6695
    },
    {
      "entropy": 5.193804168701172,
      "epoch": 6.085415720127215,
      "grad_norm": 1.5,
      "learning_rate": 0.00022598675659767515,
      "loss": 4.6034,
      "mean_token_accuracy": 0.21862859427928924,
      "num_tokens": 13868724.0,
      "step": 6700
    },
    {
      "entropy": 5.253815650939941,
      "epoch": 6.089959109495684,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022564187554848663,
      "loss": 4.6598,
      "mean_token_accuracy": 0.21204373985528946,
      "num_tokens": 13879679.0,
      "step": 6705
    },
    {
      "entropy": 5.211440753936768,
      "epoch": 6.094502498864153,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022529711628556387,
      "loss": 4.5679,
      "mean_token_accuracy": 0.22105786055326462,
      "num_tokens": 13889099.0,
      "step": 6710
    },
    {
      "entropy": 5.209904909133911,
      "epoch": 6.099045888232622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022495247965956567,
      "loss": 4.6377,
      "mean_token_accuracy": 0.21251746714115144,
      "num_tokens": 13899912.0,
      "step": 6715
    },
    {
      "entropy": 5.214570999145508,
      "epoch": 6.10358927760109,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002246079665208489,
      "loss": 4.562,
      "mean_token_accuracy": 0.22048781663179398,
      "num_tokens": 13908923.0,
      "step": 6720
    },
    {
      "entropy": 5.199292469024658,
      "epoch": 6.108132666969559,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022426357771946544,
      "loss": 4.5607,
      "mean_token_accuracy": 0.22620125561952592,
      "num_tokens": 13918718.0,
      "step": 6725
    },
    {
      "entropy": 5.191080808639526,
      "epoch": 6.112676056338028,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022391931410516036,
      "loss": 4.5401,
      "mean_token_accuracy": 0.23284049481153488,
      "num_tokens": 13928926.0,
      "step": 6730
    },
    {
      "entropy": 5.110253429412841,
      "epoch": 6.117219445706497,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022357517652736992,
      "loss": 4.5219,
      "mean_token_accuracy": 0.2285371705889702,
      "num_tokens": 13938916.0,
      "step": 6735
    },
    {
      "entropy": 5.1582049369812015,
      "epoch": 6.121762835074966,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022323116583521948,
      "loss": 4.5864,
      "mean_token_accuracy": 0.2191261127591133,
      "num_tokens": 13949271.0,
      "step": 6740
    },
    {
      "entropy": 5.2070784091949465,
      "epoch": 6.126306224443435,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022288728287752109,
      "loss": 4.572,
      "mean_token_accuracy": 0.2254108801484108,
      "num_tokens": 13959266.0,
      "step": 6745
    },
    {
      "entropy": 5.204643106460571,
      "epoch": 6.130849613811904,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002225435285027719,
      "loss": 4.5319,
      "mean_token_accuracy": 0.21901249885559082,
      "num_tokens": 13969011.0,
      "step": 6750
    },
    {
      "entropy": 5.180943202972412,
      "epoch": 6.135393003180373,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022219990355915148,
      "loss": 4.6346,
      "mean_token_accuracy": 0.21731972098350524,
      "num_tokens": 13980137.0,
      "step": 6755
    },
    {
      "entropy": 5.123016834259033,
      "epoch": 6.139936392548841,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022185640889452036,
      "loss": 4.5308,
      "mean_token_accuracy": 0.22864564657211303,
      "num_tokens": 13990008.0,
      "step": 6760
    },
    {
      "entropy": 5.20939154624939,
      "epoch": 6.14447978191731,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022151304535641747,
      "loss": 4.5694,
      "mean_token_accuracy": 0.21451642811298371,
      "num_tokens": 14001524.0,
      "step": 6765
    },
    {
      "entropy": 5.202929878234864,
      "epoch": 6.149023171285779,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002211698137920582,
      "loss": 4.592,
      "mean_token_accuracy": 0.2177274525165558,
      "num_tokens": 14012286.0,
      "step": 6770
    },
    {
      "entropy": 5.195061111450196,
      "epoch": 6.153566560654248,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022082671504833223,
      "loss": 4.5371,
      "mean_token_accuracy": 0.23354398608207702,
      "num_tokens": 14023164.0,
      "step": 6775
    },
    {
      "entropy": 5.1999462127685545,
      "epoch": 6.158109950022717,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022048374997180175,
      "loss": 4.6128,
      "mean_token_accuracy": 0.2129863142967224,
      "num_tokens": 14033360.0,
      "step": 6780
    },
    {
      "entropy": 5.187428522109985,
      "epoch": 6.162653339391186,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022014091940869885,
      "loss": 4.61,
      "mean_token_accuracy": 0.22031975090503692,
      "num_tokens": 14044318.0,
      "step": 6785
    },
    {
      "entropy": 5.183869075775147,
      "epoch": 6.167196728759655,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021979822420492402,
      "loss": 4.5206,
      "mean_token_accuracy": 0.2311058446764946,
      "num_tokens": 14053600.0,
      "step": 6790
    },
    {
      "entropy": 5.203868293762207,
      "epoch": 6.171740118128124,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021945566520604337,
      "loss": 4.5723,
      "mean_token_accuracy": 0.22663809955120087,
      "num_tokens": 14063897.0,
      "step": 6795
    },
    {
      "entropy": 5.172455263137818,
      "epoch": 6.176283507496592,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002191132432572875,
      "loss": 4.5792,
      "mean_token_accuracy": 0.21421003341674805,
      "num_tokens": 14075037.0,
      "step": 6800
    },
    {
      "entropy": 5.307517576217651,
      "epoch": 6.180826896865061,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021877095920354818,
      "loss": 4.7408,
      "mean_token_accuracy": 0.2116205185651779,
      "num_tokens": 14085512.0,
      "step": 6805
    },
    {
      "entropy": 5.23793511390686,
      "epoch": 6.18537028623353,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021842881388937764,
      "loss": 4.5738,
      "mean_token_accuracy": 0.22223070710897447,
      "num_tokens": 14096337.0,
      "step": 6810
    },
    {
      "entropy": 5.190524578094482,
      "epoch": 6.189913675601999,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021808680815898508,
      "loss": 4.5931,
      "mean_token_accuracy": 0.22074858844280243,
      "num_tokens": 14106422.0,
      "step": 6815
    },
    {
      "entropy": 5.133832263946533,
      "epoch": 6.194457064970468,
      "grad_norm": 1.375,
      "learning_rate": 0.00021774494285623598,
      "loss": 4.5404,
      "mean_token_accuracy": 0.22152386456727982,
      "num_tokens": 14115227.0,
      "step": 6820
    },
    {
      "entropy": 5.20292387008667,
      "epoch": 6.199000454338937,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002174032188246487,
      "loss": 4.6158,
      "mean_token_accuracy": 0.2216602385044098,
      "num_tokens": 14125095.0,
      "step": 6825
    },
    {
      "entropy": 5.253089904785156,
      "epoch": 6.203543843707406,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021706163690739344,
      "loss": 4.6314,
      "mean_token_accuracy": 0.22048637121915818,
      "num_tokens": 14136708.0,
      "step": 6830
    },
    {
      "entropy": 5.176157236099243,
      "epoch": 6.208087233075875,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021672019794728963,
      "loss": 4.5225,
      "mean_token_accuracy": 0.22786477357149124,
      "num_tokens": 14146259.0,
      "step": 6835
    },
    {
      "entropy": 5.170056676864624,
      "epoch": 6.212630622444343,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021637890278680406,
      "loss": 4.5625,
      "mean_token_accuracy": 0.2291735291481018,
      "num_tokens": 14156848.0,
      "step": 6840
    },
    {
      "entropy": 5.20801682472229,
      "epoch": 6.217174011812812,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021603775226804845,
      "loss": 4.6229,
      "mean_token_accuracy": 0.2162877470254898,
      "num_tokens": 14168898.0,
      "step": 6845
    },
    {
      "entropy": 5.214771127700805,
      "epoch": 6.221717401181281,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021569674723277783,
      "loss": 4.5589,
      "mean_token_accuracy": 0.22559811174869537,
      "num_tokens": 14179322.0,
      "step": 6850
    },
    {
      "entropy": 5.189191055297852,
      "epoch": 6.22626079054975,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021535588852238823,
      "loss": 4.547,
      "mean_token_accuracy": 0.22731879949569703,
      "num_tokens": 14189884.0,
      "step": 6855
    },
    {
      "entropy": 5.198185110092163,
      "epoch": 6.230804179918219,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021501517697791479,
      "loss": 4.5296,
      "mean_token_accuracy": 0.2174621492624283,
      "num_tokens": 14200166.0,
      "step": 6860
    },
    {
      "entropy": 5.1773172378540036,
      "epoch": 6.235347569286688,
      "grad_norm": 1.375,
      "learning_rate": 0.0002146746134400291,
      "loss": 4.6311,
      "mean_token_accuracy": 0.21665179282426833,
      "num_tokens": 14210056.0,
      "step": 6865
    },
    {
      "entropy": 5.182350778579712,
      "epoch": 6.239890958655157,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021433419874903803,
      "loss": 4.5756,
      "mean_token_accuracy": 0.2310296878218651,
      "num_tokens": 14220384.0,
      "step": 6870
    },
    {
      "entropy": 5.23394284248352,
      "epoch": 6.244434348023626,
      "grad_norm": 1.25,
      "learning_rate": 0.00021399393374488097,
      "loss": 4.5906,
      "mean_token_accuracy": 0.22090001851320268,
      "num_tokens": 14231712.0,
      "step": 6875
    },
    {
      "entropy": 5.156463670730591,
      "epoch": 6.248977737392094,
      "grad_norm": 1.125,
      "learning_rate": 0.00021365381926712797,
      "loss": 4.5805,
      "mean_token_accuracy": 0.22230938374996184,
      "num_tokens": 14242581.0,
      "step": 6880
    },
    {
      "entropy": 5.201946592330932,
      "epoch": 6.253521126760563,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021331385615497756,
      "loss": 4.6529,
      "mean_token_accuracy": 0.22092657089233397,
      "num_tokens": 14253690.0,
      "step": 6885
    },
    {
      "entropy": 5.234377336502075,
      "epoch": 6.258064516129032,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002129740452472552,
      "loss": 4.6283,
      "mean_token_accuracy": 0.2196796253323555,
      "num_tokens": 14264256.0,
      "step": 6890
    },
    {
      "entropy": 5.235438203811645,
      "epoch": 6.262607905497501,
      "grad_norm": 1.375,
      "learning_rate": 0.00021263438738241026,
      "loss": 4.6082,
      "mean_token_accuracy": 0.23004007786512376,
      "num_tokens": 14273784.0,
      "step": 6895
    },
    {
      "entropy": 5.185305547714234,
      "epoch": 6.26715129486597,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021229488339851498,
      "loss": 4.5869,
      "mean_token_accuracy": 0.2250821128487587,
      "num_tokens": 14284769.0,
      "step": 6900
    },
    {
      "entropy": 5.151624011993408,
      "epoch": 6.271694684234439,
      "grad_norm": 1.640625,
      "learning_rate": 0.00021195553413326146,
      "loss": 4.5589,
      "mean_token_accuracy": 0.22108794152736663,
      "num_tokens": 14295451.0,
      "step": 6905
    },
    {
      "entropy": 5.169211578369141,
      "epoch": 6.276238073602908,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002116163404239605,
      "loss": 4.5585,
      "mean_token_accuracy": 0.22184546738862992,
      "num_tokens": 14305088.0,
      "step": 6910
    },
    {
      "entropy": 5.169395399093628,
      "epoch": 6.280781462971377,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021127730310753866,
      "loss": 4.5655,
      "mean_token_accuracy": 0.22305043637752534,
      "num_tokens": 14315834.0,
      "step": 6915
    },
    {
      "entropy": 5.209618997573853,
      "epoch": 6.285324852339846,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021093842302053706,
      "loss": 4.6219,
      "mean_token_accuracy": 0.2196890741586685,
      "num_tokens": 14326040.0,
      "step": 6920
    },
    {
      "entropy": 5.192116785049438,
      "epoch": 6.289868241708314,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021059970099910835,
      "loss": 4.6107,
      "mean_token_accuracy": 0.2158440500497818,
      "num_tokens": 14337088.0,
      "step": 6925
    },
    {
      "entropy": 5.188809108734131,
      "epoch": 6.294411631076783,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021026113787901568,
      "loss": 4.5755,
      "mean_token_accuracy": 0.23072212785482407,
      "num_tokens": 14348107.0,
      "step": 6930
    },
    {
      "entropy": 5.1460202693939205,
      "epoch": 6.298955020445252,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020992273449562966,
      "loss": 4.5743,
      "mean_token_accuracy": 0.2195840746164322,
      "num_tokens": 14358457.0,
      "step": 6935
    },
    {
      "entropy": 5.214573764801026,
      "epoch": 6.303498409813721,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020958449168392723,
      "loss": 4.6024,
      "mean_token_accuracy": 0.2181102529168129,
      "num_tokens": 14368639.0,
      "step": 6940
    },
    {
      "entropy": 5.223721647262574,
      "epoch": 6.30804179918219,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002092464102784888,
      "loss": 4.6296,
      "mean_token_accuracy": 0.23415992259979249,
      "num_tokens": 14379611.0,
      "step": 6945
    },
    {
      "entropy": 5.250877571105957,
      "epoch": 6.312585188550659,
      "grad_norm": 1.625,
      "learning_rate": 0.00020890849111349663,
      "loss": 4.6262,
      "mean_token_accuracy": 0.22078102231025695,
      "num_tokens": 14389927.0,
      "step": 6950
    },
    {
      "entropy": 5.1555828094482425,
      "epoch": 6.317128577919128,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020857073502273266,
      "loss": 4.5123,
      "mean_token_accuracy": 0.2285353347659111,
      "num_tokens": 14400097.0,
      "step": 6955
    },
    {
      "entropy": 5.24519305229187,
      "epoch": 6.321671967287597,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002082331428395765,
      "loss": 4.6175,
      "mean_token_accuracy": 0.22524985522031785,
      "num_tokens": 14411392.0,
      "step": 6960
    },
    {
      "entropy": 5.203271198272705,
      "epoch": 6.326215356656065,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020789571539700326,
      "loss": 4.6435,
      "mean_token_accuracy": 0.21559442579746246,
      "num_tokens": 14422961.0,
      "step": 6965
    },
    {
      "entropy": 5.201879835128784,
      "epoch": 6.330758746024534,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020755845352758175,
      "loss": 4.5779,
      "mean_token_accuracy": 0.22470675110816957,
      "num_tokens": 14432319.0,
      "step": 6970
    },
    {
      "entropy": 5.208508825302124,
      "epoch": 6.335302135393003,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002072213580634718,
      "loss": 4.5647,
      "mean_token_accuracy": 0.2216217115521431,
      "num_tokens": 14442756.0,
      "step": 6975
    },
    {
      "entropy": 5.158717346191406,
      "epoch": 6.339845524761472,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002068844298364233,
      "loss": 4.6253,
      "mean_token_accuracy": 0.21594420969486236,
      "num_tokens": 14453960.0,
      "step": 6980
    },
    {
      "entropy": 5.133931732177734,
      "epoch": 6.344388914129941,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020654766967777288,
      "loss": 4.5317,
      "mean_token_accuracy": 0.22412099391222,
      "num_tokens": 14464472.0,
      "step": 6985
    },
    {
      "entropy": 5.201199245452881,
      "epoch": 6.34893230349841,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002062110784184431,
      "loss": 4.5906,
      "mean_token_accuracy": 0.21805181354284286,
      "num_tokens": 14474289.0,
      "step": 6990
    },
    {
      "entropy": 5.168363332748413,
      "epoch": 6.353475692866879,
      "grad_norm": 1.375,
      "learning_rate": 0.00020587465688893904,
      "loss": 4.5038,
      "mean_token_accuracy": 0.22496067136526107,
      "num_tokens": 14484121.0,
      "step": 6995
    },
    {
      "entropy": 5.168064451217651,
      "epoch": 6.358019082235348,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020553840591934765,
      "loss": 4.5687,
      "mean_token_accuracy": 0.22602293491363526,
      "num_tokens": 14493780.0,
      "step": 7000
    },
    {
      "epoch": 6.358019082235348,
      "eval_entropy": 5.091943261789721,
      "eval_loss": 5.674976348876953,
      "eval_mean_token_accuracy": 0.17420489148691642,
      "eval_num_tokens": 14493780.0,
      "eval_runtime": 2.5884,
      "eval_samples_per_second": 1327.055,
      "eval_steps_per_second": 166.123,
      "step": 7000
    },
    {
      "entropy": 5.0987560749053955,
      "epoch": 6.362562471603816,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002052023263393346,
      "loss": 4.5479,
      "mean_token_accuracy": 0.22534846663475036,
      "num_tokens": 14504150.0,
      "step": 7005
    },
    {
      "entropy": 5.205683898925781,
      "epoch": 6.367105860972285,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020486641897814295,
      "loss": 4.5689,
      "mean_token_accuracy": 0.2224227473139763,
      "num_tokens": 14514516.0,
      "step": 7010
    },
    {
      "entropy": 5.227892541885376,
      "epoch": 6.371649250340754,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020453068466459067,
      "loss": 4.6438,
      "mean_token_accuracy": 0.2159521162509918,
      "num_tokens": 14525418.0,
      "step": 7015
    },
    {
      "entropy": 5.196643686294555,
      "epoch": 6.376192639709223,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020419512422706893,
      "loss": 4.5598,
      "mean_token_accuracy": 0.22821896970272065,
      "num_tokens": 14535330.0,
      "step": 7020
    },
    {
      "entropy": 5.2433568954467775,
      "epoch": 6.380736029077692,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020385973849353957,
      "loss": 4.6387,
      "mean_token_accuracy": 0.22332401424646378,
      "num_tokens": 14546513.0,
      "step": 7025
    },
    {
      "entropy": 5.128661298751831,
      "epoch": 6.385279418446161,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020352452829153369,
      "loss": 4.5616,
      "mean_token_accuracy": 0.2276337817311287,
      "num_tokens": 14556393.0,
      "step": 7030
    },
    {
      "entropy": 5.121362400054932,
      "epoch": 6.38982280781463,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000203189494448149,
      "loss": 4.6083,
      "mean_token_accuracy": 0.2177527517080307,
      "num_tokens": 14566458.0,
      "step": 7035
    },
    {
      "entropy": 5.190993213653565,
      "epoch": 6.394366197183099,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020285463779004837,
      "loss": 4.5276,
      "mean_token_accuracy": 0.21768778264522554,
      "num_tokens": 14577298.0,
      "step": 7040
    },
    {
      "entropy": 5.153841400146485,
      "epoch": 6.398909586551568,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020251995914345722,
      "loss": 4.5589,
      "mean_token_accuracy": 0.21602552831172944,
      "num_tokens": 14587015.0,
      "step": 7045
    },
    {
      "entropy": 5.118944454193115,
      "epoch": 6.403452975920036,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020218545933416181,
      "loss": 4.5938,
      "mean_token_accuracy": 0.217490154504776,
      "num_tokens": 14597568.0,
      "step": 7050
    },
    {
      "entropy": 5.1107264995574955,
      "epoch": 6.407996365288505,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020185113918750724,
      "loss": 4.5032,
      "mean_token_accuracy": 0.22895032465457915,
      "num_tokens": 14607393.0,
      "step": 7055
    },
    {
      "entropy": 5.2118395328521725,
      "epoch": 6.412539754656974,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002015169995283952,
      "loss": 4.588,
      "mean_token_accuracy": 0.22019348591566085,
      "num_tokens": 14617291.0,
      "step": 7060
    },
    {
      "entropy": 5.169604825973511,
      "epoch": 6.417083144025443,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020118304118128211,
      "loss": 4.6037,
      "mean_token_accuracy": 0.22006657719612122,
      "num_tokens": 14628124.0,
      "step": 7065
    },
    {
      "entropy": 5.207346057891845,
      "epoch": 6.421626533393912,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020084926497017703,
      "loss": 4.6714,
      "mean_token_accuracy": 0.21138775646686553,
      "num_tokens": 14638625.0,
      "step": 7070
    },
    {
      "entropy": 5.172139501571655,
      "epoch": 6.426169922762381,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020051567171863944,
      "loss": 4.5809,
      "mean_token_accuracy": 0.21871691793203354,
      "num_tokens": 14648407.0,
      "step": 7075
    },
    {
      "entropy": 5.257504272460937,
      "epoch": 6.43071331213085,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002001822622497777,
      "loss": 4.6155,
      "mean_token_accuracy": 0.21402187049388885,
      "num_tokens": 14658617.0,
      "step": 7080
    },
    {
      "entropy": 5.211090087890625,
      "epoch": 6.435256701499318,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019984903738624632,
      "loss": 4.6049,
      "mean_token_accuracy": 0.2191697508096695,
      "num_tokens": 14669542.0,
      "step": 7085
    },
    {
      "entropy": 5.159584045410156,
      "epoch": 6.439800090867787,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019951599795024478,
      "loss": 4.573,
      "mean_token_accuracy": 0.22474441230297088,
      "num_tokens": 14680265.0,
      "step": 7090
    },
    {
      "entropy": 5.222048568725586,
      "epoch": 6.444343480236256,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019918314476351462,
      "loss": 4.6116,
      "mean_token_accuracy": 0.22043169289827347,
      "num_tokens": 14690103.0,
      "step": 7095
    },
    {
      "entropy": 5.200143098831177,
      "epoch": 6.448886869604725,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019885047864733803,
      "loss": 4.6459,
      "mean_token_accuracy": 0.2163871020078659,
      "num_tokens": 14699626.0,
      "step": 7100
    },
    {
      "entropy": 5.146953868865967,
      "epoch": 6.453430258973194,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019851800042253553,
      "loss": 4.5489,
      "mean_token_accuracy": 0.22394299060106276,
      "num_tokens": 14710173.0,
      "step": 7105
    },
    {
      "entropy": 5.165923595428467,
      "epoch": 6.457973648341663,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001981857109094642,
      "loss": 4.5856,
      "mean_token_accuracy": 0.21562654674053192,
      "num_tokens": 14720730.0,
      "step": 7110
    },
    {
      "entropy": 5.144653081893921,
      "epoch": 6.462517037710132,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019785361092801524,
      "loss": 4.5792,
      "mean_token_accuracy": 0.2233627602458,
      "num_tokens": 14731473.0,
      "step": 7115
    },
    {
      "entropy": 5.124156856536866,
      "epoch": 6.467060427078601,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019752170129761244,
      "loss": 4.5515,
      "mean_token_accuracy": 0.23074091374874114,
      "num_tokens": 14742093.0,
      "step": 7120
    },
    {
      "entropy": 5.125227737426758,
      "epoch": 6.47160381644707,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001971899828372098,
      "loss": 4.6376,
      "mean_token_accuracy": 0.21322681456804277,
      "num_tokens": 14753636.0,
      "step": 7125
    },
    {
      "entropy": 5.194995546340943,
      "epoch": 6.476147205815538,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019685845636528972,
      "loss": 4.5559,
      "mean_token_accuracy": 0.2284410297870636,
      "num_tokens": 14763251.0,
      "step": 7130
    },
    {
      "entropy": 5.224172782897949,
      "epoch": 6.480690595184007,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019652712269986074,
      "loss": 4.5942,
      "mean_token_accuracy": 0.2239474281668663,
      "num_tokens": 14773340.0,
      "step": 7135
    },
    {
      "entropy": 5.207349491119385,
      "epoch": 6.485233984552476,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019619598265845556,
      "loss": 4.6687,
      "mean_token_accuracy": 0.21406734138727188,
      "num_tokens": 14784451.0,
      "step": 7140
    },
    {
      "entropy": 5.122167015075684,
      "epoch": 6.489777373920945,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019586503705812969,
      "loss": 4.471,
      "mean_token_accuracy": 0.23433852791786194,
      "num_tokens": 14794142.0,
      "step": 7145
    },
    {
      "entropy": 5.153581285476685,
      "epoch": 6.494320763289414,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001955342867154582,
      "loss": 4.6465,
      "mean_token_accuracy": 0.22095524072647094,
      "num_tokens": 14804638.0,
      "step": 7150
    },
    {
      "entropy": 5.238812255859375,
      "epoch": 6.498864152657883,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019520373244653484,
      "loss": 4.6513,
      "mean_token_accuracy": 0.21803468614816665,
      "num_tokens": 14815295.0,
      "step": 7155
    },
    {
      "entropy": 5.2626025676727295,
      "epoch": 6.503407542026352,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001948733750669693,
      "loss": 4.6194,
      "mean_token_accuracy": 0.22134263068437576,
      "num_tokens": 14826212.0,
      "step": 7160
    },
    {
      "entropy": 5.258536052703858,
      "epoch": 6.50795093139482,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019454321539188568,
      "loss": 4.6289,
      "mean_token_accuracy": 0.2202504098415375,
      "num_tokens": 14836375.0,
      "step": 7165
    },
    {
      "entropy": 5.129872798919678,
      "epoch": 6.512494320763289,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019421325423592007,
      "loss": 4.5537,
      "mean_token_accuracy": 0.22856810688972473,
      "num_tokens": 14846800.0,
      "step": 7170
    },
    {
      "entropy": 5.176173734664917,
      "epoch": 6.517037710131758,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001938834924132189,
      "loss": 4.6091,
      "mean_token_accuracy": 0.21375721991062163,
      "num_tokens": 14858059.0,
      "step": 7175
    },
    {
      "entropy": 5.155012321472168,
      "epoch": 6.521581099500227,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001935539307374366,
      "loss": 4.625,
      "mean_token_accuracy": 0.22147468626499175,
      "num_tokens": 14868211.0,
      "step": 7180
    },
    {
      "entropy": 5.121175527572632,
      "epoch": 6.526124488868696,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019322457002173395,
      "loss": 4.4553,
      "mean_token_accuracy": 0.23349898904561997,
      "num_tokens": 14877130.0,
      "step": 7185
    },
    {
      "entropy": 5.2116203784942625,
      "epoch": 6.530667878237165,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019289541107877562,
      "loss": 4.5823,
      "mean_token_accuracy": 0.23042782992124558,
      "num_tokens": 14886864.0,
      "step": 7190
    },
    {
      "entropy": 5.184208869934082,
      "epoch": 6.535211267605634,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019256645472072876,
      "loss": 4.7355,
      "mean_token_accuracy": 0.2135016366839409,
      "num_tokens": 14897839.0,
      "step": 7195
    },
    {
      "entropy": 5.231349134445191,
      "epoch": 6.539754656974103,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019223770175926032,
      "loss": 4.5795,
      "mean_token_accuracy": 0.22261989265680313,
      "num_tokens": 14908594.0,
      "step": 7200
    },
    {
      "entropy": 5.189267778396607,
      "epoch": 6.544298046342572,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019190915300553574,
      "loss": 4.6489,
      "mean_token_accuracy": 0.20772677958011626,
      "num_tokens": 14919517.0,
      "step": 7205
    },
    {
      "entropy": 5.1563159942626955,
      "epoch": 6.54884143571104,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001915808092702162,
      "loss": 4.5984,
      "mean_token_accuracy": 0.2219265192747116,
      "num_tokens": 14929952.0,
      "step": 7210
    },
    {
      "entropy": 5.203795146942139,
      "epoch": 6.553384825079509,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001912526713634574,
      "loss": 4.6128,
      "mean_token_accuracy": 0.21955978721380234,
      "num_tokens": 14940600.0,
      "step": 7215
    },
    {
      "entropy": 5.233231925964356,
      "epoch": 6.557928214447978,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001909247400949069,
      "loss": 4.5391,
      "mean_token_accuracy": 0.2261827439069748,
      "num_tokens": 14951051.0,
      "step": 7220
    },
    {
      "entropy": 5.176024007797241,
      "epoch": 6.562471603816447,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019059701627370263,
      "loss": 4.5971,
      "mean_token_accuracy": 0.22293343693017958,
      "num_tokens": 14961309.0,
      "step": 7225
    },
    {
      "entropy": 5.172501039505005,
      "epoch": 6.567014993184916,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019026950070847025,
      "loss": 4.6311,
      "mean_token_accuracy": 0.21460071057081223,
      "num_tokens": 14971814.0,
      "step": 7230
    },
    {
      "entropy": 5.139586544036865,
      "epoch": 6.571558382553385,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018994219420732228,
      "loss": 4.5766,
      "mean_token_accuracy": 0.2271139770746231,
      "num_tokens": 14981776.0,
      "step": 7235
    },
    {
      "entropy": 5.181930494308472,
      "epoch": 6.576101771921854,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018961509757785462,
      "loss": 4.5854,
      "mean_token_accuracy": 0.2189624860882759,
      "num_tokens": 14991634.0,
      "step": 7240
    },
    {
      "entropy": 5.195900344848633,
      "epoch": 6.580645161290323,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018928821162714587,
      "loss": 4.592,
      "mean_token_accuracy": 0.22367185801267625,
      "num_tokens": 15002847.0,
      "step": 7245
    },
    {
      "entropy": 5.223254585266114,
      "epoch": 6.585188550658792,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018896153716175458,
      "loss": 4.6708,
      "mean_token_accuracy": 0.20723884403705597,
      "num_tokens": 15012968.0,
      "step": 7250
    },
    {
      "entropy": 5.1899271488189695,
      "epoch": 6.58973194002726,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001886350749877175,
      "loss": 4.5881,
      "mean_token_accuracy": 0.2216276228427887,
      "num_tokens": 15023523.0,
      "step": 7255
    },
    {
      "entropy": 5.18328595161438,
      "epoch": 6.594275329395729,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001883088259105476,
      "loss": 4.6044,
      "mean_token_accuracy": 0.22303034961223603,
      "num_tokens": 15033502.0,
      "step": 7260
    },
    {
      "entropy": 5.138995742797851,
      "epoch": 6.598818718764198,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018798279073523204,
      "loss": 4.5279,
      "mean_token_accuracy": 0.22692831605672836,
      "num_tokens": 15043641.0,
      "step": 7265
    },
    {
      "entropy": 5.189411544799805,
      "epoch": 6.603362108132667,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00018765697026623017,
      "loss": 4.6139,
      "mean_token_accuracy": 0.22267412543296813,
      "num_tokens": 15054063.0,
      "step": 7270
    },
    {
      "entropy": 5.22999906539917,
      "epoch": 6.607905497501136,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001873313653074718,
      "loss": 4.5851,
      "mean_token_accuracy": 0.22349095046520234,
      "num_tokens": 15064606.0,
      "step": 7275
    },
    {
      "entropy": 5.236790943145752,
      "epoch": 6.612448886869605,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018700597666235452,
      "loss": 4.6648,
      "mean_token_accuracy": 0.2196704089641571,
      "num_tokens": 15074871.0,
      "step": 7280
    },
    {
      "entropy": 5.246225738525391,
      "epoch": 6.616992276238074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018668080513374268,
      "loss": 4.6319,
      "mean_token_accuracy": 0.2161237731575966,
      "num_tokens": 15086112.0,
      "step": 7285
    },
    {
      "entropy": 5.171166563034058,
      "epoch": 6.621535665606542,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018635585152396457,
      "loss": 4.5911,
      "mean_token_accuracy": 0.22142619341611863,
      "num_tokens": 15096081.0,
      "step": 7290
    },
    {
      "entropy": 5.181109762191772,
      "epoch": 6.626079054975011,
      "grad_norm": 2.40625,
      "learning_rate": 0.000186031116634811,
      "loss": 4.544,
      "mean_token_accuracy": 0.22300596088171004,
      "num_tokens": 15105975.0,
      "step": 7295
    },
    {
      "entropy": 5.197613286972046,
      "epoch": 6.63062244434348,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018570660126753291,
      "loss": 4.6402,
      "mean_token_accuracy": 0.21844936460256575,
      "num_tokens": 15116428.0,
      "step": 7300
    },
    {
      "entropy": 5.242282724380493,
      "epoch": 6.635165833711949,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001853823062228399,
      "loss": 4.6049,
      "mean_token_accuracy": 0.22933634370565414,
      "num_tokens": 15127434.0,
      "step": 7305
    },
    {
      "entropy": 5.207888889312744,
      "epoch": 6.639709223080418,
      "grad_norm": 1.25,
      "learning_rate": 0.00018505823230089758,
      "loss": 4.6014,
      "mean_token_accuracy": 0.22107418477535248,
      "num_tokens": 15139220.0,
      "step": 7310
    },
    {
      "entropy": 5.162822628021241,
      "epoch": 6.644252612448887,
      "grad_norm": 1.5625,
      "learning_rate": 0.00018473438030132633,
      "loss": 4.5987,
      "mean_token_accuracy": 0.21979332119226455,
      "num_tokens": 15150154.0,
      "step": 7315
    },
    {
      "entropy": 5.221553468704224,
      "epoch": 6.648796001817356,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001844107510231985,
      "loss": 4.6247,
      "mean_token_accuracy": 0.2131568118929863,
      "num_tokens": 15160170.0,
      "step": 7320
    },
    {
      "entropy": 5.194230270385742,
      "epoch": 6.653339391185825,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018408734526503744,
      "loss": 4.6403,
      "mean_token_accuracy": 0.21666666865348816,
      "num_tokens": 15171817.0,
      "step": 7325
    },
    {
      "entropy": 5.108583450317383,
      "epoch": 6.657882780554294,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018376416382481442,
      "loss": 4.5013,
      "mean_token_accuracy": 0.2276254415512085,
      "num_tokens": 15181980.0,
      "step": 7330
    },
    {
      "entropy": 5.143319129943848,
      "epoch": 6.662426169922762,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018344120749994787,
      "loss": 4.5286,
      "mean_token_accuracy": 0.22677031010389329,
      "num_tokens": 15192056.0,
      "step": 7335
    },
    {
      "entropy": 5.206723213195801,
      "epoch": 6.666969559291231,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018311847708730005,
      "loss": 4.5817,
      "mean_token_accuracy": 0.2231358602643013,
      "num_tokens": 15202107.0,
      "step": 7340
    },
    {
      "entropy": 5.20794415473938,
      "epoch": 6.6715129486597,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001827959733831765,
      "loss": 4.6686,
      "mean_token_accuracy": 0.21760102212429047,
      "num_tokens": 15213574.0,
      "step": 7345
    },
    {
      "entropy": 5.126482629776001,
      "epoch": 6.676056338028169,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018247369718332274,
      "loss": 4.4869,
      "mean_token_accuracy": 0.2327742248773575,
      "num_tokens": 15223637.0,
      "step": 7350
    },
    {
      "entropy": 5.11256513595581,
      "epoch": 6.680599727396638,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018215164928292354,
      "loss": 4.5919,
      "mean_token_accuracy": 0.21699206680059432,
      "num_tokens": 15233472.0,
      "step": 7355
    },
    {
      "entropy": 5.109126996994019,
      "epoch": 6.685143116765107,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018182983047659984,
      "loss": 4.5972,
      "mean_token_accuracy": 0.21771769374608993,
      "num_tokens": 15243912.0,
      "step": 7360
    },
    {
      "entropy": 5.136125278472901,
      "epoch": 6.689686506133576,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018150824155840778,
      "loss": 4.5602,
      "mean_token_accuracy": 0.2235122561454773,
      "num_tokens": 15253771.0,
      "step": 7365
    },
    {
      "entropy": 5.2284064292907715,
      "epoch": 6.694229895502044,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001811868833218358,
      "loss": 4.6119,
      "mean_token_accuracy": 0.22236407846212386,
      "num_tokens": 15264153.0,
      "step": 7370
    },
    {
      "entropy": 5.164019393920898,
      "epoch": 6.698773284870513,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018086575655980376,
      "loss": 4.5431,
      "mean_token_accuracy": 0.22705066353082656,
      "num_tokens": 15274274.0,
      "step": 7375
    },
    {
      "entropy": 5.184140539169311,
      "epoch": 6.703316674238982,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018054486206465952,
      "loss": 4.6004,
      "mean_token_accuracy": 0.22022962272167207,
      "num_tokens": 15284284.0,
      "step": 7380
    },
    {
      "entropy": 5.128257799148559,
      "epoch": 6.707860063607451,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001802242006281788,
      "loss": 4.5923,
      "mean_token_accuracy": 0.22352205216884613,
      "num_tokens": 15295457.0,
      "step": 7385
    },
    {
      "entropy": 5.117687654495239,
      "epoch": 6.71240345297592,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017990377304156165,
      "loss": 4.5632,
      "mean_token_accuracy": 0.22293044328689576,
      "num_tokens": 15306091.0,
      "step": 7390
    },
    {
      "entropy": 5.13040680885315,
      "epoch": 6.716946842344389,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001795835800954312,
      "loss": 4.5885,
      "mean_token_accuracy": 0.22671784460544586,
      "num_tokens": 15315997.0,
      "step": 7395
    },
    {
      "entropy": 5.1597188949584964,
      "epoch": 6.721490231712858,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017926362257983174,
      "loss": 4.4971,
      "mean_token_accuracy": 0.23017482906579972,
      "num_tokens": 15325784.0,
      "step": 7400
    },
    {
      "entropy": 5.151299381256104,
      "epoch": 6.726033621081327,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001789439012842269,
      "loss": 4.548,
      "mean_token_accuracy": 0.2235654890537262,
      "num_tokens": 15336529.0,
      "step": 7405
    },
    {
      "entropy": 5.165550661087036,
      "epoch": 6.730577010449796,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017862441699749687,
      "loss": 4.5414,
      "mean_token_accuracy": 0.22446956783533095,
      "num_tokens": 15346225.0,
      "step": 7410
    },
    {
      "entropy": 5.18557186126709,
      "epoch": 6.735120399818264,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001783051705079375,
      "loss": 4.6026,
      "mean_token_accuracy": 0.21836166828870773,
      "num_tokens": 15357107.0,
      "step": 7415
    },
    {
      "entropy": 5.190396785736084,
      "epoch": 6.739663789186733,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017798616260325767,
      "loss": 4.5702,
      "mean_token_accuracy": 0.22598574161529542,
      "num_tokens": 15366575.0,
      "step": 7420
    },
    {
      "entropy": 5.151611328125,
      "epoch": 6.744207178555202,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001776673940705778,
      "loss": 4.4741,
      "mean_token_accuracy": 0.23158927112817765,
      "num_tokens": 15376067.0,
      "step": 7425
    },
    {
      "entropy": 5.263339424133301,
      "epoch": 6.748750567923671,
      "grad_norm": 1.640625,
      "learning_rate": 0.0001773488656964276,
      "loss": 4.6989,
      "mean_token_accuracy": 0.21261636018753052,
      "num_tokens": 15386936.0,
      "step": 7430
    },
    {
      "entropy": 5.146401596069336,
      "epoch": 6.75329395729214,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000177030578266744,
      "loss": 4.6014,
      "mean_token_accuracy": 0.22373099476099015,
      "num_tokens": 15397064.0,
      "step": 7435
    },
    {
      "entropy": 5.123068189620971,
      "epoch": 6.757837346660609,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017671253256686963,
      "loss": 4.5175,
      "mean_token_accuracy": 0.23300340324640273,
      "num_tokens": 15406620.0,
      "step": 7440
    },
    {
      "entropy": 5.182918739318848,
      "epoch": 6.762380736029078,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017639472938155075,
      "loss": 4.5341,
      "mean_token_accuracy": 0.23126738369464875,
      "num_tokens": 15416606.0,
      "step": 7445
    },
    {
      "entropy": 5.225580263137817,
      "epoch": 6.766924125397547,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017607716949493514,
      "loss": 4.6304,
      "mean_token_accuracy": 0.21704081892967225,
      "num_tokens": 15427438.0,
      "step": 7450
    },
    {
      "entropy": 5.163024187088013,
      "epoch": 6.771467514766016,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017575985369057008,
      "loss": 4.5991,
      "mean_token_accuracy": 0.21682407557964326,
      "num_tokens": 15438983.0,
      "step": 7455
    },
    {
      "entropy": 5.1535844802856445,
      "epoch": 6.776010904134484,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001754427827514008,
      "loss": 4.5367,
      "mean_token_accuracy": 0.22670070976018905,
      "num_tokens": 15449345.0,
      "step": 7460
    },
    {
      "entropy": 5.211802339553833,
      "epoch": 6.780554293502953,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017512595745976855,
      "loss": 4.6027,
      "mean_token_accuracy": 0.216689033806324,
      "num_tokens": 15459589.0,
      "step": 7465
    },
    {
      "entropy": 5.167437744140625,
      "epoch": 6.785097682871422,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017480937859740793,
      "loss": 4.5773,
      "mean_token_accuracy": 0.22208358645439147,
      "num_tokens": 15468765.0,
      "step": 7470
    },
    {
      "entropy": 5.108882713317871,
      "epoch": 6.789641072239891,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017449304694544608,
      "loss": 4.5726,
      "mean_token_accuracy": 0.22055044919252395,
      "num_tokens": 15479065.0,
      "step": 7475
    },
    {
      "entropy": 5.146591329574585,
      "epoch": 6.79418446160836,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017417696328439974,
      "loss": 4.5745,
      "mean_token_accuracy": 0.22215359658002853,
      "num_tokens": 15489028.0,
      "step": 7480
    },
    {
      "entropy": 5.176337146759034,
      "epoch": 6.798727850976829,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017386112839417402,
      "loss": 4.5042,
      "mean_token_accuracy": 0.23382718712091446,
      "num_tokens": 15497979.0,
      "step": 7485
    },
    {
      "entropy": 5.188509225845337,
      "epoch": 6.803271240345298,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017354554305405994,
      "loss": 4.6339,
      "mean_token_accuracy": 0.21927012503147125,
      "num_tokens": 15509690.0,
      "step": 7490
    },
    {
      "entropy": 5.160838556289673,
      "epoch": 6.807814629713766,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001732302080427332,
      "loss": 4.6184,
      "mean_token_accuracy": 0.22014106065034866,
      "num_tokens": 15519881.0,
      "step": 7495
    },
    {
      "entropy": 5.1574078559875485,
      "epoch": 6.812358019082235,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017291512413825148,
      "loss": 4.6093,
      "mean_token_accuracy": 0.21424129754304885,
      "num_tokens": 15528924.0,
      "step": 7500
    },
    {
      "epoch": 6.812358019082235,
      "eval_entropy": 5.081966128460197,
      "eval_loss": 5.656538963317871,
      "eval_mean_token_accuracy": 0.17496644106022147,
      "eval_num_tokens": 15528924.0,
      "eval_runtime": 2.7523,
      "eval_samples_per_second": 1248.033,
      "eval_steps_per_second": 156.231,
      "step": 7500
    }
  ],
  "logging_steps": 5,
  "max_steps": 11000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 2.2509855378432e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}