eng-latn-100mb-after-ppt-shuff-dyck-disjoint-tok-ckpt500_seed3407_seed3407/checkpoint-21000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.3585106980400967,
  "eval_steps": 3000,
  "global_step": 21000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.25639886856079,
      "epoch": 0.0005615769079575448,
      "grad_norm": 3.71875,
      "learning_rate": 2e-06,
      "loss": 11.7409,
      "mean_token_accuracy": 0.0026910859567578884,
      "num_tokens": 11659.0,
      "step": 5
    },
    {
      "entropy": 10.252698040008545,
      "epoch": 0.0011231538159150896,
      "grad_norm": 4.25,
      "learning_rate": 4.5e-06,
      "loss": 11.7094,
      "mean_token_accuracy": 0.0012989068985916675,
      "num_tokens": 21794.0,
      "step": 10
    },
    {
      "entropy": 10.256062889099121,
      "epoch": 0.0016847307238726343,
      "grad_norm": 4.09375,
      "learning_rate": 7e-06,
      "loss": 11.6893,
      "mean_token_accuracy": 0.002295289159519598,
      "num_tokens": 33206.0,
      "step": 15
    },
    {
      "entropy": 10.28339958190918,
      "epoch": 0.0022463076318301792,
      "grad_norm": 3.96875,
      "learning_rate": 9.5e-06,
      "loss": 11.5471,
      "mean_token_accuracy": 0.003405291523085907,
      "num_tokens": 44487.0,
      "step": 20
    },
    {
      "entropy": 10.283167362213135,
      "epoch": 0.002807884539787724,
      "grad_norm": 3.625,
      "learning_rate": 1.2e-05,
      "loss": 11.3861,
      "mean_token_accuracy": 0.006172553007490933,
      "num_tokens": 56062.0,
      "step": 25
    },
    {
      "entropy": 10.233351135253907,
      "epoch": 0.0033694614477452686,
      "grad_norm": 3.59375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 11.2207,
      "mean_token_accuracy": 0.022563868667930365,
      "num_tokens": 67453.0,
      "step": 30
    },
    {
      "entropy": 10.17716064453125,
      "epoch": 0.003931038355702813,
      "grad_norm": 3.25,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.9781,
      "mean_token_accuracy": 0.02911745272576809,
      "num_tokens": 78527.0,
      "step": 35
    },
    {
      "entropy": 10.325019550323486,
      "epoch": 0.0044926152636603585,
      "grad_norm": 2.640625,
      "learning_rate": 1.95e-05,
      "loss": 10.6651,
      "mean_token_accuracy": 0.030208175629377367,
      "num_tokens": 89267.0,
      "step": 40
    },
    {
      "entropy": 10.383963871002198,
      "epoch": 0.005054192171617903,
      "grad_norm": 2.59375,
      "learning_rate": 2.2e-05,
      "loss": 10.5824,
      "mean_token_accuracy": 0.02886309754103422,
      "num_tokens": 102035.0,
      "step": 45
    },
    {
      "entropy": 10.449808597564697,
      "epoch": 0.005615769079575448,
      "grad_norm": 2.546875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.4605,
      "mean_token_accuracy": 0.038186770118772986,
      "num_tokens": 112734.0,
      "step": 50
    },
    {
      "entropy": 10.461854267120362,
      "epoch": 0.006177345987532993,
      "grad_norm": 2.4375,
      "learning_rate": 2.7e-05,
      "loss": 10.3849,
      "mean_token_accuracy": 0.040889688022434714,
      "num_tokens": 123854.0,
      "step": 55
    },
    {
      "entropy": 10.472981071472168,
      "epoch": 0.006738922895490537,
      "grad_norm": 2.390625,
      "learning_rate": 2.95e-05,
      "loss": 10.295,
      "mean_token_accuracy": 0.037749376334249976,
      "num_tokens": 133444.0,
      "step": 60
    },
    {
      "entropy": 10.491286468505859,
      "epoch": 0.007300499803448083,
      "grad_norm": 2.515625,
      "learning_rate": 3.2e-05,
      "loss": 10.2382,
      "mean_token_accuracy": 0.03989502489566803,
      "num_tokens": 144678.0,
      "step": 65
    },
    {
      "entropy": 10.506498050689697,
      "epoch": 0.007862076711405626,
      "grad_norm": 2.546875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 10.1488,
      "mean_token_accuracy": 0.03764319252222777,
      "num_tokens": 155866.0,
      "step": 70
    },
    {
      "entropy": 10.518252658843995,
      "epoch": 0.008423653619363172,
      "grad_norm": 2.4375,
      "learning_rate": 3.7e-05,
      "loss": 10.0531,
      "mean_token_accuracy": 0.038073062710464,
      "num_tokens": 166965.0,
      "step": 75
    },
    {
      "entropy": 10.528139972686768,
      "epoch": 0.008985230527320717,
      "grad_norm": 2.34375,
      "learning_rate": 3.95e-05,
      "loss": 9.9445,
      "mean_token_accuracy": 0.039117959514260295,
      "num_tokens": 177423.0,
      "step": 80
    },
    {
      "entropy": 10.524243068695068,
      "epoch": 0.00954680743527826,
      "grad_norm": 2.375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.8407,
      "mean_token_accuracy": 0.044676617905497554,
      "num_tokens": 188687.0,
      "step": 85
    },
    {
      "entropy": 10.510469150543212,
      "epoch": 0.010108384343235806,
      "grad_norm": 2.328125,
      "learning_rate": 4.45e-05,
      "loss": 9.7592,
      "mean_token_accuracy": 0.038761510699987414,
      "num_tokens": 200173.0,
      "step": 90
    },
    {
      "entropy": 10.509880542755127,
      "epoch": 0.010669961251193351,
      "grad_norm": 2.453125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.6288,
      "mean_token_accuracy": 0.042973919026553634,
      "num_tokens": 210877.0,
      "step": 95
    },
    {
      "entropy": 10.515023994445801,
      "epoch": 0.011231538159150897,
      "grad_norm": 2.40625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.5036,
      "mean_token_accuracy": 0.039228996820747854,
      "num_tokens": 221770.0,
      "step": 100
    },
    {
      "entropy": 10.511959838867188,
      "epoch": 0.01179311506710844,
      "grad_norm": 2.359375,
      "learning_rate": 5.2e-05,
      "loss": 9.3655,
      "mean_token_accuracy": 0.041880004480481146,
      "num_tokens": 232140.0,
      "step": 105
    },
    {
      "entropy": 10.486880874633789,
      "epoch": 0.012354691975065986,
      "grad_norm": 2.234375,
      "learning_rate": 5.45e-05,
      "loss": 9.2756,
      "mean_token_accuracy": 0.03760720025748014,
      "num_tokens": 244186.0,
      "step": 110
    },
    {
      "entropy": 10.462724304199218,
      "epoch": 0.012916268883023531,
      "grad_norm": 2.25,
      "learning_rate": 5.7e-05,
      "loss": 9.084,
      "mean_token_accuracy": 0.038566050492227076,
      "num_tokens": 254805.0,
      "step": 115
    },
    {
      "entropy": 10.42501459121704,
      "epoch": 0.013477845790981074,
      "grad_norm": 2.34375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.8662,
      "mean_token_accuracy": 0.041522559337317945,
      "num_tokens": 265055.0,
      "step": 120
    },
    {
      "entropy": 10.37605791091919,
      "epoch": 0.01403942269893862,
      "grad_norm": 2.21875,
      "learning_rate": 6.2e-05,
      "loss": 8.7259,
      "mean_token_accuracy": 0.035992196202278136,
      "num_tokens": 275850.0,
      "step": 125
    },
    {
      "entropy": 10.301450824737548,
      "epoch": 0.014600999606896165,
      "grad_norm": 1.9609375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.5828,
      "mean_token_accuracy": 0.03876302167773247,
      "num_tokens": 286802.0,
      "step": 130
    },
    {
      "entropy": 10.22424955368042,
      "epoch": 0.015162576514853709,
      "grad_norm": 1.984375,
      "learning_rate": 6.7e-05,
      "loss": 8.4842,
      "mean_token_accuracy": 0.03725951910018921,
      "num_tokens": 297070.0,
      "step": 135
    },
    {
      "entropy": 10.137630462646484,
      "epoch": 0.015724153422811252,
      "grad_norm": 1.890625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.3182,
      "mean_token_accuracy": 0.04086887463927269,
      "num_tokens": 309617.0,
      "step": 140
    },
    {
      "entropy": 9.952799987792968,
      "epoch": 0.0162857303307688,
      "grad_norm": 1.734375,
      "learning_rate": 7.2e-05,
      "loss": 8.1687,
      "mean_token_accuracy": 0.04170650318264961,
      "num_tokens": 321310.0,
      "step": 145
    },
    {
      "entropy": 9.707593727111817,
      "epoch": 0.016847307238726343,
      "grad_norm": 1.484375,
      "learning_rate": 7.45e-05,
      "loss": 7.9867,
      "mean_token_accuracy": 0.04252769723534584,
      "num_tokens": 332243.0,
      "step": 150
    },
    {
      "entropy": 9.44668788909912,
      "epoch": 0.017408884146683887,
      "grad_norm": 1.15625,
      "learning_rate": 7.7e-05,
      "loss": 7.8717,
      "mean_token_accuracy": 0.03686592467129231,
      "num_tokens": 342847.0,
      "step": 155
    },
    {
      "entropy": 9.195470142364503,
      "epoch": 0.017970461054641434,
      "grad_norm": 1.140625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.7169,
      "mean_token_accuracy": 0.0400078721344471,
      "num_tokens": 353570.0,
      "step": 160
    },
    {
      "entropy": 8.776480579376221,
      "epoch": 0.018532037962598977,
      "grad_norm": 0.9453125,
      "learning_rate": 8.2e-05,
      "loss": 7.6303,
      "mean_token_accuracy": 0.04083114080131054,
      "num_tokens": 364330.0,
      "step": 165
    },
    {
      "entropy": 8.479571533203124,
      "epoch": 0.01909361487055652,
      "grad_norm": 0.98828125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.6035,
      "mean_token_accuracy": 0.039075618609786034,
      "num_tokens": 374930.0,
      "step": 170
    },
    {
      "entropy": 8.282562637329102,
      "epoch": 0.019655191778514068,
      "grad_norm": 0.9453125,
      "learning_rate": 8.7e-05,
      "loss": 7.5491,
      "mean_token_accuracy": 0.03750791251659393,
      "num_tokens": 384690.0,
      "step": 175
    },
    {
      "entropy": 8.006285572052002,
      "epoch": 0.020216768686471612,
      "grad_norm": 0.99609375,
      "learning_rate": 8.95e-05,
      "loss": 7.5439,
      "mean_token_accuracy": 0.038402245566248897,
      "num_tokens": 395753.0,
      "step": 180
    },
    {
      "entropy": 7.9631373405456545,
      "epoch": 0.020778345594429155,
      "grad_norm": 0.98046875,
      "learning_rate": 9.2e-05,
      "loss": 7.493,
      "mean_token_accuracy": 0.0437284167855978,
      "num_tokens": 407049.0,
      "step": 185
    },
    {
      "entropy": 7.781239318847656,
      "epoch": 0.021339922502386702,
      "grad_norm": 0.88671875,
      "learning_rate": 9.45e-05,
      "loss": 7.4957,
      "mean_token_accuracy": 0.04678881764411926,
      "num_tokens": 417942.0,
      "step": 190
    },
    {
      "entropy": 7.785053539276123,
      "epoch": 0.021901499410344246,
      "grad_norm": 0.94921875,
      "learning_rate": 9.7e-05,
      "loss": 7.5136,
      "mean_token_accuracy": 0.04727039746940136,
      "num_tokens": 428151.0,
      "step": 195
    },
    {
      "entropy": 7.740858125686645,
      "epoch": 0.022463076318301793,
      "grad_norm": 0.828125,
      "learning_rate": 9.95e-05,
      "loss": 7.5696,
      "mean_token_accuracy": 0.04193144179880619,
      "num_tokens": 439837.0,
      "step": 200
    },
    {
      "entropy": 7.758722639083862,
      "epoch": 0.023024653226259337,
      "grad_norm": 0.7578125,
      "learning_rate": 0.000102,
      "loss": 7.4302,
      "mean_token_accuracy": 0.045322371646761894,
      "num_tokens": 450357.0,
      "step": 205
    },
    {
      "entropy": 7.715395164489746,
      "epoch": 0.02358623013421688,
      "grad_norm": 0.859375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4423,
      "mean_token_accuracy": 0.058253410086035726,
      "num_tokens": 461174.0,
      "step": 210
    },
    {
      "entropy": 7.59510555267334,
      "epoch": 0.024147807042174427,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000107,
      "loss": 7.5059,
      "mean_token_accuracy": 0.058244950324296954,
      "num_tokens": 472785.0,
      "step": 215
    },
    {
      "entropy": 7.6779868602752686,
      "epoch": 0.02470938395013197,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0001095,
      "loss": 7.458,
      "mean_token_accuracy": 0.06584881767630577,
      "num_tokens": 483248.0,
      "step": 220
    },
    {
      "entropy": 7.6262555599212645,
      "epoch": 0.025270960858089515,
      "grad_norm": 0.80078125,
      "learning_rate": 0.000112,
      "loss": 7.3474,
      "mean_token_accuracy": 0.06667502745985984,
      "num_tokens": 494690.0,
      "step": 225
    },
    {
      "entropy": 7.683849191665649,
      "epoch": 0.025832537766047062,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0001145,
      "loss": 7.5293,
      "mean_token_accuracy": 0.0678394366055727,
      "num_tokens": 506192.0,
      "step": 230
    },
    {
      "entropy": 7.675125026702881,
      "epoch": 0.026394114674004605,
      "grad_norm": 0.875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.3906,
      "mean_token_accuracy": 0.07262160144746303,
      "num_tokens": 517152.0,
      "step": 235
    },
    {
      "entropy": 7.534460210800171,
      "epoch": 0.02695569158196215,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.3269,
      "mean_token_accuracy": 0.0788414567708969,
      "num_tokens": 527836.0,
      "step": 240
    },
    {
      "entropy": 7.556740760803223,
      "epoch": 0.027517268489919696,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000122,
      "loss": 7.4559,
      "mean_token_accuracy": 0.07313848771154881,
      "num_tokens": 539491.0,
      "step": 245
    },
    {
      "entropy": 7.651741409301758,
      "epoch": 0.02807884539787724,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0001245,
      "loss": 7.4231,
      "mean_token_accuracy": 0.06984101124107837,
      "num_tokens": 550816.0,
      "step": 250
    },
    {
      "entropy": 7.559566640853882,
      "epoch": 0.028640422305834783,
      "grad_norm": 0.765625,
      "learning_rate": 0.000127,
      "loss": 7.4417,
      "mean_token_accuracy": 0.07528160624206066,
      "num_tokens": 561566.0,
      "step": 255
    },
    {
      "entropy": 7.60768313407898,
      "epoch": 0.02920199921379233,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0001295,
      "loss": 7.3847,
      "mean_token_accuracy": 0.07447496987879276,
      "num_tokens": 573032.0,
      "step": 260
    },
    {
      "entropy": 7.602000761032104,
      "epoch": 0.029763576121749874,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000132,
      "loss": 7.3083,
      "mean_token_accuracy": 0.07689321786165237,
      "num_tokens": 584140.0,
      "step": 265
    },
    {
      "entropy": 7.546748113632202,
      "epoch": 0.030325153029707418,
      "grad_norm": 0.74609375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.3863,
      "mean_token_accuracy": 0.0756129652261734,
      "num_tokens": 595097.0,
      "step": 270
    },
    {
      "entropy": 7.626759958267212,
      "epoch": 0.030886729937664965,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.3776,
      "mean_token_accuracy": 0.07868644669651985,
      "num_tokens": 606797.0,
      "step": 275
    },
    {
      "entropy": 7.603859090805054,
      "epoch": 0.031448306845622505,
      "grad_norm": 0.921875,
      "learning_rate": 0.0001395,
      "loss": 7.3381,
      "mean_token_accuracy": 0.0748847633600235,
      "num_tokens": 618187.0,
      "step": 280
    },
    {
      "entropy": 7.569405174255371,
      "epoch": 0.032009883753580055,
      "grad_norm": 0.7734375,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.2915,
      "mean_token_accuracy": 0.08101793453097343,
      "num_tokens": 629829.0,
      "step": 285
    },
    {
      "entropy": 7.554937267303467,
      "epoch": 0.0325714606615376,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0001445,
      "loss": 7.2766,
      "mean_token_accuracy": 0.08518879413604737,
      "num_tokens": 640914.0,
      "step": 290
    },
    {
      "entropy": 7.543597173690796,
      "epoch": 0.03313303756949514,
      "grad_norm": 0.9375,
      "learning_rate": 0.000147,
      "loss": 7.2827,
      "mean_token_accuracy": 0.0838129960000515,
      "num_tokens": 653057.0,
      "step": 295
    },
    {
      "entropy": 7.536286020278931,
      "epoch": 0.033694614477452686,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0001495,
      "loss": 7.2208,
      "mean_token_accuracy": 0.08279290944337844,
      "num_tokens": 663437.0,
      "step": 300
    },
    {
      "entropy": 7.500700044631958,
      "epoch": 0.03425619138541023,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000152,
      "loss": 7.2719,
      "mean_token_accuracy": 0.08277729079127312,
      "num_tokens": 675416.0,
      "step": 305
    },
    {
      "entropy": 7.403319644927978,
      "epoch": 0.03481776829336777,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.0849,
      "mean_token_accuracy": 0.08483515977859497,
      "num_tokens": 686633.0,
      "step": 310
    },
    {
      "entropy": 7.422772455215454,
      "epoch": 0.035379345201325324,
      "grad_norm": 1.015625,
      "learning_rate": 0.000157,
      "loss": 7.173,
      "mean_token_accuracy": 0.08649655506014824,
      "num_tokens": 698065.0,
      "step": 315
    },
    {
      "entropy": 7.5332183837890625,
      "epoch": 0.03594092210928287,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0001595,
      "loss": 7.1756,
      "mean_token_accuracy": 0.08641982153058052,
      "num_tokens": 708652.0,
      "step": 320
    },
    {
      "entropy": 7.479076910018921,
      "epoch": 0.03650249901724041,
      "grad_norm": 0.9375,
      "learning_rate": 0.000162,
      "loss": 7.2977,
      "mean_token_accuracy": 0.08264525830745698,
      "num_tokens": 720199.0,
      "step": 325
    },
    {
      "entropy": 7.496834087371826,
      "epoch": 0.037064075925197955,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.2999,
      "mean_token_accuracy": 0.08251902386546135,
      "num_tokens": 732964.0,
      "step": 330
    },
    {
      "entropy": 7.395641613006592,
      "epoch": 0.0376256528331555,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.0579,
      "mean_token_accuracy": 0.08676571622490883,
      "num_tokens": 743528.0,
      "step": 335
    },
    {
      "entropy": 7.39327449798584,
      "epoch": 0.03818722974111304,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.1187,
      "mean_token_accuracy": 0.0880549892783165,
      "num_tokens": 754365.0,
      "step": 340
    },
    {
      "entropy": 7.318950414657593,
      "epoch": 0.03874880664907059,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.0332,
      "mean_token_accuracy": 0.08635925874114037,
      "num_tokens": 766094.0,
      "step": 345
    },
    {
      "entropy": 7.353262042999267,
      "epoch": 0.039310383557028136,
      "grad_norm": 0.984375,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.0952,
      "mean_token_accuracy": 0.09182592332363129,
      "num_tokens": 777275.0,
      "step": 350
    },
    {
      "entropy": 7.3257616519927975,
      "epoch": 0.03987196046498568,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000177,
      "loss": 7.136,
      "mean_token_accuracy": 0.08920007646083832,
      "num_tokens": 789436.0,
      "step": 355
    },
    {
      "entropy": 7.399164628982544,
      "epoch": 0.040433537372943223,
      "grad_norm": 0.921875,
      "learning_rate": 0.0001795,
      "loss": 7.1632,
      "mean_token_accuracy": 0.08764473721385002,
      "num_tokens": 800164.0,
      "step": 360
    },
    {
      "entropy": 7.224786233901978,
      "epoch": 0.04099511428090077,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000182,
      "loss": 7.0352,
      "mean_token_accuracy": 0.09125346541404725,
      "num_tokens": 811797.0,
      "step": 365
    },
    {
      "entropy": 7.286064577102661,
      "epoch": 0.04155669118885831,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001845,
      "loss": 7.0431,
      "mean_token_accuracy": 0.09055957570672035,
      "num_tokens": 824158.0,
      "step": 370
    },
    {
      "entropy": 7.305629539489746,
      "epoch": 0.04211826809681586,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000187,
      "loss": 7.0457,
      "mean_token_accuracy": 0.09202495589852333,
      "num_tokens": 835803.0,
      "step": 375
    },
    {
      "entropy": 7.371310424804688,
      "epoch": 0.042679845004773405,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0001895,
      "loss": 7.1098,
      "mean_token_accuracy": 0.08957164734601974,
      "num_tokens": 848106.0,
      "step": 380
    },
    {
      "entropy": 7.336284589767456,
      "epoch": 0.04324142191273095,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000192,
      "loss": 7.063,
      "mean_token_accuracy": 0.09114922881126404,
      "num_tokens": 859101.0,
      "step": 385
    },
    {
      "entropy": 7.226832294464112,
      "epoch": 0.04380299882068849,
      "grad_norm": 0.796875,
      "learning_rate": 0.0001945,
      "loss": 6.9601,
      "mean_token_accuracy": 0.0964093454182148,
      "num_tokens": 870309.0,
      "step": 390
    },
    {
      "entropy": 7.222394752502441,
      "epoch": 0.044364575728646036,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0239,
      "mean_token_accuracy": 0.10131923705339432,
      "num_tokens": 881515.0,
      "step": 395
    },
    {
      "entropy": 7.244077253341675,
      "epoch": 0.044926152636603586,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.9904,
      "mean_token_accuracy": 0.09794511124491692,
      "num_tokens": 892129.0,
      "step": 400
    },
    {
      "entropy": 7.264095592498779,
      "epoch": 0.04548772954456113,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000202,
      "loss": 7.0037,
      "mean_token_accuracy": 0.09216101467609406,
      "num_tokens": 903186.0,
      "step": 405
    },
    {
      "entropy": 7.212151622772216,
      "epoch": 0.046049306452518673,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.9559,
      "mean_token_accuracy": 0.10168264731764794,
      "num_tokens": 913471.0,
      "step": 410
    },
    {
      "entropy": 7.21423306465149,
      "epoch": 0.04661088336047622,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000207,
      "loss": 6.9665,
      "mean_token_accuracy": 0.09732877537608146,
      "num_tokens": 924283.0,
      "step": 415
    },
    {
      "entropy": 7.138846921920776,
      "epoch": 0.04717246026843376,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0002095,
      "loss": 6.9948,
      "mean_token_accuracy": 0.09952060505747795,
      "num_tokens": 934975.0,
      "step": 420
    },
    {
      "entropy": 7.175388956069947,
      "epoch": 0.047734037176391304,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000212,
      "loss": 6.8805,
      "mean_token_accuracy": 0.10059212520718575,
      "num_tokens": 945587.0,
      "step": 425
    },
    {
      "entropy": 7.089182806015015,
      "epoch": 0.048295614084348855,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0002145,
      "loss": 6.9372,
      "mean_token_accuracy": 0.09849796593189239,
      "num_tokens": 955935.0,
      "step": 430
    },
    {
      "entropy": 7.184536552429199,
      "epoch": 0.0488571909923064,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.889,
      "mean_token_accuracy": 0.10151714682579041,
      "num_tokens": 967960.0,
      "step": 435
    },
    {
      "entropy": 7.127185440063476,
      "epoch": 0.04941876790026394,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0002195,
      "loss": 6.9301,
      "mean_token_accuracy": 0.10072232410311699,
      "num_tokens": 979680.0,
      "step": 440
    },
    {
      "entropy": 7.1458930492401125,
      "epoch": 0.049980344808221486,
      "grad_norm": 0.79296875,
      "learning_rate": 0.000222,
      "loss": 7.0148,
      "mean_token_accuracy": 0.1004769742488861,
      "num_tokens": 990600.0,
      "step": 445
    },
    {
      "entropy": 7.197234964370727,
      "epoch": 0.05054192171617903,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002245,
      "loss": 7.0244,
      "mean_token_accuracy": 0.10357746407389641,
      "num_tokens": 1002331.0,
      "step": 450
    },
    {
      "entropy": 6.973642158508301,
      "epoch": 0.05110349862413657,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.8315,
      "mean_token_accuracy": 0.1050386182963848,
      "num_tokens": 1013473.0,
      "step": 455
    },
    {
      "entropy": 7.1875935077667235,
      "epoch": 0.051665075532094124,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.9257,
      "mean_token_accuracy": 0.09776141494512558,
      "num_tokens": 1024434.0,
      "step": 460
    },
    {
      "entropy": 7.147069692611694,
      "epoch": 0.05222665244005167,
      "grad_norm": 0.84375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.9539,
      "mean_token_accuracy": 0.09493176713585853,
      "num_tokens": 1035452.0,
      "step": 465
    },
    {
      "entropy": 7.040191555023194,
      "epoch": 0.05278822934800921,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8166,
      "mean_token_accuracy": 0.10042703226208687,
      "num_tokens": 1046213.0,
      "step": 470
    },
    {
      "entropy": 7.027799129486084,
      "epoch": 0.053349806255966754,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000237,
      "loss": 6.8431,
      "mean_token_accuracy": 0.1014677569270134,
      "num_tokens": 1057469.0,
      "step": 475
    },
    {
      "entropy": 7.141836309432984,
      "epoch": 0.0539113831639243,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0002395,
      "loss": 6.9903,
      "mean_token_accuracy": 0.10095212161540985,
      "num_tokens": 1069543.0,
      "step": 480
    },
    {
      "entropy": 6.945070791244507,
      "epoch": 0.05447296007188184,
      "grad_norm": 1.015625,
      "learning_rate": 0.000242,
      "loss": 6.8419,
      "mean_token_accuracy": 0.10394882559776306,
      "num_tokens": 1080476.0,
      "step": 485
    },
    {
      "entropy": 7.250262975692749,
      "epoch": 0.05503453697983939,
      "grad_norm": 0.90625,
      "learning_rate": 0.0002445,
      "loss": 6.9611,
      "mean_token_accuracy": 0.10339747741818428,
      "num_tokens": 1091700.0,
      "step": 490
    },
    {
      "entropy": 6.982399654388428,
      "epoch": 0.055596113887796936,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000247,
      "loss": 6.8471,
      "mean_token_accuracy": 0.0988765001296997,
      "num_tokens": 1103056.0,
      "step": 495
    },
    {
      "entropy": 7.01871132850647,
      "epoch": 0.05615769079575448,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002495,
      "loss": 6.9121,
      "mean_token_accuracy": 0.1074085734784603,
      "num_tokens": 1114836.0,
      "step": 500
    },
    {
      "entropy": 7.023473834991455,
      "epoch": 0.05671926770371202,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000252,
      "loss": 6.8057,
      "mean_token_accuracy": 0.10938784778118134,
      "num_tokens": 1125993.0,
      "step": 505
    },
    {
      "entropy": 7.023718929290771,
      "epoch": 0.05728084461166957,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002545,
      "loss": 6.8839,
      "mean_token_accuracy": 0.11009863540530204,
      "num_tokens": 1136719.0,
      "step": 510
    },
    {
      "entropy": 6.948973846435547,
      "epoch": 0.05784242151962711,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000257,
      "loss": 6.7549,
      "mean_token_accuracy": 0.10685163512825965,
      "num_tokens": 1146947.0,
      "step": 515
    },
    {
      "entropy": 7.0727332592010494,
      "epoch": 0.05840399842758466,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002595,
      "loss": 6.8908,
      "mean_token_accuracy": 0.10683172717690467,
      "num_tokens": 1157974.0,
      "step": 520
    },
    {
      "entropy": 6.95055775642395,
      "epoch": 0.058965575335542204,
      "grad_norm": 1.046875,
      "learning_rate": 0.000262,
      "loss": 6.8077,
      "mean_token_accuracy": 0.10525572448968887,
      "num_tokens": 1168573.0,
      "step": 525
    },
    {
      "entropy": 7.000631856918335,
      "epoch": 0.05952715224349975,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7251,
      "mean_token_accuracy": 0.10736186131834984,
      "num_tokens": 1178446.0,
      "step": 530
    },
    {
      "entropy": 6.959245443344116,
      "epoch": 0.06008872915145729,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.7478,
      "mean_token_accuracy": 0.10534990206360817,
      "num_tokens": 1190558.0,
      "step": 535
    },
    {
      "entropy": 6.902206563949585,
      "epoch": 0.060650306059414835,
      "grad_norm": 0.875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.9124,
      "mean_token_accuracy": 0.10460811406373978,
      "num_tokens": 1201929.0,
      "step": 540
    },
    {
      "entropy": 6.9541998386383055,
      "epoch": 0.06121188296737238,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.7425,
      "mean_token_accuracy": 0.11181432232260705,
      "num_tokens": 1213155.0,
      "step": 545
    },
    {
      "entropy": 7.080395078659057,
      "epoch": 0.06177345987532993,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002745,
      "loss": 6.8042,
      "mean_token_accuracy": 0.10288820937275886,
      "num_tokens": 1223961.0,
      "step": 550
    },
    {
      "entropy": 6.755504512786866,
      "epoch": 0.06233503678328747,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000277,
      "loss": 6.7008,
      "mean_token_accuracy": 0.11182279288768768,
      "num_tokens": 1233888.0,
      "step": 555
    },
    {
      "entropy": 6.958348941802979,
      "epoch": 0.06289661369124501,
      "grad_norm": 0.84375,
      "learning_rate": 0.0002795,
      "loss": 6.8077,
      "mean_token_accuracy": 0.1030250370502472,
      "num_tokens": 1244405.0,
      "step": 560
    },
    {
      "entropy": 7.0104207515716555,
      "epoch": 0.06345819059920256,
      "grad_norm": 0.796875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.7471,
      "mean_token_accuracy": 0.1139664702117443,
      "num_tokens": 1257099.0,
      "step": 565
    },
    {
      "entropy": 6.860793161392212,
      "epoch": 0.06401976750716011,
      "grad_norm": 0.90625,
      "learning_rate": 0.0002845,
      "loss": 6.8584,
      "mean_token_accuracy": 0.11190064251422882,
      "num_tokens": 1267965.0,
      "step": 570
    },
    {
      "entropy": 6.909835290908814,
      "epoch": 0.06458134441511765,
      "grad_norm": 0.953125,
      "learning_rate": 0.000287,
      "loss": 6.747,
      "mean_token_accuracy": 0.11047227904200554,
      "num_tokens": 1280075.0,
      "step": 575
    },
    {
      "entropy": 6.996085596084595,
      "epoch": 0.0651429213230752,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002895,
      "loss": 6.7767,
      "mean_token_accuracy": 0.11513174846768379,
      "num_tokens": 1290574.0,
      "step": 580
    },
    {
      "entropy": 6.839344120025634,
      "epoch": 0.06570449823103273,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000292,
      "loss": 6.7449,
      "mean_token_accuracy": 0.11083959862589836,
      "num_tokens": 1302680.0,
      "step": 585
    },
    {
      "entropy": 6.863824939727783,
      "epoch": 0.06626607513899029,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002945,
      "loss": 6.727,
      "mean_token_accuracy": 0.11484655365347862,
      "num_tokens": 1314577.0,
      "step": 590
    },
    {
      "entropy": 6.9280845642089846,
      "epoch": 0.06682765204694784,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000297,
      "loss": 6.8363,
      "mean_token_accuracy": 0.11602571532130242,
      "num_tokens": 1326560.0,
      "step": 595
    },
    {
      "entropy": 7.001092004776001,
      "epoch": 0.06738922895490537,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0002995,
      "loss": 6.9057,
      "mean_token_accuracy": 0.10357883349061012,
      "num_tokens": 1339105.0,
      "step": 600
    },
    {
      "entropy": 6.722957897186279,
      "epoch": 0.06795080586286292,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000302,
      "loss": 6.5765,
      "mean_token_accuracy": 0.11965182796120644,
      "num_tokens": 1349914.0,
      "step": 605
    },
    {
      "entropy": 6.931371068954467,
      "epoch": 0.06851238277082046,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0003045,
      "loss": 6.7924,
      "mean_token_accuracy": 0.11453562378883361,
      "num_tokens": 1361700.0,
      "step": 610
    },
    {
      "entropy": 6.858430242538452,
      "epoch": 0.06907395967877801,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000307,
      "loss": 6.8079,
      "mean_token_accuracy": 0.1133706197142601,
      "num_tokens": 1373103.0,
      "step": 615
    },
    {
      "entropy": 6.822653102874756,
      "epoch": 0.06963553658673555,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003095,
      "loss": 6.6143,
      "mean_token_accuracy": 0.11548932567238808,
      "num_tokens": 1383742.0,
      "step": 620
    },
    {
      "entropy": 6.799988889694214,
      "epoch": 0.0701971134946931,
      "grad_norm": 0.8046875,
      "learning_rate": 0.000312,
      "loss": 6.7658,
      "mean_token_accuracy": 0.11335846111178398,
      "num_tokens": 1396656.0,
      "step": 625
    },
    {
      "entropy": 6.776160717010498,
      "epoch": 0.07075869040265065,
      "grad_norm": 1.0,
      "learning_rate": 0.0003145,
      "loss": 6.6291,
      "mean_token_accuracy": 0.12655726373195647,
      "num_tokens": 1406709.0,
      "step": 630
    },
    {
      "entropy": 6.81230092048645,
      "epoch": 0.07132026731060818,
      "grad_norm": 0.79296875,
      "learning_rate": 0.000317,
      "loss": 6.677,
      "mean_token_accuracy": 0.120645322650671,
      "num_tokens": 1417793.0,
      "step": 635
    },
    {
      "entropy": 6.884833669662475,
      "epoch": 0.07188184421856574,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003195,
      "loss": 6.7582,
      "mean_token_accuracy": 0.11574307680130005,
      "num_tokens": 1429093.0,
      "step": 640
    },
    {
      "entropy": 6.699341487884522,
      "epoch": 0.07244342112652327,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000322,
      "loss": 6.6679,
      "mean_token_accuracy": 0.12265671268105507,
      "num_tokens": 1440907.0,
      "step": 645
    },
    {
      "entropy": 6.890548658370972,
      "epoch": 0.07300499803448082,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.7498,
      "mean_token_accuracy": 0.1140442967414856,
      "num_tokens": 1452089.0,
      "step": 650
    },
    {
      "entropy": 6.8591334342956545,
      "epoch": 0.07356657494243837,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7635,
      "mean_token_accuracy": 0.11165911853313445,
      "num_tokens": 1464378.0,
      "step": 655
    },
    {
      "entropy": 6.80465030670166,
      "epoch": 0.07412815185039591,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.687,
      "mean_token_accuracy": 0.11683566421270371,
      "num_tokens": 1476678.0,
      "step": 660
    },
    {
      "entropy": 6.709757089614868,
      "epoch": 0.07468972875835346,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5363,
      "mean_token_accuracy": 0.121712426841259,
      "num_tokens": 1486808.0,
      "step": 665
    },
    {
      "entropy": 6.838316631317139,
      "epoch": 0.075251305666311,
      "grad_norm": 0.859375,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.8144,
      "mean_token_accuracy": 0.1120360791683197,
      "num_tokens": 1499281.0,
      "step": 670
    },
    {
      "entropy": 6.873385381698609,
      "epoch": 0.07581288257426855,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000337,
      "loss": 6.7205,
      "mean_token_accuracy": 0.11793086603283882,
      "num_tokens": 1510202.0,
      "step": 675
    },
    {
      "entropy": 6.792718982696533,
      "epoch": 0.07637445948222608,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0003395,
      "loss": 6.5831,
      "mean_token_accuracy": 0.12022969275712966,
      "num_tokens": 1521492.0,
      "step": 680
    },
    {
      "entropy": 6.644273567199707,
      "epoch": 0.07693603639018363,
      "grad_norm": 1.046875,
      "learning_rate": 0.000342,
      "loss": 6.6175,
      "mean_token_accuracy": 0.12695625126361848,
      "num_tokens": 1532856.0,
      "step": 685
    },
    {
      "entropy": 6.711900901794434,
      "epoch": 0.07749761329814119,
      "grad_norm": 0.984375,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6209,
      "mean_token_accuracy": 0.1197846882045269,
      "num_tokens": 1543408.0,
      "step": 690
    },
    {
      "entropy": 6.7214684009552,
      "epoch": 0.07805919020609872,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000347,
      "loss": 6.6278,
      "mean_token_accuracy": 0.11783769056200981,
      "num_tokens": 1555474.0,
      "step": 695
    },
    {
      "entropy": 6.796803903579712,
      "epoch": 0.07862076711405627,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003495,
      "loss": 6.7268,
      "mean_token_accuracy": 0.1210656888782978,
      "num_tokens": 1567484.0,
      "step": 700
    },
    {
      "entropy": 6.696943473815918,
      "epoch": 0.07918234402201381,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000352,
      "loss": 6.5718,
      "mean_token_accuracy": 0.1262438789010048,
      "num_tokens": 1578204.0,
      "step": 705
    },
    {
      "entropy": 6.750957107543945,
      "epoch": 0.07974392092997136,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003545,
      "loss": 6.5652,
      "mean_token_accuracy": 0.12253414019942284,
      "num_tokens": 1589012.0,
      "step": 710
    },
    {
      "entropy": 6.546931219100952,
      "epoch": 0.08030549783792891,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000357,
      "loss": 6.5747,
      "mean_token_accuracy": 0.12381779551506042,
      "num_tokens": 1601099.0,
      "step": 715
    },
    {
      "entropy": 6.817351818084717,
      "epoch": 0.08086707474588645,
      "grad_norm": 0.859375,
      "learning_rate": 0.0003595,
      "loss": 6.7402,
      "mean_token_accuracy": 0.11719322726130485,
      "num_tokens": 1612814.0,
      "step": 720
    },
    {
      "entropy": 6.713132858276367,
      "epoch": 0.081428651653844,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000362,
      "loss": 6.6325,
      "mean_token_accuracy": 0.11970261111855507,
      "num_tokens": 1624953.0,
      "step": 725
    },
    {
      "entropy": 6.730032444000244,
      "epoch": 0.08199022856180153,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0003645,
      "loss": 6.6488,
      "mean_token_accuracy": 0.12532211914658548,
      "num_tokens": 1635673.0,
      "step": 730
    },
    {
      "entropy": 6.764851093292236,
      "epoch": 0.08255180546975908,
      "grad_norm": 0.875,
      "learning_rate": 0.000367,
      "loss": 6.5684,
      "mean_token_accuracy": 0.1264908105134964,
      "num_tokens": 1647282.0,
      "step": 735
    },
    {
      "entropy": 6.644136762619018,
      "epoch": 0.08311338237771662,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003695,
      "loss": 6.576,
      "mean_token_accuracy": 0.12845728248357774,
      "num_tokens": 1658426.0,
      "step": 740
    },
    {
      "entropy": 6.66550931930542,
      "epoch": 0.08367495928567417,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000372,
      "loss": 6.6157,
      "mean_token_accuracy": 0.1237633615732193,
      "num_tokens": 1669958.0,
      "step": 745
    },
    {
      "entropy": 6.739824438095093,
      "epoch": 0.08423653619363172,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003745,
      "loss": 6.6554,
      "mean_token_accuracy": 0.12233619689941407,
      "num_tokens": 1682409.0,
      "step": 750
    },
    {
      "entropy": 6.677445888519287,
      "epoch": 0.08479811310158926,
      "grad_norm": 0.96875,
      "learning_rate": 0.000377,
      "loss": 6.5781,
      "mean_token_accuracy": 0.13024282306432725,
      "num_tokens": 1692409.0,
      "step": 755
    },
    {
      "entropy": 6.733970499038696,
      "epoch": 0.08535969000954681,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003795,
      "loss": 6.6821,
      "mean_token_accuracy": 0.12274667993187904,
      "num_tokens": 1705018.0,
      "step": 760
    },
    {
      "entropy": 6.672050523757934,
      "epoch": 0.08592126691750435,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000382,
      "loss": 6.6111,
      "mean_token_accuracy": 0.12385557517409325,
      "num_tokens": 1715625.0,
      "step": 765
    },
    {
      "entropy": 6.624249505996704,
      "epoch": 0.0864828438254619,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003845,
      "loss": 6.6708,
      "mean_token_accuracy": 0.11638602316379547,
      "num_tokens": 1727045.0,
      "step": 770
    },
    {
      "entropy": 6.845441389083862,
      "epoch": 0.08704442073341945,
      "grad_norm": 0.921875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.7396,
      "mean_token_accuracy": 0.12483799159526825,
      "num_tokens": 1737374.0,
      "step": 775
    },
    {
      "entropy": 6.763451385498047,
      "epoch": 0.08760599764137698,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.7368,
      "mean_token_accuracy": 0.1217820793390274,
      "num_tokens": 1747788.0,
      "step": 780
    },
    {
      "entropy": 6.595237827301025,
      "epoch": 0.08816757454933453,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.4325,
      "mean_token_accuracy": 0.13330992460250854,
      "num_tokens": 1757769.0,
      "step": 785
    },
    {
      "entropy": 6.58812689781189,
      "epoch": 0.08872915145729207,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.5469,
      "mean_token_accuracy": 0.12191837057471275,
      "num_tokens": 1769246.0,
      "step": 790
    },
    {
      "entropy": 6.780652809143066,
      "epoch": 0.08929072836524962,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.6634,
      "mean_token_accuracy": 0.123942082375288,
      "num_tokens": 1781306.0,
      "step": 795
    },
    {
      "entropy": 6.602289724349975,
      "epoch": 0.08985230527320717,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0003995,
      "loss": 6.5703,
      "mean_token_accuracy": 0.12607999593019487,
      "num_tokens": 1792888.0,
      "step": 800
    },
    {
      "entropy": 6.695853853225708,
      "epoch": 0.09041388218116471,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000402,
      "loss": 6.5673,
      "mean_token_accuracy": 0.12844647020101546,
      "num_tokens": 1802938.0,
      "step": 805
    },
    {
      "entropy": 6.667530870437622,
      "epoch": 0.09097545908912226,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004045,
      "loss": 6.6049,
      "mean_token_accuracy": 0.12256239578127862,
      "num_tokens": 1814738.0,
      "step": 810
    },
    {
      "entropy": 6.709681272506714,
      "epoch": 0.0915370359970798,
      "grad_norm": 0.890625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.6675,
      "mean_token_accuracy": 0.12430930361151696,
      "num_tokens": 1825100.0,
      "step": 815
    },
    {
      "entropy": 6.590344858169556,
      "epoch": 0.09209861290503735,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004095,
      "loss": 6.5098,
      "mean_token_accuracy": 0.12951224520802498,
      "num_tokens": 1835879.0,
      "step": 820
    },
    {
      "entropy": 6.738299226760864,
      "epoch": 0.09266018981299488,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000412,
      "loss": 6.6605,
      "mean_token_accuracy": 0.11758637428283691,
      "num_tokens": 1848352.0,
      "step": 825
    },
    {
      "entropy": 6.652737951278686,
      "epoch": 0.09322176672095243,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004145,
      "loss": 6.6038,
      "mean_token_accuracy": 0.12358284816145897,
      "num_tokens": 1859687.0,
      "step": 830
    },
    {
      "entropy": 6.627905178070068,
      "epoch": 0.09378334362890998,
      "grad_norm": 0.921875,
      "learning_rate": 0.000417,
      "loss": 6.598,
      "mean_token_accuracy": 0.12505071982741356,
      "num_tokens": 1871438.0,
      "step": 835
    },
    {
      "entropy": 6.549618434906006,
      "epoch": 0.09434492053686752,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004195,
      "loss": 6.4806,
      "mean_token_accuracy": 0.13302377462387086,
      "num_tokens": 1881519.0,
      "step": 840
    },
    {
      "entropy": 6.733680200576782,
      "epoch": 0.09490649744482507,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000422,
      "loss": 6.6641,
      "mean_token_accuracy": 0.12029701620340347,
      "num_tokens": 1892428.0,
      "step": 845
    },
    {
      "entropy": 6.674782085418701,
      "epoch": 0.09546807435278261,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004245,
      "loss": 6.5505,
      "mean_token_accuracy": 0.13130439519882203,
      "num_tokens": 1903760.0,
      "step": 850
    },
    {
      "entropy": 6.632183504104614,
      "epoch": 0.09602965126074016,
      "grad_norm": 0.75390625,
      "learning_rate": 0.000427,
      "loss": 6.6123,
      "mean_token_accuracy": 0.11986933276057243,
      "num_tokens": 1915430.0,
      "step": 855
    },
    {
      "entropy": 6.6506672382354735,
      "epoch": 0.09659122816869771,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004295,
      "loss": 6.566,
      "mean_token_accuracy": 0.128663220256567,
      "num_tokens": 1925709.0,
      "step": 860
    },
    {
      "entropy": 6.659067535400391,
      "epoch": 0.09715280507665525,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000432,
      "loss": 6.6133,
      "mean_token_accuracy": 0.12502505853772164,
      "num_tokens": 1938148.0,
      "step": 865
    },
    {
      "entropy": 6.522532606124878,
      "epoch": 0.0977143819846128,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004345,
      "loss": 6.465,
      "mean_token_accuracy": 0.1305969774723053,
      "num_tokens": 1949028.0,
      "step": 870
    },
    {
      "entropy": 6.521027231216431,
      "epoch": 0.09827595889257033,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000437,
      "loss": 6.5332,
      "mean_token_accuracy": 0.12862171232700348,
      "num_tokens": 1960353.0,
      "step": 875
    },
    {
      "entropy": 6.733491516113281,
      "epoch": 0.09883753580052788,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004395,
      "loss": 6.6259,
      "mean_token_accuracy": 0.12553723603487016,
      "num_tokens": 1972335.0,
      "step": 880
    },
    {
      "entropy": 6.552764797210694,
      "epoch": 0.09939911270848542,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000442,
      "loss": 6.4797,
      "mean_token_accuracy": 0.12476090043783188,
      "num_tokens": 1982851.0,
      "step": 885
    },
    {
      "entropy": 6.5761137962341305,
      "epoch": 0.09996068961644297,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004445,
      "loss": 6.4659,
      "mean_token_accuracy": 0.13477788046002387,
      "num_tokens": 1993931.0,
      "step": 890
    },
    {
      "entropy": 6.574522161483765,
      "epoch": 0.10052226652440052,
      "grad_norm": 0.828125,
      "learning_rate": 0.000447,
      "loss": 6.5948,
      "mean_token_accuracy": 0.12630576267838478,
      "num_tokens": 2004838.0,
      "step": 895
    },
    {
      "entropy": 6.519236660003662,
      "epoch": 0.10108384343235806,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3958,
      "mean_token_accuracy": 0.13307079225778579,
      "num_tokens": 2015620.0,
      "step": 900
    },
    {
      "entropy": 6.607462596893311,
      "epoch": 0.10164542034031561,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5494,
      "mean_token_accuracy": 0.13406692892313005,
      "num_tokens": 2027489.0,
      "step": 905
    },
    {
      "entropy": 6.4195150375366214,
      "epoch": 0.10220699724827315,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.3915,
      "mean_token_accuracy": 0.13798633739352226,
      "num_tokens": 2038475.0,
      "step": 910
    },
    {
      "entropy": 6.49401707649231,
      "epoch": 0.1027685741562307,
      "grad_norm": 0.890625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.411,
      "mean_token_accuracy": 0.13152770772576333,
      "num_tokens": 2048938.0,
      "step": 915
    },
    {
      "entropy": 6.484944677352905,
      "epoch": 0.10333015106418825,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4981,
      "mean_token_accuracy": 0.12505212873220445,
      "num_tokens": 2060372.0,
      "step": 920
    },
    {
      "entropy": 6.572954750061035,
      "epoch": 0.10389172797214578,
      "grad_norm": 0.9375,
      "learning_rate": 0.000462,
      "loss": 6.5125,
      "mean_token_accuracy": 0.12695388346910477,
      "num_tokens": 2071558.0,
      "step": 925
    },
    {
      "entropy": 6.553452968597412,
      "epoch": 0.10445330488010333,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004645,
      "loss": 6.5288,
      "mean_token_accuracy": 0.12948471456766128,
      "num_tokens": 2082350.0,
      "step": 930
    },
    {
      "entropy": 6.6269059658050535,
      "epoch": 0.10501488178806087,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000467,
      "loss": 6.5882,
      "mean_token_accuracy": 0.12736724987626075,
      "num_tokens": 2093448.0,
      "step": 935
    },
    {
      "entropy": 6.718510961532592,
      "epoch": 0.10557645869601842,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004695,
      "loss": 6.6336,
      "mean_token_accuracy": 0.12538549676537514,
      "num_tokens": 2104171.0,
      "step": 940
    },
    {
      "entropy": 6.5074928283691404,
      "epoch": 0.10613803560397596,
      "grad_norm": 1.09375,
      "learning_rate": 0.000472,
      "loss": 6.5012,
      "mean_token_accuracy": 0.13048772215843202,
      "num_tokens": 2114496.0,
      "step": 945
    },
    {
      "entropy": 6.615377426147461,
      "epoch": 0.10669961251193351,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004745,
      "loss": 6.506,
      "mean_token_accuracy": 0.1323799431324005,
      "num_tokens": 2125930.0,
      "step": 950
    },
    {
      "entropy": 6.48550271987915,
      "epoch": 0.10726118941989106,
      "grad_norm": 1.015625,
      "learning_rate": 0.000477,
      "loss": 6.4768,
      "mean_token_accuracy": 0.13284032344818114,
      "num_tokens": 2136452.0,
      "step": 955
    },
    {
      "entropy": 6.5667078495025635,
      "epoch": 0.1078227663278486,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004795,
      "loss": 6.5651,
      "mean_token_accuracy": 0.12846669927239418,
      "num_tokens": 2148233.0,
      "step": 960
    },
    {
      "entropy": 6.492109537124634,
      "epoch": 0.10838434323580615,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000482,
      "loss": 6.3508,
      "mean_token_accuracy": 0.13522292152047158,
      "num_tokens": 2159475.0,
      "step": 965
    },
    {
      "entropy": 6.475024604797364,
      "epoch": 0.10894592014376368,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004845,
      "loss": 6.4622,
      "mean_token_accuracy": 0.12998763769865035,
      "num_tokens": 2170042.0,
      "step": 970
    },
    {
      "entropy": 6.565961837768555,
      "epoch": 0.10950749705172123,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000487,
      "loss": 6.4639,
      "mean_token_accuracy": 0.12638997063040733,
      "num_tokens": 2182234.0,
      "step": 975
    },
    {
      "entropy": 6.398307752609253,
      "epoch": 0.11006907395967878,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895,
      "loss": 6.4561,
      "mean_token_accuracy": 0.13463474065065384,
      "num_tokens": 2192477.0,
      "step": 980
    },
    {
      "entropy": 6.4558258056640625,
      "epoch": 0.11063065086763632,
      "grad_norm": 1.15625,
      "learning_rate": 0.000492,
      "loss": 6.4511,
      "mean_token_accuracy": 0.12646445855498314,
      "num_tokens": 2203082.0,
      "step": 985
    },
    {
      "entropy": 6.489946889877319,
      "epoch": 0.11119222777559387,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004945,
      "loss": 6.4149,
      "mean_token_accuracy": 0.13445554673671722,
      "num_tokens": 2213684.0,
      "step": 990
    },
    {
      "entropy": 6.453236436843872,
      "epoch": 0.11175380468355141,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000497,
      "loss": 6.4207,
      "mean_token_accuracy": 0.13301730901002884,
      "num_tokens": 2225524.0,
      "step": 995
    },
    {
      "entropy": 6.585108804702759,
      "epoch": 0.11231538159150896,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004995,
      "loss": 6.3966,
      "mean_token_accuracy": 0.1316801317036152,
      "num_tokens": 2236004.0,
      "step": 1000
    },
    {
      "entropy": 6.456496858596802,
      "epoch": 0.1128769584994665,
      "grad_norm": 0.875,
      "learning_rate": 0.000499999997707492,
      "loss": 6.5247,
      "mean_token_accuracy": 0.12453527525067329,
      "num_tokens": 2247730.0,
      "step": 1005
    },
    {
      "entropy": 6.560186100006104,
      "epoch": 0.11343853540742405,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999999883941787,
      "loss": 6.491,
      "mean_token_accuracy": 0.1362232081592083,
      "num_tokens": 2260132.0,
      "step": 1010
    },
    {
      "entropy": 6.490068674087524,
      "epoch": 0.1140001123153816,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999999719167785,
      "loss": 6.464,
      "mean_token_accuracy": 0.1341004975140095,
      "num_tokens": 2270060.0,
      "step": 1015
    },
    {
      "entropy": 6.507891464233398,
      "epoch": 0.11456168922333913,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999999482752917,
      "loss": 6.4733,
      "mean_token_accuracy": 0.12934465631842612,
      "num_tokens": 2281568.0,
      "step": 1020
    },
    {
      "entropy": 6.524950933456421,
      "epoch": 0.11512326613129668,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999999174697195,
      "loss": 6.4998,
      "mean_token_accuracy": 0.127571739256382,
      "num_tokens": 2293325.0,
      "step": 1025
    },
    {
      "entropy": 6.621273231506348,
      "epoch": 0.11568484303925422,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999998795000627,
      "loss": 6.5086,
      "mean_token_accuracy": 0.1368153616786003,
      "num_tokens": 2304780.0,
      "step": 1030
    },
    {
      "entropy": 6.402975273132324,
      "epoch": 0.11624641994721177,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999998343663223,
      "loss": 6.4343,
      "mean_token_accuracy": 0.1310793451964855,
      "num_tokens": 2315999.0,
      "step": 1035
    },
    {
      "entropy": 6.539117860794067,
      "epoch": 0.11680799685516932,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999997820685,
      "loss": 6.5788,
      "mean_token_accuracy": 0.131030685454607,
      "num_tokens": 2326695.0,
      "step": 1040
    },
    {
      "entropy": 6.555915832519531,
      "epoch": 0.11736957376312686,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999997226065974,
      "loss": 6.4067,
      "mean_token_accuracy": 0.14168997928500177,
      "num_tokens": 2336925.0,
      "step": 1045
    },
    {
      "entropy": 6.3779439449310305,
      "epoch": 0.11793115067108441,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999996559806163,
      "loss": 6.3148,
      "mean_token_accuracy": 0.13549636751413346,
      "num_tokens": 2347431.0,
      "step": 1050
    },
    {
      "entropy": 6.440852689743042,
      "epoch": 0.11849272757904195,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999995821905589,
      "loss": 6.4194,
      "mean_token_accuracy": 0.13899283483624458,
      "num_tokens": 2359894.0,
      "step": 1055
    },
    {
      "entropy": 6.419764804840088,
      "epoch": 0.1190543044869995,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999995012364276,
      "loss": 6.4644,
      "mean_token_accuracy": 0.1340063691139221,
      "num_tokens": 2371181.0,
      "step": 1060
    },
    {
      "entropy": 6.476871919631958,
      "epoch": 0.11961588139495703,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999994131182248,
      "loss": 6.3616,
      "mean_token_accuracy": 0.1311580441892147,
      "num_tokens": 2381531.0,
      "step": 1065
    },
    {
      "entropy": 6.585131025314331,
      "epoch": 0.12017745830291458,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999993178359535,
      "loss": 6.4729,
      "mean_token_accuracy": 0.13767909109592438,
      "num_tokens": 2391573.0,
      "step": 1070
    },
    {
      "entropy": 6.413916444778442,
      "epoch": 0.12073903521087213,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999992153896166,
      "loss": 6.446,
      "mean_token_accuracy": 0.13483774214982985,
      "num_tokens": 2402822.0,
      "step": 1075
    },
    {
      "entropy": 6.425882148742676,
      "epoch": 0.12130061211882967,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999991057792175,
      "loss": 6.3556,
      "mean_token_accuracy": 0.13548125475645065,
      "num_tokens": 2414322.0,
      "step": 1080
    },
    {
      "entropy": 6.4980425357818605,
      "epoch": 0.12186218902678722,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004999989890047596,
      "loss": 6.5464,
      "mean_token_accuracy": 0.1351483039557934,
      "num_tokens": 2426560.0,
      "step": 1085
    },
    {
      "entropy": 6.573987579345703,
      "epoch": 0.12242376593474476,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999988650662467,
      "loss": 6.4631,
      "mean_token_accuracy": 0.13326464518904685,
      "num_tokens": 2437723.0,
      "step": 1090
    },
    {
      "entropy": 6.402809953689575,
      "epoch": 0.12298534284270231,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999987339636825,
      "loss": 6.4139,
      "mean_token_accuracy": 0.13775320574641228,
      "num_tokens": 2448497.0,
      "step": 1095
    },
    {
      "entropy": 6.311347532272339,
      "epoch": 0.12354691975065986,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999985956970713,
      "loss": 6.2693,
      "mean_token_accuracy": 0.14299433752894403,
      "num_tokens": 2459089.0,
      "step": 1100
    },
    {
      "entropy": 6.469817876815796,
      "epoch": 0.1241084966586174,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999984502664178,
      "loss": 6.3627,
      "mean_token_accuracy": 0.14109985455870627,
      "num_tokens": 2469882.0,
      "step": 1105
    },
    {
      "entropy": 6.417138671875,
      "epoch": 0.12467007356657495,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999982976717261,
      "loss": 6.415,
      "mean_token_accuracy": 0.13988902568817138,
      "num_tokens": 2482120.0,
      "step": 1110
    },
    {
      "entropy": 6.481253623962402,
      "epoch": 0.1252316504745325,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999981379130016,
      "loss": 6.4102,
      "mean_token_accuracy": 0.13518037721514703,
      "num_tokens": 2492626.0,
      "step": 1115
    },
    {
      "entropy": 6.59227557182312,
      "epoch": 0.12579322738249002,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499997970990249,
      "loss": 6.4646,
      "mean_token_accuracy": 0.1327376738190651,
      "num_tokens": 2506347.0,
      "step": 1120
    },
    {
      "entropy": 6.266043615341187,
      "epoch": 0.12635480429044757,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999977969034737,
      "loss": 6.3464,
      "mean_token_accuracy": 0.13951586112380027,
      "num_tokens": 2516584.0,
      "step": 1125
    },
    {
      "entropy": 6.528832149505615,
      "epoch": 0.12691638119840512,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999976156526814,
      "loss": 6.3984,
      "mean_token_accuracy": 0.13773541748523713,
      "num_tokens": 2526731.0,
      "step": 1130
    },
    {
      "entropy": 6.33425726890564,
      "epoch": 0.12747795810636267,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999974272378776,
      "loss": 6.2973,
      "mean_token_accuracy": 0.1425212301313877,
      "num_tokens": 2537180.0,
      "step": 1135
    },
    {
      "entropy": 6.518994569778442,
      "epoch": 0.12803953501432022,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999972316590685,
      "loss": 6.3819,
      "mean_token_accuracy": 0.13330503478646277,
      "num_tokens": 2547617.0,
      "step": 1140
    },
    {
      "entropy": 6.3555159091949465,
      "epoch": 0.12860111192227774,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999970289162604,
      "loss": 6.2686,
      "mean_token_accuracy": 0.1448584720492363,
      "num_tokens": 2557743.0,
      "step": 1145
    },
    {
      "entropy": 6.3979815483093265,
      "epoch": 0.1291626888302353,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999968190094595,
      "loss": 6.3601,
      "mean_token_accuracy": 0.13572428226470948,
      "num_tokens": 2568351.0,
      "step": 1150
    },
    {
      "entropy": 6.344897937774658,
      "epoch": 0.12972426573819285,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999966019386727,
      "loss": 6.4169,
      "mean_token_accuracy": 0.14001458808779715,
      "num_tokens": 2579703.0,
      "step": 1155
    },
    {
      "entropy": 6.554988098144531,
      "epoch": 0.1302858426461504,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999963777039067,
      "loss": 6.3852,
      "mean_token_accuracy": 0.13615019991993904,
      "num_tokens": 2591197.0,
      "step": 1160
    },
    {
      "entropy": 6.3376793384552,
      "epoch": 0.13084741955410795,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499996146305169,
      "loss": 6.3748,
      "mean_token_accuracy": 0.13186048045754434,
      "num_tokens": 2601836.0,
      "step": 1165
    },
    {
      "entropy": 6.344990921020508,
      "epoch": 0.13140899646206547,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999959077424665,
      "loss": 6.3009,
      "mean_token_accuracy": 0.1399838626384735,
      "num_tokens": 2613513.0,
      "step": 1170
    },
    {
      "entropy": 6.420182085037231,
      "epoch": 0.13197057337002302,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999956620158072,
      "loss": 6.3476,
      "mean_token_accuracy": 0.13971240893006326,
      "num_tokens": 2625246.0,
      "step": 1175
    },
    {
      "entropy": 6.348382759094238,
      "epoch": 0.13253215027798057,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004999954091251987,
      "loss": 6.3225,
      "mean_token_accuracy": 0.13963563665747641,
      "num_tokens": 2636375.0,
      "step": 1180
    },
    {
      "entropy": 6.378366088867187,
      "epoch": 0.13309372718593812,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999951490706491,
      "loss": 6.4077,
      "mean_token_accuracy": 0.13064486235380174,
      "num_tokens": 2648586.0,
      "step": 1185
    },
    {
      "entropy": 6.471061944961548,
      "epoch": 0.13365530409389567,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999948818521667,
      "loss": 6.4102,
      "mean_token_accuracy": 0.142090106010437,
      "num_tokens": 2660847.0,
      "step": 1190
    },
    {
      "entropy": 6.4110862731933596,
      "epoch": 0.1342168810018532,
      "grad_norm": 0.828125,
      "learning_rate": 0.00049999460746976,
      "loss": 6.3302,
      "mean_token_accuracy": 0.14113106280565263,
      "num_tokens": 2671160.0,
      "step": 1195
    },
    {
      "entropy": 6.469895267486573,
      "epoch": 0.13477845790981074,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999943259234377,
      "loss": 6.4551,
      "mean_token_accuracy": 0.1292402721941471,
      "num_tokens": 2681622.0,
      "step": 1200
    },
    {
      "entropy": 6.354708957672119,
      "epoch": 0.1353400348177683,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999940372132089,
      "loss": 6.2578,
      "mean_token_accuracy": 0.14201848655939103,
      "num_tokens": 2691503.0,
      "step": 1205
    },
    {
      "entropy": 6.4176633834838865,
      "epoch": 0.13590161172572585,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999937413390826,
      "loss": 6.3486,
      "mean_token_accuracy": 0.13264842703938484,
      "num_tokens": 2702207.0,
      "step": 1210
    },
    {
      "entropy": 6.377297592163086,
      "epoch": 0.1364631886336834,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999934383010683,
      "loss": 6.3505,
      "mean_token_accuracy": 0.13756700977683067,
      "num_tokens": 2713178.0,
      "step": 1215
    },
    {
      "entropy": 6.4666331768035885,
      "epoch": 0.13702476554164092,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999931280991757,
      "loss": 6.3452,
      "mean_token_accuracy": 0.14598208293318748,
      "num_tokens": 2723717.0,
      "step": 1220
    },
    {
      "entropy": 6.428584241867066,
      "epoch": 0.13758634244959847,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999928107334146,
      "loss": 6.4134,
      "mean_token_accuracy": 0.1348503641784191,
      "num_tokens": 2735669.0,
      "step": 1225
    },
    {
      "entropy": 6.266319847106933,
      "epoch": 0.13814791935755602,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999924862037952,
      "loss": 6.3068,
      "mean_token_accuracy": 0.1396687351167202,
      "num_tokens": 2746087.0,
      "step": 1230
    },
    {
      "entropy": 6.460708808898926,
      "epoch": 0.13870949626551357,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999921545103278,
      "loss": 6.3242,
      "mean_token_accuracy": 0.13702112659811974,
      "num_tokens": 2757143.0,
      "step": 1235
    },
    {
      "entropy": 6.414711141586304,
      "epoch": 0.1392710731734711,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999918156530229,
      "loss": 6.4365,
      "mean_token_accuracy": 0.13701202794909478,
      "num_tokens": 2769340.0,
      "step": 1240
    },
    {
      "entropy": 6.432314443588257,
      "epoch": 0.13983265008142864,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999914696318913,
      "loss": 6.3401,
      "mean_token_accuracy": 0.13881635367870332,
      "num_tokens": 2779514.0,
      "step": 1245
    },
    {
      "entropy": 6.358312511444092,
      "epoch": 0.1403942269893862,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999911164469441,
      "loss": 6.2974,
      "mean_token_accuracy": 0.14529771953821183,
      "num_tokens": 2790801.0,
      "step": 1250
    },
    {
      "entropy": 6.419577169418335,
      "epoch": 0.14095580389734375,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999907560981925,
      "loss": 6.2782,
      "mean_token_accuracy": 0.14040456488728523,
      "num_tokens": 2801082.0,
      "step": 1255
    },
    {
      "entropy": 6.296146583557129,
      "epoch": 0.1415173808053013,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999903885856479,
      "loss": 6.2855,
      "mean_token_accuracy": 0.14076389670372008,
      "num_tokens": 2812719.0,
      "step": 1260
    },
    {
      "entropy": 6.348859882354736,
      "epoch": 0.14207895771325882,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499990013909322,
      "loss": 6.2849,
      "mean_token_accuracy": 0.1368976838886738,
      "num_tokens": 2823766.0,
      "step": 1265
    },
    {
      "entropy": 6.348705673217774,
      "epoch": 0.14264053462121637,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499989632069227,
      "loss": 6.3652,
      "mean_token_accuracy": 0.13619414418935777,
      "num_tokens": 2834801.0,
      "step": 1270
    },
    {
      "entropy": 6.347746229171753,
      "epoch": 0.14320211152917392,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999892430653747,
      "loss": 6.2028,
      "mean_token_accuracy": 0.14546942338347435,
      "num_tokens": 2845769.0,
      "step": 1275
    },
    {
      "entropy": 6.289702272415161,
      "epoch": 0.14376368843713147,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999888468977776,
      "loss": 6.2401,
      "mean_token_accuracy": 0.14375891238451005,
      "num_tokens": 2856001.0,
      "step": 1280
    },
    {
      "entropy": 6.35533881187439,
      "epoch": 0.14432526534508902,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999884435664485,
      "loss": 6.2663,
      "mean_token_accuracy": 0.14633105844259262,
      "num_tokens": 2867703.0,
      "step": 1285
    },
    {
      "entropy": 6.233186101913452,
      "epoch": 0.14488684225304654,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999880330714,
      "loss": 6.234,
      "mean_token_accuracy": 0.14503639936447144,
      "num_tokens": 2878671.0,
      "step": 1290
    },
    {
      "entropy": 6.362540769577026,
      "epoch": 0.1454484191610041,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999876154126453,
      "loss": 6.3403,
      "mean_token_accuracy": 0.13664028346538543,
      "num_tokens": 2890137.0,
      "step": 1295
    },
    {
      "entropy": 6.306020212173462,
      "epoch": 0.14600999606896164,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999871905901977,
      "loss": 6.2357,
      "mean_token_accuracy": 0.13921404853463173,
      "num_tokens": 2900344.0,
      "step": 1300
    },
    {
      "entropy": 6.284024572372436,
      "epoch": 0.1465715729769192,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999867586040706,
      "loss": 6.2557,
      "mean_token_accuracy": 0.14088367968797683,
      "num_tokens": 2912270.0,
      "step": 1305
    },
    {
      "entropy": 6.376524925231934,
      "epoch": 0.14713314988487675,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999863194542779,
      "loss": 6.3348,
      "mean_token_accuracy": 0.13199515640735626,
      "num_tokens": 2924677.0,
      "step": 1310
    },
    {
      "entropy": 6.2676990032196045,
      "epoch": 0.14769472679283427,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999858731408336,
      "loss": 6.2504,
      "mean_token_accuracy": 0.1430131196975708,
      "num_tokens": 2935794.0,
      "step": 1315
    },
    {
      "entropy": 6.3416272640228275,
      "epoch": 0.14825630370079182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999854196637518,
      "loss": 6.3627,
      "mean_token_accuracy": 0.13728150352835655,
      "num_tokens": 2945958.0,
      "step": 1320
    },
    {
      "entropy": 6.376453304290772,
      "epoch": 0.14881788060874937,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999849590230469,
      "loss": 6.3296,
      "mean_token_accuracy": 0.13831673040986062,
      "num_tokens": 2957536.0,
      "step": 1325
    },
    {
      "entropy": 6.417525577545166,
      "epoch": 0.14937945751670692,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999844912187336,
      "loss": 6.313,
      "mean_token_accuracy": 0.13861921206116676,
      "num_tokens": 2967787.0,
      "step": 1330
    },
    {
      "entropy": 6.407558107376099,
      "epoch": 0.14994103442466447,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499984016250827,
      "loss": 6.3224,
      "mean_token_accuracy": 0.13716086149215698,
      "num_tokens": 2980522.0,
      "step": 1335
    },
    {
      "entropy": 6.188078784942627,
      "epoch": 0.150502611332622,
      "grad_norm": 1.0,
      "learning_rate": 0.000499983534119342,
      "loss": 6.2513,
      "mean_token_accuracy": 0.13945341408252715,
      "num_tokens": 2990497.0,
      "step": 1340
    },
    {
      "entropy": 6.482673597335816,
      "epoch": 0.15106418824057954,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999830448242941,
      "loss": 6.353,
      "mean_token_accuracy": 0.14002300277352334,
      "num_tokens": 3002931.0,
      "step": 1345
    },
    {
      "entropy": 6.254575872421265,
      "epoch": 0.1516257651485371,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999825483656986,
      "loss": 6.2748,
      "mean_token_accuracy": 0.14595276713371277,
      "num_tokens": 3014137.0,
      "step": 1350
    },
    {
      "entropy": 6.553881549835205,
      "epoch": 0.15218734205649465,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999820447435716,
      "loss": 6.4367,
      "mean_token_accuracy": 0.13388766795396806,
      "num_tokens": 3026145.0,
      "step": 1355
    },
    {
      "entropy": 6.27535490989685,
      "epoch": 0.15274891896445217,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999815339579291,
      "loss": 6.2853,
      "mean_token_accuracy": 0.13878872245550156,
      "num_tokens": 3037068.0,
      "step": 1360
    },
    {
      "entropy": 6.293557262420654,
      "epoch": 0.15331049587240972,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999810160087872,
      "loss": 6.2339,
      "mean_token_accuracy": 0.14105164185166358,
      "num_tokens": 3047836.0,
      "step": 1365
    },
    {
      "entropy": 6.372267818450927,
      "epoch": 0.15387207278036727,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004999804908961626,
      "loss": 6.3584,
      "mean_token_accuracy": 0.1421847365796566,
      "num_tokens": 3060309.0,
      "step": 1370
    },
    {
      "entropy": 6.353737211227417,
      "epoch": 0.15443364968832482,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999799586200719,
      "loss": 6.3003,
      "mean_token_accuracy": 0.1419496037065983,
      "num_tokens": 3071789.0,
      "step": 1375
    },
    {
      "entropy": 6.360533714294434,
      "epoch": 0.15499522659628237,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499979419180532,
      "loss": 6.2236,
      "mean_token_accuracy": 0.1449585184454918,
      "num_tokens": 3081756.0,
      "step": 1380
    },
    {
      "entropy": 6.235655450820923,
      "epoch": 0.1555568035042399,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999788725775601,
      "loss": 6.3434,
      "mean_token_accuracy": 0.13456685617566108,
      "num_tokens": 3093456.0,
      "step": 1385
    },
    {
      "entropy": 6.440259408950806,
      "epoch": 0.15611838041219744,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999783188111737,
      "loss": 6.265,
      "mean_token_accuracy": 0.13899147212505342,
      "num_tokens": 3104565.0,
      "step": 1390
    },
    {
      "entropy": 6.34439787864685,
      "epoch": 0.156679957320155,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999777578813904,
      "loss": 6.2861,
      "mean_token_accuracy": 0.14071865901350974,
      "num_tokens": 3115577.0,
      "step": 1395
    },
    {
      "entropy": 6.224101924896241,
      "epoch": 0.15724153422811254,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499977189788228,
      "loss": 6.1836,
      "mean_token_accuracy": 0.14295614287257194,
      "num_tokens": 3126019.0,
      "step": 1400
    },
    {
      "entropy": 6.388071966171265,
      "epoch": 0.1578031111360701,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999766145317046,
      "loss": 6.3394,
      "mean_token_accuracy": 0.13431284055113793,
      "num_tokens": 3136191.0,
      "step": 1405
    },
    {
      "entropy": 6.397411298751831,
      "epoch": 0.15836468804402762,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999760321118385,
      "loss": 6.2113,
      "mean_token_accuracy": 0.143301859498024,
      "num_tokens": 3147737.0,
      "step": 1410
    },
    {
      "entropy": 6.158065271377564,
      "epoch": 0.15892626495198517,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999754425286483,
      "loss": 6.2684,
      "mean_token_accuracy": 0.13974310755729674,
      "num_tokens": 3158494.0,
      "step": 1415
    },
    {
      "entropy": 6.432788991928101,
      "epoch": 0.15948784185994272,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999748457821529,
      "loss": 6.315,
      "mean_token_accuracy": 0.14211769849061967,
      "num_tokens": 3170550.0,
      "step": 1420
    },
    {
      "entropy": 6.21512804031372,
      "epoch": 0.16004941876790027,
      "grad_norm": 0.8125,
      "learning_rate": 0.000499974241872371,
      "loss": 6.2877,
      "mean_token_accuracy": 0.14602836072444916,
      "num_tokens": 3182860.0,
      "step": 1425
    },
    {
      "entropy": 6.338991689682007,
      "epoch": 0.16061099567585782,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999736307993221,
      "loss": 6.2843,
      "mean_token_accuracy": 0.13677069991827012,
      "num_tokens": 3193807.0,
      "step": 1430
    },
    {
      "entropy": 6.18875823020935,
      "epoch": 0.16117257258381534,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999730125630255,
      "loss": 6.1569,
      "mean_token_accuracy": 0.13699881434440614,
      "num_tokens": 3203959.0,
      "step": 1435
    },
    {
      "entropy": 6.351223182678223,
      "epoch": 0.1617341494917729,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999723871635008,
      "loss": 6.1745,
      "mean_token_accuracy": 0.14297635555267335,
      "num_tokens": 3215416.0,
      "step": 1440
    },
    {
      "entropy": 6.344785833358765,
      "epoch": 0.16229572639973044,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999717546007681,
      "loss": 6.3179,
      "mean_token_accuracy": 0.13555386140942574,
      "num_tokens": 3226938.0,
      "step": 1445
    },
    {
      "entropy": 6.229445600509644,
      "epoch": 0.162857303307688,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999711148748476,
      "loss": 6.1367,
      "mean_token_accuracy": 0.14848765432834626,
      "num_tokens": 3238406.0,
      "step": 1450
    },
    {
      "entropy": 6.131477737426758,
      "epoch": 0.16341888021564555,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999704679857594,
      "loss": 6.1427,
      "mean_token_accuracy": 0.14189836829900743,
      "num_tokens": 3249695.0,
      "step": 1455
    },
    {
      "entropy": 6.446213865280152,
      "epoch": 0.16398045712360307,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999698139335243,
      "loss": 6.3058,
      "mean_token_accuracy": 0.1473018154501915,
      "num_tokens": 3259656.0,
      "step": 1460
    },
    {
      "entropy": 6.181229066848755,
      "epoch": 0.16454203403156062,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999691527181632,
      "loss": 6.1311,
      "mean_token_accuracy": 0.15136088579893112,
      "num_tokens": 3269712.0,
      "step": 1465
    },
    {
      "entropy": 6.132031583786011,
      "epoch": 0.16510361093951817,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999684843396968,
      "loss": 6.1254,
      "mean_token_accuracy": 0.1496811881661415,
      "num_tokens": 3280343.0,
      "step": 1470
    },
    {
      "entropy": 6.244436836242675,
      "epoch": 0.16566518784747572,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999678087981467,
      "loss": 6.1549,
      "mean_token_accuracy": 0.14772350192070008,
      "num_tokens": 3291649.0,
      "step": 1475
    },
    {
      "entropy": 6.332507038116455,
      "epoch": 0.16622676475543324,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004999671260935343,
      "loss": 6.3432,
      "mean_token_accuracy": 0.14269206747412683,
      "num_tokens": 3304190.0,
      "step": 1480
    },
    {
      "entropy": 6.224244928359985,
      "epoch": 0.1667883416633908,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999664362258814,
      "loss": 6.0663,
      "mean_token_accuracy": 0.1542764872312546,
      "num_tokens": 3313966.0,
      "step": 1485
    },
    {
      "entropy": 6.275435161590576,
      "epoch": 0.16734991857134834,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999657391952099,
      "loss": 6.3114,
      "mean_token_accuracy": 0.14152566939592362,
      "num_tokens": 3326017.0,
      "step": 1490
    },
    {
      "entropy": 6.349188041687012,
      "epoch": 0.1679114954793059,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499965035001542,
      "loss": 6.2104,
      "mean_token_accuracy": 0.13980446606874466,
      "num_tokens": 3338264.0,
      "step": 1495
    },
    {
      "entropy": 6.26988787651062,
      "epoch": 0.16847307238726345,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999643236449001,
      "loss": 6.2569,
      "mean_token_accuracy": 0.15189649015665055,
      "num_tokens": 3348901.0,
      "step": 1500
    },
    {
      "entropy": 6.249851369857788,
      "epoch": 0.16903464929522097,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999636051253069,
      "loss": 6.1784,
      "mean_token_accuracy": 0.14386393129825592,
      "num_tokens": 3358441.0,
      "step": 1505
    },
    {
      "entropy": 6.336975526809693,
      "epoch": 0.16959622620317852,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999628794427854,
      "loss": 6.2653,
      "mean_token_accuracy": 0.1504651814699173,
      "num_tokens": 3369334.0,
      "step": 1510
    },
    {
      "entropy": 6.2502185821533205,
      "epoch": 0.17015780311113607,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999621465973583,
      "loss": 6.2748,
      "mean_token_accuracy": 0.1409212589263916,
      "num_tokens": 3380698.0,
      "step": 1515
    },
    {
      "entropy": 6.462999057769776,
      "epoch": 0.17071938001909362,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999614065890495,
      "loss": 6.288,
      "mean_token_accuracy": 0.1381421558558941,
      "num_tokens": 3391061.0,
      "step": 1520
    },
    {
      "entropy": 6.246134901046753,
      "epoch": 0.17128095692705117,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999606594178821,
      "loss": 6.3371,
      "mean_token_accuracy": 0.13986739218235017,
      "num_tokens": 3403983.0,
      "step": 1525
    },
    {
      "entropy": 6.3834983825683596,
      "epoch": 0.1718425338350087,
      "grad_norm": 0.984375,
      "learning_rate": 0.00049995990508388,
      "loss": 6.2555,
      "mean_token_accuracy": 0.1448330372571945,
      "num_tokens": 3413838.0,
      "step": 1530
    },
    {
      "entropy": 6.145442771911621,
      "epoch": 0.17240411074296624,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999591435870674,
      "loss": 6.0926,
      "mean_token_accuracy": 0.14555919021368027,
      "num_tokens": 3423978.0,
      "step": 1535
    },
    {
      "entropy": 6.297583770751953,
      "epoch": 0.1729656876509238,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999583749274684,
      "loss": 6.2691,
      "mean_token_accuracy": 0.1472916215658188,
      "num_tokens": 3433416.0,
      "step": 1540
    },
    {
      "entropy": 6.213202953338623,
      "epoch": 0.17352726455888134,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004999575991051074,
      "loss": 6.1183,
      "mean_token_accuracy": 0.1482724741101265,
      "num_tokens": 3444983.0,
      "step": 1545
    },
    {
      "entropy": 6.228994274139405,
      "epoch": 0.1740888414668389,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999568161200093,
      "loss": 6.0923,
      "mean_token_accuracy": 0.1479077994823456,
      "num_tokens": 3455808.0,
      "step": 1550
    },
    {
      "entropy": 6.121140384674073,
      "epoch": 0.17465041837479642,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999560259721989,
      "loss": 6.2406,
      "mean_token_accuracy": 0.14061250612139703,
      "num_tokens": 3467503.0,
      "step": 1555
    },
    {
      "entropy": 6.5244382381439205,
      "epoch": 0.17521199528275397,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999552286617014,
      "loss": 6.3998,
      "mean_token_accuracy": 0.14157005473971368,
      "num_tokens": 3479941.0,
      "step": 1560
    },
    {
      "entropy": 6.189446020126343,
      "epoch": 0.17577357219071152,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999544241885421,
      "loss": 6.1028,
      "mean_token_accuracy": 0.15490597486495972,
      "num_tokens": 3491069.0,
      "step": 1565
    },
    {
      "entropy": 6.284439182281494,
      "epoch": 0.17633514909866907,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999536125527467,
      "loss": 6.3132,
      "mean_token_accuracy": 0.13803954273462296,
      "num_tokens": 3502065.0,
      "step": 1570
    },
    {
      "entropy": 6.291994380950928,
      "epoch": 0.17689672600662662,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499952793754341,
      "loss": 6.2551,
      "mean_token_accuracy": 0.14772795587778093,
      "num_tokens": 3514228.0,
      "step": 1575
    },
    {
      "entropy": 6.30311017036438,
      "epoch": 0.17745830291458414,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999519677933512,
      "loss": 6.1707,
      "mean_token_accuracy": 0.14605900645256042,
      "num_tokens": 3525801.0,
      "step": 1580
    },
    {
      "entropy": 6.187708377838135,
      "epoch": 0.1780198798225417,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999511346698035,
      "loss": 6.1216,
      "mean_token_accuracy": 0.1448603853583336,
      "num_tokens": 3536431.0,
      "step": 1585
    },
    {
      "entropy": 6.21040096282959,
      "epoch": 0.17858145673049924,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004999502943837244,
      "loss": 6.2162,
      "mean_token_accuracy": 0.14277410134673119,
      "num_tokens": 3549098.0,
      "step": 1590
    },
    {
      "entropy": 6.308737564086914,
      "epoch": 0.1791430336384568,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999494469351406,
      "loss": 6.187,
      "mean_token_accuracy": 0.1415993995964527,
      "num_tokens": 3560532.0,
      "step": 1595
    },
    {
      "entropy": 6.151534795761108,
      "epoch": 0.17970461054641435,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999485923240793,
      "loss": 6.1807,
      "mean_token_accuracy": 0.14444077014923096,
      "num_tokens": 3572175.0,
      "step": 1600
    },
    {
      "entropy": 6.327757406234741,
      "epoch": 0.18026618745437187,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999477305505675,
      "loss": 6.1126,
      "mean_token_accuracy": 0.15372487977147103,
      "num_tokens": 3582862.0,
      "step": 1605
    },
    {
      "entropy": 6.3059488296508786,
      "epoch": 0.18082776436232942,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999468616146328,
      "loss": 6.2841,
      "mean_token_accuracy": 0.14689953327178956,
      "num_tokens": 3594685.0,
      "step": 1610
    },
    {
      "entropy": 6.120879793167115,
      "epoch": 0.18138934127028697,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999459855163026,
      "loss": 6.0994,
      "mean_token_accuracy": 0.152867791056633,
      "num_tokens": 3604847.0,
      "step": 1615
    },
    {
      "entropy": 6.267522573471069,
      "epoch": 0.18195091817824452,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499945102255605,
      "loss": 6.1771,
      "mean_token_accuracy": 0.14639166221022606,
      "num_tokens": 3615549.0,
      "step": 1620
    },
    {
      "entropy": 6.195144271850586,
      "epoch": 0.18251249508620204,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999442118325682,
      "loss": 6.0792,
      "mean_token_accuracy": 0.1583167091012001,
      "num_tokens": 3626300.0,
      "step": 1625
    },
    {
      "entropy": 6.165605783462524,
      "epoch": 0.1830740719941596,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999433142472205,
      "loss": 6.1142,
      "mean_token_accuracy": 0.1476339131593704,
      "num_tokens": 3637832.0,
      "step": 1630
    },
    {
      "entropy": 6.1696093559265135,
      "epoch": 0.18363564890211714,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999424094995902,
      "loss": 6.1569,
      "mean_token_accuracy": 0.1420426405966282,
      "num_tokens": 3649179.0,
      "step": 1635
    },
    {
      "entropy": 6.245587396621704,
      "epoch": 0.1841972258100747,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999414975897066,
      "loss": 6.198,
      "mean_token_accuracy": 0.1402585081756115,
      "num_tokens": 3660011.0,
      "step": 1640
    },
    {
      "entropy": 6.302711725234985,
      "epoch": 0.18475880271803224,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999405785175983,
      "loss": 6.2802,
      "mean_token_accuracy": 0.14278458282351494,
      "num_tokens": 3672791.0,
      "step": 1645
    },
    {
      "entropy": 6.166055631637573,
      "epoch": 0.18532037962598977,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999396522832949,
      "loss": 6.0875,
      "mean_token_accuracy": 0.1390815943479538,
      "num_tokens": 3684709.0,
      "step": 1650
    },
    {
      "entropy": 6.152737283706665,
      "epoch": 0.18588195653394732,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999387188868256,
      "loss": 6.0627,
      "mean_token_accuracy": 0.15113601088523865,
      "num_tokens": 3694675.0,
      "step": 1655
    },
    {
      "entropy": 6.079391670227051,
      "epoch": 0.18644353344190487,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999377783282203,
      "loss": 6.0739,
      "mean_token_accuracy": 0.14676920920610428,
      "num_tokens": 3706384.0,
      "step": 1660
    },
    {
      "entropy": 6.2547712326049805,
      "epoch": 0.18700511034986242,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999368306075089,
      "loss": 6.1527,
      "mean_token_accuracy": 0.14625280648469924,
      "num_tokens": 3717419.0,
      "step": 1665
    },
    {
      "entropy": 6.255711317062378,
      "epoch": 0.18756668725781997,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999358757247215,
      "loss": 6.1702,
      "mean_token_accuracy": 0.1443650782108307,
      "num_tokens": 3727306.0,
      "step": 1670
    },
    {
      "entropy": 6.278718280792236,
      "epoch": 0.1881282641657775,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999349136798886,
      "loss": 6.2849,
      "mean_token_accuracy": 0.14049319103360175,
      "num_tokens": 3739476.0,
      "step": 1675
    },
    {
      "entropy": 6.170718765258789,
      "epoch": 0.18868984107373504,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999339444730409,
      "loss": 6.0598,
      "mean_token_accuracy": 0.15319431126117705,
      "num_tokens": 3751035.0,
      "step": 1680
    },
    {
      "entropy": 6.1924333572387695,
      "epoch": 0.1892514179816926,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999329681042092,
      "loss": 6.1581,
      "mean_token_accuracy": 0.1471312403678894,
      "num_tokens": 3762881.0,
      "step": 1685
    },
    {
      "entropy": 6.224253702163696,
      "epoch": 0.18981299488965014,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999319845734243,
      "loss": 6.1835,
      "mean_token_accuracy": 0.14361522495746612,
      "num_tokens": 3774180.0,
      "step": 1690
    },
    {
      "entropy": 6.230266857147217,
      "epoch": 0.1903745717976077,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004999309938807181,
      "loss": 6.1492,
      "mean_token_accuracy": 0.1505300983786583,
      "num_tokens": 3784864.0,
      "step": 1695
    },
    {
      "entropy": 6.259658145904541,
      "epoch": 0.19093614870556522,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999299960261216,
      "loss": 6.1322,
      "mean_token_accuracy": 0.14401278346776963,
      "num_tokens": 3797693.0,
      "step": 1700
    },
    {
      "entropy": 6.167710828781128,
      "epoch": 0.19149772561352277,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999289910096669,
      "loss": 6.1574,
      "mean_token_accuracy": 0.14654720053076745,
      "num_tokens": 3809379.0,
      "step": 1705
    },
    {
      "entropy": 6.081889629364014,
      "epoch": 0.19205930252148032,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999279788313858,
      "loss": 5.999,
      "mean_token_accuracy": 0.15570705085992814,
      "num_tokens": 3819473.0,
      "step": 1710
    },
    {
      "entropy": 6.077426338195801,
      "epoch": 0.19262087942943787,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999269594913107,
      "loss": 5.9932,
      "mean_token_accuracy": 0.15049063116312028,
      "num_tokens": 3831037.0,
      "step": 1715
    },
    {
      "entropy": 6.112188291549683,
      "epoch": 0.19318245633739542,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999259329894739,
      "loss": 6.1417,
      "mean_token_accuracy": 0.148158148676157,
      "num_tokens": 3841767.0,
      "step": 1720
    },
    {
      "entropy": 6.089222717285156,
      "epoch": 0.19374403324535294,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999248993259081,
      "loss": 5.9764,
      "mean_token_accuracy": 0.1548941805958748,
      "num_tokens": 3851629.0,
      "step": 1725
    },
    {
      "entropy": 6.0848329067230225,
      "epoch": 0.1943056101533105,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999238585006463,
      "loss": 6.0616,
      "mean_token_accuracy": 0.14957866221666336,
      "num_tokens": 3863247.0,
      "step": 1730
    },
    {
      "entropy": 6.147818183898925,
      "epoch": 0.19486718706126804,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999228105137218,
      "loss": 6.0335,
      "mean_token_accuracy": 0.1490088775753975,
      "num_tokens": 3873949.0,
      "step": 1735
    },
    {
      "entropy": 6.188397979736328,
      "epoch": 0.1954287639692256,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999217553651676,
      "loss": 6.1385,
      "mean_token_accuracy": 0.15051164627075195,
      "num_tokens": 3884992.0,
      "step": 1740
    },
    {
      "entropy": 6.246903657913208,
      "epoch": 0.19599034087718312,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999206930550175,
      "loss": 6.2617,
      "mean_token_accuracy": 0.15150685235857964,
      "num_tokens": 3897029.0,
      "step": 1745
    },
    {
      "entropy": 6.131383800506592,
      "epoch": 0.19655191778514067,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999196235833054,
      "loss": 6.105,
      "mean_token_accuracy": 0.14395989701151848,
      "num_tokens": 3909510.0,
      "step": 1750
    },
    {
      "entropy": 6.214809513092041,
      "epoch": 0.19711349469309822,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999185469500651,
      "loss": 6.0185,
      "mean_token_accuracy": 0.1482173793017864,
      "num_tokens": 3920593.0,
      "step": 1755
    },
    {
      "entropy": 6.142779541015625,
      "epoch": 0.19767507160105577,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000499917463155331,
      "loss": 6.1469,
      "mean_token_accuracy": 0.14746149256825447,
      "num_tokens": 3932215.0,
      "step": 1760
    },
    {
      "entropy": 6.216650819778442,
      "epoch": 0.19823664850901332,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999163721991378,
      "loss": 6.1024,
      "mean_token_accuracy": 0.14692942351102828,
      "num_tokens": 3943876.0,
      "step": 1765
    },
    {
      "entropy": 6.062715196609497,
      "epoch": 0.19879822541697084,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049991527408152,
      "loss": 6.0895,
      "mean_token_accuracy": 0.15319886654615403,
      "num_tokens": 3954533.0,
      "step": 1770
    },
    {
      "entropy": 6.210793972015381,
      "epoch": 0.1993598023249284,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999141688025127,
      "loss": 6.148,
      "mean_token_accuracy": 0.15144960433244706,
      "num_tokens": 3965722.0,
      "step": 1775
    },
    {
      "entropy": 6.174288940429688,
      "epoch": 0.19992137923288594,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499913056362151,
      "loss": 6.088,
      "mean_token_accuracy": 0.14908579289913176,
      "num_tokens": 3976737.0,
      "step": 1780
    },
    {
      "entropy": 6.225276279449463,
      "epoch": 0.2004829561408435,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999119367604703,
      "loss": 6.1585,
      "mean_token_accuracy": 0.1512651652097702,
      "num_tokens": 3986102.0,
      "step": 1785
    },
    {
      "entropy": 6.183860206604004,
      "epoch": 0.20104453304880104,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999108099975064,
      "loss": 6.1194,
      "mean_token_accuracy": 0.15166645050048827,
      "num_tokens": 3996534.0,
      "step": 1790
    },
    {
      "entropy": 6.145701837539673,
      "epoch": 0.20160610995675857,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499909676073295,
      "loss": 6.0466,
      "mean_token_accuracy": 0.14939896166324615,
      "num_tokens": 4006947.0,
      "step": 1795
    },
    {
      "entropy": 6.029658460617066,
      "epoch": 0.20216768686471612,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999085349878723,
      "loss": 6.0612,
      "mean_token_accuracy": 0.1588248908519745,
      "num_tokens": 4017723.0,
      "step": 1800
    },
    {
      "entropy": 6.135398435592651,
      "epoch": 0.20272926377267367,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999073867412746,
      "loss": 6.1535,
      "mean_token_accuracy": 0.14331067353487015,
      "num_tokens": 4028347.0,
      "step": 1805
    },
    {
      "entropy": 6.15322699546814,
      "epoch": 0.20329084068063122,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999062313335385,
      "loss": 6.0206,
      "mean_token_accuracy": 0.1573499321937561,
      "num_tokens": 4039648.0,
      "step": 1810
    },
    {
      "entropy": 6.208154487609863,
      "epoch": 0.20385241758858877,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999050687647008,
      "loss": 6.071,
      "mean_token_accuracy": 0.1487251937389374,
      "num_tokens": 4050786.0,
      "step": 1815
    },
    {
      "entropy": 6.128590059280396,
      "epoch": 0.2044139944965463,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999038990347984,
      "loss": 6.1181,
      "mean_token_accuracy": 0.1585391640663147,
      "num_tokens": 4061184.0,
      "step": 1820
    },
    {
      "entropy": 6.332269668579102,
      "epoch": 0.20497557140450384,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999027221438688,
      "loss": 6.2477,
      "mean_token_accuracy": 0.1459768995642662,
      "num_tokens": 4073607.0,
      "step": 1825
    },
    {
      "entropy": 6.105072927474976,
      "epoch": 0.2055371483124614,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499901538091949,
      "loss": 6.063,
      "mean_token_accuracy": 0.14765892326831817,
      "num_tokens": 4083924.0,
      "step": 1830
    },
    {
      "entropy": 6.109910249710083,
      "epoch": 0.20609872522041894,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999003468790773,
      "loss": 6.0556,
      "mean_token_accuracy": 0.15023878514766692,
      "num_tokens": 4094083.0,
      "step": 1835
    },
    {
      "entropy": 6.175615119934082,
      "epoch": 0.2066603021283765,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998991485052911,
      "loss": 6.0894,
      "mean_token_accuracy": 0.157446925342083,
      "num_tokens": 4105124.0,
      "step": 1840
    },
    {
      "entropy": 6.236642742156983,
      "epoch": 0.20722187903633402,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000499897942970629,
      "loss": 6.2515,
      "mean_token_accuracy": 0.14365296810865402,
      "num_tokens": 4116792.0,
      "step": 1845
    },
    {
      "entropy": 6.346523237228394,
      "epoch": 0.20778345594429157,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000499896730275129,
      "loss": 6.2425,
      "mean_token_accuracy": 0.14365033358335494,
      "num_tokens": 4127441.0,
      "step": 1850
    },
    {
      "entropy": 6.223315095901489,
      "epoch": 0.20834503285224912,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00049989551041883,
      "loss": 6.216,
      "mean_token_accuracy": 0.14539413154125214,
      "num_tokens": 4138696.0,
      "step": 1855
    },
    {
      "entropy": 6.098178768157959,
      "epoch": 0.20890660976020667,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998942834017707,
      "loss": 6.0751,
      "mean_token_accuracy": 0.1485121563076973,
      "num_tokens": 4151254.0,
      "step": 1860
    },
    {
      "entropy": 6.124366617202758,
      "epoch": 0.2094681866681642,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998930492239901,
      "loss": 6.0368,
      "mean_token_accuracy": 0.1491617165505886,
      "num_tokens": 4162046.0,
      "step": 1865
    },
    {
      "entropy": 6.119881010055542,
      "epoch": 0.21002976357612174,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998918078855277,
      "loss": 6.0023,
      "mean_token_accuracy": 0.15643363893032075,
      "num_tokens": 4172409.0,
      "step": 1870
    },
    {
      "entropy": 6.118890285491943,
      "epoch": 0.2105913404840793,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004998905593864227,
      "loss": 6.1515,
      "mean_token_accuracy": 0.1560173213481903,
      "num_tokens": 4182949.0,
      "step": 1875
    },
    {
      "entropy": 6.123640012741089,
      "epoch": 0.21115291739203684,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998893037267153,
      "loss": 6.1166,
      "mean_token_accuracy": 0.15241918265819548,
      "num_tokens": 4194164.0,
      "step": 1880
    },
    {
      "entropy": 6.308550214767456,
      "epoch": 0.2117144942999944,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998880409064452,
      "loss": 6.0822,
      "mean_token_accuracy": 0.15168551802635194,
      "num_tokens": 4206056.0,
      "step": 1885
    },
    {
      "entropy": 6.042774629592896,
      "epoch": 0.21227607120795192,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998867709256527,
      "loss": 6.1398,
      "mean_token_accuracy": 0.14869935661554337,
      "num_tokens": 4216660.0,
      "step": 1890
    },
    {
      "entropy": 6.086363315582275,
      "epoch": 0.21283764811590947,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000499885493784378,
      "loss": 6.0304,
      "mean_token_accuracy": 0.15354848504066468,
      "num_tokens": 4228495.0,
      "step": 1895
    },
    {
      "entropy": 6.137011957168579,
      "epoch": 0.21339922502386702,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998842094826621,
      "loss": 6.0173,
      "mean_token_accuracy": 0.15438924431800843,
      "num_tokens": 4238560.0,
      "step": 1900
    },
    {
      "entropy": 6.263278245925903,
      "epoch": 0.21396080193182457,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998829180205458,
      "loss": 6.2303,
      "mean_token_accuracy": 0.14329903498291968,
      "num_tokens": 4250553.0,
      "step": 1905
    },
    {
      "entropy": 6.198366451263428,
      "epoch": 0.21452237883978212,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998816193980701,
      "loss": 6.161,
      "mean_token_accuracy": 0.14747311770915986,
      "num_tokens": 4262050.0,
      "step": 1910
    },
    {
      "entropy": 6.277743291854859,
      "epoch": 0.21508395574773964,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998803136152764,
      "loss": 6.1105,
      "mean_token_accuracy": 0.14608076959848404,
      "num_tokens": 4273019.0,
      "step": 1915
    },
    {
      "entropy": 6.069996547698975,
      "epoch": 0.2156455326556972,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998790006722063,
      "loss": 6.0012,
      "mean_token_accuracy": 0.15598759353160857,
      "num_tokens": 4283585.0,
      "step": 1920
    },
    {
      "entropy": 6.026457405090332,
      "epoch": 0.21620710956365474,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998776805689016,
      "loss": 6.0773,
      "mean_token_accuracy": 0.15107496082782745,
      "num_tokens": 4294905.0,
      "step": 1925
    },
    {
      "entropy": 6.290648698806763,
      "epoch": 0.2167686864716123,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998763533054045,
      "loss": 6.1873,
      "mean_token_accuracy": 0.15128218084573747,
      "num_tokens": 4305926.0,
      "step": 1930
    },
    {
      "entropy": 6.10584077835083,
      "epoch": 0.21733026337956984,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998750188817568,
      "loss": 6.1502,
      "mean_token_accuracy": 0.14796533063054085,
      "num_tokens": 4318297.0,
      "step": 1935
    },
    {
      "entropy": 6.090063381195068,
      "epoch": 0.21789184028752737,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998736772980015,
      "loss": 5.979,
      "mean_token_accuracy": 0.15715189576148986,
      "num_tokens": 4328795.0,
      "step": 1940
    },
    {
      "entropy": 6.042044544219971,
      "epoch": 0.21845341719548492,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499872328554181,
      "loss": 5.9812,
      "mean_token_accuracy": 0.15147724151611328,
      "num_tokens": 4340279.0,
      "step": 1945
    },
    {
      "entropy": 6.2164843559265135,
      "epoch": 0.21901499410344247,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998709726503383,
      "loss": 6.1623,
      "mean_token_accuracy": 0.1436479665338993,
      "num_tokens": 4352157.0,
      "step": 1950
    },
    {
      "entropy": 6.175703525543213,
      "epoch": 0.21957657101140002,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998696095865168,
      "loss": 6.0064,
      "mean_token_accuracy": 0.1556989625096321,
      "num_tokens": 4361909.0,
      "step": 1955
    },
    {
      "entropy": 5.869294500350952,
      "epoch": 0.22013814791935757,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998682393627595,
      "loss": 5.9129,
      "mean_token_accuracy": 0.15484603345394135,
      "num_tokens": 4372177.0,
      "step": 1960
    },
    {
      "entropy": 6.039911985397339,
      "epoch": 0.2206997248273151,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998668619791104,
      "loss": 5.982,
      "mean_token_accuracy": 0.1563266322016716,
      "num_tokens": 4382872.0,
      "step": 1965
    },
    {
      "entropy": 6.159225559234619,
      "epoch": 0.22126130173527264,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998654774356131,
      "loss": 6.0804,
      "mean_token_accuracy": 0.1488205224275589,
      "num_tokens": 4393623.0,
      "step": 1970
    },
    {
      "entropy": 6.015854930877685,
      "epoch": 0.2218228786432302,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004998640857323118,
      "loss": 6.0076,
      "mean_token_accuracy": 0.14695556685328484,
      "num_tokens": 4404215.0,
      "step": 1975
    },
    {
      "entropy": 6.244098138809204,
      "epoch": 0.22238445555118774,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998626868692509,
      "loss": 6.2583,
      "mean_token_accuracy": 0.14658438861370088,
      "num_tokens": 4415978.0,
      "step": 1980
    },
    {
      "entropy": 6.135475778579712,
      "epoch": 0.2229460324591453,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998612808464748,
      "loss": 6.0107,
      "mean_token_accuracy": 0.15329247564077378,
      "num_tokens": 4427060.0,
      "step": 1985
    },
    {
      "entropy": 6.063277769088745,
      "epoch": 0.22350760936710282,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004998598676640283,
      "loss": 6.0722,
      "mean_token_accuracy": 0.14928158596158028,
      "num_tokens": 4438145.0,
      "step": 1990
    },
    {
      "entropy": 6.266687870025635,
      "epoch": 0.22406918627506037,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998584473219563,
      "loss": 6.1493,
      "mean_token_accuracy": 0.15549051016569138,
      "num_tokens": 4449324.0,
      "step": 1995
    },
    {
      "entropy": 6.027319240570068,
      "epoch": 0.22463076318301792,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998570198203043,
      "loss": 5.8902,
      "mean_token_accuracy": 0.15551431477069855,
      "num_tokens": 4459728.0,
      "step": 2000
    },
    {
      "entropy": 6.019188499450683,
      "epoch": 0.22519234009097547,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998555851591175,
      "loss": 6.1022,
      "mean_token_accuracy": 0.15041673630476,
      "num_tokens": 4471164.0,
      "step": 2005
    },
    {
      "entropy": 6.231358623504638,
      "epoch": 0.225753916998933,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004998541433384416,
      "loss": 6.1488,
      "mean_token_accuracy": 0.15111713707447053,
      "num_tokens": 4483349.0,
      "step": 2010
    },
    {
      "entropy": 6.196571016311646,
      "epoch": 0.22631549390689054,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998526943583226,
      "loss": 6.1081,
      "mean_token_accuracy": 0.1539139837026596,
      "num_tokens": 4495311.0,
      "step": 2015
    },
    {
      "entropy": 6.084535074234009,
      "epoch": 0.2268770708148481,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998512382188066,
      "loss": 6.0404,
      "mean_token_accuracy": 0.15315846651792525,
      "num_tokens": 4507017.0,
      "step": 2020
    },
    {
      "entropy": 6.096588039398194,
      "epoch": 0.22743864772280564,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998497749199401,
      "loss": 6.0761,
      "mean_token_accuracy": 0.15145348012447357,
      "num_tokens": 4517746.0,
      "step": 2025
    },
    {
      "entropy": 6.213549852371216,
      "epoch": 0.2280002246307632,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998483044617694,
      "loss": 6.1664,
      "mean_token_accuracy": 0.145511694252491,
      "num_tokens": 4529149.0,
      "step": 2030
    },
    {
      "entropy": 6.154675054550171,
      "epoch": 0.22856180153872072,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998468268443416,
      "loss": 6.051,
      "mean_token_accuracy": 0.15400808304548264,
      "num_tokens": 4540630.0,
      "step": 2035
    },
    {
      "entropy": 6.01413254737854,
      "epoch": 0.22912337844667827,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998453420677034,
      "loss": 5.9758,
      "mean_token_accuracy": 0.1529553934931755,
      "num_tokens": 4552877.0,
      "step": 2040
    },
    {
      "entropy": 6.154169607162475,
      "epoch": 0.22968495535463582,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998438501319025,
      "loss": 6.0693,
      "mean_token_accuracy": 0.15515930503606795,
      "num_tokens": 4563634.0,
      "step": 2045
    },
    {
      "entropy": 6.079659175872803,
      "epoch": 0.23024653226259337,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998423510369862,
      "loss": 6.0185,
      "mean_token_accuracy": 0.16131191849708557,
      "num_tokens": 4574007.0,
      "step": 2050
    },
    {
      "entropy": 6.063276243209839,
      "epoch": 0.23080810917055092,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998408447830022,
      "loss": 6.0141,
      "mean_token_accuracy": 0.1526772379875183,
      "num_tokens": 4584146.0,
      "step": 2055
    },
    {
      "entropy": 6.107676076889038,
      "epoch": 0.23136968607850844,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998393313699985,
      "loss": 6.0371,
      "mean_token_accuracy": 0.1490740567445755,
      "num_tokens": 4594839.0,
      "step": 2060
    },
    {
      "entropy": 6.250979852676392,
      "epoch": 0.231931262986466,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004998378107980233,
      "loss": 6.1898,
      "mean_token_accuracy": 0.14502785950899125,
      "num_tokens": 4608229.0,
      "step": 2065
    },
    {
      "entropy": 6.109724712371826,
      "epoch": 0.23249283989442354,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998362830671249,
      "loss": 5.9936,
      "mean_token_accuracy": 0.15285840034484863,
      "num_tokens": 4619206.0,
      "step": 2070
    },
    {
      "entropy": 6.003591156005859,
      "epoch": 0.2330544168023811,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998347481773522,
      "loss": 6.0237,
      "mean_token_accuracy": 0.15680260062217713,
      "num_tokens": 4628966.0,
      "step": 2075
    },
    {
      "entropy": 6.083307552337646,
      "epoch": 0.23361599371033864,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998332061287538,
      "loss": 5.9268,
      "mean_token_accuracy": 0.1513105571269989,
      "num_tokens": 4639935.0,
      "step": 2080
    },
    {
      "entropy": 6.155953454971313,
      "epoch": 0.23417757061829617,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499831656921379,
      "loss": 6.1565,
      "mean_token_accuracy": 0.14754048883914947,
      "num_tokens": 4651998.0,
      "step": 2085
    },
    {
      "entropy": 6.174464988708496,
      "epoch": 0.23473914752625372,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499830100555277,
      "loss": 6.1424,
      "mean_token_accuracy": 0.1461272232234478,
      "num_tokens": 4664002.0,
      "step": 2090
    },
    {
      "entropy": 6.1856427669525145,
      "epoch": 0.23530072443421127,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998285370304974,
      "loss": 6.0264,
      "mean_token_accuracy": 0.15473159700632094,
      "num_tokens": 4675343.0,
      "step": 2095
    },
    {
      "entropy": 6.077418518066406,
      "epoch": 0.23586230134216882,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00049982696634709,
      "loss": 6.0756,
      "mean_token_accuracy": 0.15717330425977707,
      "num_tokens": 4686410.0,
      "step": 2100
    },
    {
      "entropy": 6.082688808441162,
      "epoch": 0.23642387825012637,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998253885051048,
      "loss": 6.0996,
      "mean_token_accuracy": 0.15195055305957794,
      "num_tokens": 4697768.0,
      "step": 2105
    },
    {
      "entropy": 6.22024130821228,
      "epoch": 0.2369854551580839,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499823803504592,
      "loss": 6.1022,
      "mean_token_accuracy": 0.1504859909415245,
      "num_tokens": 4709933.0,
      "step": 2110
    },
    {
      "entropy": 6.151685571670532,
      "epoch": 0.23754703206604144,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998222113456022,
      "loss": 6.1375,
      "mean_token_accuracy": 0.14684750735759736,
      "num_tokens": 4721385.0,
      "step": 2115
    },
    {
      "entropy": 6.061194324493409,
      "epoch": 0.238108608973999,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998206120281858,
      "loss": 5.956,
      "mean_token_accuracy": 0.1638572856783867,
      "num_tokens": 4732562.0,
      "step": 2120
    },
    {
      "entropy": 6.022126531600952,
      "epoch": 0.23867018588195654,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499819005552394,
      "loss": 5.9696,
      "mean_token_accuracy": 0.15314062237739562,
      "num_tokens": 4742794.0,
      "step": 2125
    },
    {
      "entropy": 5.996016979217529,
      "epoch": 0.23923176278991407,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004998173919182779,
      "loss": 5.934,
      "mean_token_accuracy": 0.1606994539499283,
      "num_tokens": 4754903.0,
      "step": 2130
    },
    {
      "entropy": 6.060186147689819,
      "epoch": 0.23979333969787162,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499815771125889,
      "loss": 5.9783,
      "mean_token_accuracy": 0.15231975466012954,
      "num_tokens": 4766165.0,
      "step": 2135
    },
    {
      "entropy": 6.0076862335205075,
      "epoch": 0.24035491660582917,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998141431752785,
      "loss": 6.0022,
      "mean_token_accuracy": 0.1594335064291954,
      "num_tokens": 4776551.0,
      "step": 2140
    },
    {
      "entropy": 6.193838834762573,
      "epoch": 0.24091649351378672,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998125080664987,
      "loss": 6.1794,
      "mean_token_accuracy": 0.14647017419338226,
      "num_tokens": 4788344.0,
      "step": 2145
    },
    {
      "entropy": 6.1166502952575685,
      "epoch": 0.24147807042174427,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998108657996012,
      "loss": 5.9985,
      "mean_token_accuracy": 0.1459207035601139,
      "num_tokens": 4799314.0,
      "step": 2150
    },
    {
      "entropy": 6.083969259262085,
      "epoch": 0.2420396473297018,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998092163746387,
      "loss": 6.0212,
      "mean_token_accuracy": 0.15580957233905793,
      "num_tokens": 4809588.0,
      "step": 2155
    },
    {
      "entropy": 5.976114320755005,
      "epoch": 0.24260122423765934,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998075597916636,
      "loss": 5.8474,
      "mean_token_accuracy": 0.16323217898607253,
      "num_tokens": 4820076.0,
      "step": 2160
    },
    {
      "entropy": 6.061397123336792,
      "epoch": 0.2431628011456169,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998058960507286,
      "loss": 6.1208,
      "mean_token_accuracy": 0.14964161813259125,
      "num_tokens": 4831504.0,
      "step": 2165
    },
    {
      "entropy": 6.107447004318237,
      "epoch": 0.24372437805357444,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998042251518866,
      "loss": 5.9668,
      "mean_token_accuracy": 0.15928057283163072,
      "num_tokens": 4842826.0,
      "step": 2170
    },
    {
      "entropy": 5.872031927108765,
      "epoch": 0.244285954961532,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998025470951908,
      "loss": 5.9572,
      "mean_token_accuracy": 0.15852829664945603,
      "num_tokens": 4853279.0,
      "step": 2175
    },
    {
      "entropy": 6.157621288299561,
      "epoch": 0.24484753186948952,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998008618806949,
      "loss": 6.0044,
      "mean_token_accuracy": 0.16119896918535231,
      "num_tokens": 4863332.0,
      "step": 2180
    },
    {
      "entropy": 6.080954504013062,
      "epoch": 0.24540910877744707,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004997991695084523,
      "loss": 6.1308,
      "mean_token_accuracy": 0.14659972339868546,
      "num_tokens": 4874217.0,
      "step": 2185
    },
    {
      "entropy": 6.237659978866577,
      "epoch": 0.24597068568540462,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997974699785169,
      "loss": 6.0082,
      "mean_token_accuracy": 0.16113914996385575,
      "num_tokens": 4885322.0,
      "step": 2190
    },
    {
      "entropy": 5.919554424285889,
      "epoch": 0.24653226259336217,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997957632909429,
      "loss": 5.9381,
      "mean_token_accuracy": 0.15738066136837006,
      "num_tokens": 4895546.0,
      "step": 2195
    },
    {
      "entropy": 6.218607234954834,
      "epoch": 0.24709383950131972,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997940494457846,
      "loss": 6.154,
      "mean_token_accuracy": 0.14981306791305543,
      "num_tokens": 4906957.0,
      "step": 2200
    },
    {
      "entropy": 6.19651517868042,
      "epoch": 0.24765541640927724,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997923284430967,
      "loss": 6.0821,
      "mean_token_accuracy": 0.14576388597488404,
      "num_tokens": 4917965.0,
      "step": 2205
    },
    {
      "entropy": 6.0613480567932125,
      "epoch": 0.2482169933172348,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997906002829338,
      "loss": 5.9757,
      "mean_token_accuracy": 0.15590714514255524,
      "num_tokens": 4928585.0,
      "step": 2210
    },
    {
      "entropy": 5.968810176849365,
      "epoch": 0.24877857022519234,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000499788864965351,
      "loss": 5.9858,
      "mean_token_accuracy": 0.1578755110502243,
      "num_tokens": 4939818.0,
      "step": 2215
    },
    {
      "entropy": 6.131110334396363,
      "epoch": 0.2493401471331499,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997871224904036,
      "loss": 6.0622,
      "mean_token_accuracy": 0.14955551028251649,
      "num_tokens": 4951752.0,
      "step": 2220
    },
    {
      "entropy": 5.9506998538970945,
      "epoch": 0.24990172404110744,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997853728581468,
      "loss": 5.8901,
      "mean_token_accuracy": 0.16224861592054368,
      "num_tokens": 4961846.0,
      "step": 2225
    },
    {
      "entropy": 6.0852518558502195,
      "epoch": 0.250463300949065,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004997836160686368,
      "loss": 6.0521,
      "mean_token_accuracy": 0.15131033658981324,
      "num_tokens": 4973693.0,
      "step": 2230
    },
    {
      "entropy": 6.060042428970337,
      "epoch": 0.25102487785702254,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004997818521219291,
      "loss": 5.9604,
      "mean_token_accuracy": 0.1576952561736107,
      "num_tokens": 4986156.0,
      "step": 2235
    },
    {
      "entropy": 6.013088703155518,
      "epoch": 0.25158645476498004,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997800810180803,
      "loss": 6.0447,
      "mean_token_accuracy": 0.15376421064138412,
      "num_tokens": 4997615.0,
      "step": 2240
    },
    {
      "entropy": 6.1519105434417725,
      "epoch": 0.2521480316729376,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997783027571464,
      "loss": 5.9679,
      "mean_token_accuracy": 0.15947780162096023,
      "num_tokens": 5007655.0,
      "step": 2245
    },
    {
      "entropy": 6.026669692993164,
      "epoch": 0.25270960858089514,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997765173391841,
      "loss": 6.0482,
      "mean_token_accuracy": 0.15224288105964662,
      "num_tokens": 5018933.0,
      "step": 2250
    },
    {
      "entropy": 6.166677522659302,
      "epoch": 0.2532711854888527,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997747247642505,
      "loss": 6.0525,
      "mean_token_accuracy": 0.15358851104974747,
      "num_tokens": 5030495.0,
      "step": 2255
    },
    {
      "entropy": 6.102503728866577,
      "epoch": 0.25383276239681024,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997729250324023,
      "loss": 6.0817,
      "mean_token_accuracy": 0.1501933827996254,
      "num_tokens": 5041553.0,
      "step": 2260
    },
    {
      "entropy": 6.090739488601685,
      "epoch": 0.2543943393047678,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997711181436971,
      "loss": 6.007,
      "mean_token_accuracy": 0.15912422835826873,
      "num_tokens": 5052312.0,
      "step": 2265
    },
    {
      "entropy": 6.151770973205567,
      "epoch": 0.25495591621272534,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004997693040981922,
      "loss": 6.1953,
      "mean_token_accuracy": 0.14825651198625564,
      "num_tokens": 5063445.0,
      "step": 2270
    },
    {
      "entropy": 6.138964891433716,
      "epoch": 0.2555174931206829,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997674828959456,
      "loss": 5.9824,
      "mean_token_accuracy": 0.15490673780441283,
      "num_tokens": 5073052.0,
      "step": 2275
    },
    {
      "entropy": 6.038366365432739,
      "epoch": 0.25607907002864044,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004997656545370151,
      "loss": 6.0267,
      "mean_token_accuracy": 0.15771733224391937,
      "num_tokens": 5084174.0,
      "step": 2280
    },
    {
      "entropy": 6.103801107406616,
      "epoch": 0.256640646936598,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499763819021459,
      "loss": 6.0128,
      "mean_token_accuracy": 0.15576983839273453,
      "num_tokens": 5094106.0,
      "step": 2285
    },
    {
      "entropy": 6.129735326766967,
      "epoch": 0.2572022238445555,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004997619763493357,
      "loss": 6.0902,
      "mean_token_accuracy": 0.15617480129003525,
      "num_tokens": 5106963.0,
      "step": 2290
    },
    {
      "entropy": 5.987654972076416,
      "epoch": 0.25776380075251304,
      "grad_norm": 0.890625,
      "learning_rate": 0.000499760126520704,
      "loss": 5.8927,
      "mean_token_accuracy": 0.15116089582443237,
      "num_tokens": 5119476.0,
      "step": 2295
    },
    {
      "entropy": 6.157898807525635,
      "epoch": 0.2583253776604706,
      "grad_norm": 0.875,
      "learning_rate": 0.0004997582695356225,
      "loss": 6.1371,
      "mean_token_accuracy": 0.15195467919111252,
      "num_tokens": 5132169.0,
      "step": 2300
    },
    {
      "entropy": 6.094862699508667,
      "epoch": 0.25888695456842814,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997564053941507,
      "loss": 5.9806,
      "mean_token_accuracy": 0.1550524652004242,
      "num_tokens": 5143354.0,
      "step": 2305
    },
    {
      "entropy": 5.958035707473755,
      "epoch": 0.2594485314763857,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997545340963478,
      "loss": 6.0541,
      "mean_token_accuracy": 0.1468196243047714,
      "num_tokens": 5154624.0,
      "step": 2310
    },
    {
      "entropy": 6.126406049728393,
      "epoch": 0.26001010838434324,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997526556422733,
      "loss": 5.9392,
      "mean_token_accuracy": 0.15484741032123567,
      "num_tokens": 5165349.0,
      "step": 2315
    },
    {
      "entropy": 6.043291521072388,
      "epoch": 0.2605716852923008,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997507700319872,
      "loss": 6.0235,
      "mean_token_accuracy": 0.15324235707521439,
      "num_tokens": 5175874.0,
      "step": 2320
    },
    {
      "entropy": 6.055290842056275,
      "epoch": 0.26113326220025834,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997488772655492,
      "loss": 5.9108,
      "mean_token_accuracy": 0.16078570783138274,
      "num_tokens": 5186591.0,
      "step": 2325
    },
    {
      "entropy": 5.993036842346191,
      "epoch": 0.2616948391082159,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997469773430199,
      "loss": 6.0215,
      "mean_token_accuracy": 0.1626984179019928,
      "num_tokens": 5197307.0,
      "step": 2330
    },
    {
      "entropy": 6.009312629699707,
      "epoch": 0.2622564160161734,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004997450702644596,
      "loss": 5.9781,
      "mean_token_accuracy": 0.15396428406238555,
      "num_tokens": 5209868.0,
      "step": 2335
    },
    {
      "entropy": 6.020443344116211,
      "epoch": 0.26281799292413094,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997431560299292,
      "loss": 5.9995,
      "mean_token_accuracy": 0.15726363211870192,
      "num_tokens": 5220365.0,
      "step": 2340
    },
    {
      "entropy": 6.102507829666138,
      "epoch": 0.2633795698320885,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997412346394894,
      "loss": 5.9229,
      "mean_token_accuracy": 0.15023885518312455,
      "num_tokens": 5231575.0,
      "step": 2345
    },
    {
      "entropy": 6.035394668579102,
      "epoch": 0.26394114674004604,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997393060932017,
      "loss": 5.9938,
      "mean_token_accuracy": 0.15539337694644928,
      "num_tokens": 5241852.0,
      "step": 2350
    },
    {
      "entropy": 6.038948059082031,
      "epoch": 0.2645027236480036,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997373703911273,
      "loss": 6.0288,
      "mean_token_accuracy": 0.15632582753896712,
      "num_tokens": 5254536.0,
      "step": 2355
    },
    {
      "entropy": 6.083648633956909,
      "epoch": 0.26506430055596114,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997354275333277,
      "loss": 5.9606,
      "mean_token_accuracy": 0.15450441390275954,
      "num_tokens": 5264946.0,
      "step": 2360
    },
    {
      "entropy": 5.8433245658874515,
      "epoch": 0.2656258774639187,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499733477519865,
      "loss": 5.784,
      "mean_token_accuracy": 0.15305204540491105,
      "num_tokens": 5275527.0,
      "step": 2365
    },
    {
      "entropy": 6.025034093856812,
      "epoch": 0.26618745437187624,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997315203508013,
      "loss": 5.9569,
      "mean_token_accuracy": 0.15324413478374482,
      "num_tokens": 5285471.0,
      "step": 2370
    },
    {
      "entropy": 5.916028690338135,
      "epoch": 0.2667490312798338,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997295560261988,
      "loss": 5.89,
      "mean_token_accuracy": 0.15359498262405397,
      "num_tokens": 5295743.0,
      "step": 2375
    },
    {
      "entropy": 6.049281930923462,
      "epoch": 0.26731060818779134,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00049972758454612,
      "loss": 6.0335,
      "mean_token_accuracy": 0.1482474312186241,
      "num_tokens": 5307963.0,
      "step": 2380
    },
    {
      "entropy": 6.043624353408814,
      "epoch": 0.26787218509574884,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997256059106279,
      "loss": 5.9176,
      "mean_token_accuracy": 0.15914819538593292,
      "num_tokens": 5318394.0,
      "step": 2385
    },
    {
      "entropy": 6.079010963439941,
      "epoch": 0.2684337620037064,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004997236201197853,
      "loss": 6.144,
      "mean_token_accuracy": 0.1458563432097435,
      "num_tokens": 5331728.0,
      "step": 2390
    },
    {
      "entropy": 6.132135200500488,
      "epoch": 0.26899533891166394,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997216271736554,
      "loss": 6.0232,
      "mean_token_accuracy": 0.16378798931837082,
      "num_tokens": 5342825.0,
      "step": 2395
    },
    {
      "entropy": 6.050425243377686,
      "epoch": 0.2695569158196215,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997196270723018,
      "loss": 5.9357,
      "mean_token_accuracy": 0.15482408478856086,
      "num_tokens": 5354232.0,
      "step": 2400
    },
    {
      "entropy": 6.0575597286224365,
      "epoch": 0.27011849272757904,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997176198157881,
      "loss": 6.0253,
      "mean_token_accuracy": 0.1507862962782383,
      "num_tokens": 5364807.0,
      "step": 2405
    },
    {
      "entropy": 6.040148830413818,
      "epoch": 0.2706800696355366,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997156054041781,
      "loss": 5.9844,
      "mean_token_accuracy": 0.15217866897583007,
      "num_tokens": 5375593.0,
      "step": 2410
    },
    {
      "entropy": 6.045368623733521,
      "epoch": 0.27124164654349414,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997135838375362,
      "loss": 5.9655,
      "mean_token_accuracy": 0.15985994040966034,
      "num_tokens": 5386740.0,
      "step": 2415
    },
    {
      "entropy": 5.910307359695435,
      "epoch": 0.2718032234514517,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997115551159267,
      "loss": 5.9274,
      "mean_token_accuracy": 0.15814046263694764,
      "num_tokens": 5397732.0,
      "step": 2420
    },
    {
      "entropy": 6.159975671768189,
      "epoch": 0.27236480035940924,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499709519239414,
      "loss": 5.9925,
      "mean_token_accuracy": 0.14930515587329865,
      "num_tokens": 5409392.0,
      "step": 2425
    },
    {
      "entropy": 5.926601505279541,
      "epoch": 0.2729263772673668,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004997074762080632,
      "loss": 5.9607,
      "mean_token_accuracy": 0.16299881041049957,
      "num_tokens": 5420903.0,
      "step": 2430
    },
    {
      "entropy": 5.892569589614868,
      "epoch": 0.2734879541753243,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000499705426021939,
      "loss": 5.8882,
      "mean_token_accuracy": 0.15828869342803956,
      "num_tokens": 5432631.0,
      "step": 2435
    },
    {
      "entropy": 6.01947603225708,
      "epoch": 0.27404953108328184,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499703368681107,
      "loss": 5.8652,
      "mean_token_accuracy": 0.16353946775197983,
      "num_tokens": 5443463.0,
      "step": 2440
    },
    {
      "entropy": 6.1192179203033445,
      "epoch": 0.2746111079912394,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004997013041856326,
      "loss": 6.1688,
      "mean_token_accuracy": 0.15179817527532577,
      "num_tokens": 5455903.0,
      "step": 2445
    },
    {
      "entropy": 6.088233995437622,
      "epoch": 0.27517268489919694,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004996992325355815,
      "loss": 5.9814,
      "mean_token_accuracy": 0.15265644788742067,
      "num_tokens": 5467201.0,
      "step": 2450
    },
    {
      "entropy": 5.960464429855347,
      "epoch": 0.2757342618071545,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004996971537310198,
      "loss": 5.8577,
      "mean_token_accuracy": 0.15811959505081177,
      "num_tokens": 5478109.0,
      "step": 2455
    },
    {
      "entropy": 6.074854564666748,
      "epoch": 0.27629583871511204,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996950677720133,
      "loss": 5.9994,
      "mean_token_accuracy": 0.15298570096492767,
      "num_tokens": 5489621.0,
      "step": 2460
    },
    {
      "entropy": 5.961808776855468,
      "epoch": 0.2768574156230696,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499692974658629,
      "loss": 5.9402,
      "mean_token_accuracy": 0.15553341507911683,
      "num_tokens": 5500606.0,
      "step": 2465
    },
    {
      "entropy": 5.985595941543579,
      "epoch": 0.27741899253102714,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996908743909331,
      "loss": 5.8963,
      "mean_token_accuracy": 0.16250037103891374,
      "num_tokens": 5510768.0,
      "step": 2470
    },
    {
      "entropy": 5.943635702133179,
      "epoch": 0.2779805694389847,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996887669689926,
      "loss": 5.9121,
      "mean_token_accuracy": 0.15700624138116837,
      "num_tokens": 5522210.0,
      "step": 2475
    },
    {
      "entropy": 6.043688678741455,
      "epoch": 0.2785421463469422,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996866523928747,
      "loss": 6.0396,
      "mean_token_accuracy": 0.1516438513994217,
      "num_tokens": 5533055.0,
      "step": 2480
    },
    {
      "entropy": 6.1347261428833,
      "epoch": 0.27910372325489974,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996845306626467,
      "loss": 6.0314,
      "mean_token_accuracy": 0.15147426277399062,
      "num_tokens": 5544345.0,
      "step": 2485
    },
    {
      "entropy": 6.014614534378052,
      "epoch": 0.2796653001628573,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996824017783759,
      "loss": 5.9502,
      "mean_token_accuracy": 0.15230564698576926,
      "num_tokens": 5556734.0,
      "step": 2490
    },
    {
      "entropy": 6.008048486709595,
      "epoch": 0.28022687707081484,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996802657401304,
      "loss": 5.9865,
      "mean_token_accuracy": 0.15721939951181413,
      "num_tokens": 5566874.0,
      "step": 2495
    },
    {
      "entropy": 6.101916837692261,
      "epoch": 0.2807884539787724,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004996781225479781,
      "loss": 5.9791,
      "mean_token_accuracy": 0.15063416808843613,
      "num_tokens": 5578812.0,
      "step": 2500
    },
    {
      "entropy": 6.013814878463745,
      "epoch": 0.28135003088672994,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004996759722019872,
      "loss": 6.0271,
      "mean_token_accuracy": 0.15371579229831694,
      "num_tokens": 5590021.0,
      "step": 2505
    },
    {
      "entropy": 6.024594354629516,
      "epoch": 0.2819116077946875,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004996738147022262,
      "loss": 6.0567,
      "mean_token_accuracy": 0.14784704446792601,
      "num_tokens": 5600921.0,
      "step": 2510
    },
    {
      "entropy": 6.051389265060425,
      "epoch": 0.28247318470264504,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996716500487638,
      "loss": 5.9445,
      "mean_token_accuracy": 0.15591129958629607,
      "num_tokens": 5610615.0,
      "step": 2515
    },
    {
      "entropy": 5.990168857574463,
      "epoch": 0.2830347616106026,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499669478241669,
      "loss": 5.8949,
      "mean_token_accuracy": 0.15703389495611192,
      "num_tokens": 5621596.0,
      "step": 2520
    },
    {
      "entropy": 6.092303705215454,
      "epoch": 0.28359633851856014,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996672992810108,
      "loss": 5.9413,
      "mean_token_accuracy": 0.15879166424274443,
      "num_tokens": 5633047.0,
      "step": 2525
    },
    {
      "entropy": 5.900774335861206,
      "epoch": 0.28415791542651764,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996651131668587,
      "loss": 5.9041,
      "mean_token_accuracy": 0.1561056926846504,
      "num_tokens": 5644970.0,
      "step": 2530
    },
    {
      "entropy": 5.969459009170532,
      "epoch": 0.2847194923344752,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004996629198992823,
      "loss": 6.0379,
      "mean_token_accuracy": 0.15508987456560136,
      "num_tokens": 5656631.0,
      "step": 2535
    },
    {
      "entropy": 6.097274971008301,
      "epoch": 0.28528106924243274,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004996607194783513,
      "loss": 5.9743,
      "mean_token_accuracy": 0.1543091729283333,
      "num_tokens": 5668808.0,
      "step": 2540
    },
    {
      "entropy": 5.94429612159729,
      "epoch": 0.2858426461503903,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996585119041359,
      "loss": 5.9251,
      "mean_token_accuracy": 0.15240908712148665,
      "num_tokens": 5679493.0,
      "step": 2545
    },
    {
      "entropy": 6.034660053253174,
      "epoch": 0.28640422305834784,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996562971767063,
      "loss": 5.9655,
      "mean_token_accuracy": 0.15858010351657867,
      "num_tokens": 5689879.0,
      "step": 2550
    },
    {
      "entropy": 6.085075855255127,
      "epoch": 0.2869657999663054,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499654075296133,
      "loss": 6.0574,
      "mean_token_accuracy": 0.15524982064962387,
      "num_tokens": 5699934.0,
      "step": 2555
    },
    {
      "entropy": 6.04909462928772,
      "epoch": 0.28752737687426294,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499651846262487,
      "loss": 5.9431,
      "mean_token_accuracy": 0.1601421758532524,
      "num_tokens": 5710171.0,
      "step": 2560
    },
    {
      "entropy": 5.977365207672119,
      "epoch": 0.2880889537822205,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996496100758389,
      "loss": 5.9509,
      "mean_token_accuracy": 0.15049331039190292,
      "num_tokens": 5721190.0,
      "step": 2565
    },
    {
      "entropy": 6.137782859802246,
      "epoch": 0.28865053069017804,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996473667362604,
      "loss": 6.0564,
      "mean_token_accuracy": 0.15206752270460128,
      "num_tokens": 5733706.0,
      "step": 2570
    },
    {
      "entropy": 5.952969312667847,
      "epoch": 0.28921210759813554,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996451162438224,
      "loss": 6.0291,
      "mean_token_accuracy": 0.1582716077566147,
      "num_tokens": 5745712.0,
      "step": 2575
    },
    {
      "entropy": 5.9377197265625,
      "epoch": 0.2897736845060931,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996428585985967,
      "loss": 5.8341,
      "mean_token_accuracy": 0.167465503513813,
      "num_tokens": 5757915.0,
      "step": 2580
    },
    {
      "entropy": 6.0136088848114015,
      "epoch": 0.29033526141405064,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996405938006553,
      "loss": 5.983,
      "mean_token_accuracy": 0.1571892872452736,
      "num_tokens": 5768788.0,
      "step": 2585
    },
    {
      "entropy": 6.062278985977173,
      "epoch": 0.2908968383220082,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996383218500703,
      "loss": 5.9163,
      "mean_token_accuracy": 0.15861130356788636,
      "num_tokens": 5779806.0,
      "step": 2590
    },
    {
      "entropy": 5.848001956939697,
      "epoch": 0.29145841522996574,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499636042746914,
      "loss": 5.8739,
      "mean_token_accuracy": 0.1530233860015869,
      "num_tokens": 5790177.0,
      "step": 2595
    },
    {
      "entropy": 6.030651712417603,
      "epoch": 0.2920199921379233,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499633756491259,
      "loss": 5.9058,
      "mean_token_accuracy": 0.16201242804527283,
      "num_tokens": 5800251.0,
      "step": 2600
    },
    {
      "entropy": 6.013105869293213,
      "epoch": 0.29258156904588084,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004996314630831781,
      "loss": 6.0029,
      "mean_token_accuracy": 0.1577301412820816,
      "num_tokens": 5812210.0,
      "step": 2605
    },
    {
      "entropy": 6.03218035697937,
      "epoch": 0.2931431459538384,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004996291625227443,
      "loss": 5.972,
      "mean_token_accuracy": 0.15995176434516906,
      "num_tokens": 5823280.0,
      "step": 2610
    },
    {
      "entropy": 6.098472261428833,
      "epoch": 0.29370472286179594,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004996268548100307,
      "loss": 6.0516,
      "mean_token_accuracy": 0.14447705671191216,
      "num_tokens": 5835076.0,
      "step": 2615
    },
    {
      "entropy": 5.97599425315857,
      "epoch": 0.2942662997697535,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499624539945111,
      "loss": 5.8353,
      "mean_token_accuracy": 0.15990931689739227,
      "num_tokens": 5845376.0,
      "step": 2620
    },
    {
      "entropy": 5.9088949203491214,
      "epoch": 0.294827876677711,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004996222179280589,
      "loss": 5.8904,
      "mean_token_accuracy": 0.15515819191932678,
      "num_tokens": 5857124.0,
      "step": 2625
    },
    {
      "entropy": 6.012001180648804,
      "epoch": 0.29538945358566854,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996198887589481,
      "loss": 5.9792,
      "mean_token_accuracy": 0.14992934688925744,
      "num_tokens": 5868681.0,
      "step": 2630
    },
    {
      "entropy": 6.072331762313842,
      "epoch": 0.2959510304936261,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004996175524378531,
      "loss": 5.9727,
      "mean_token_accuracy": 0.1609386384487152,
      "num_tokens": 5880532.0,
      "step": 2635
    },
    {
      "entropy": 5.996478271484375,
      "epoch": 0.29651260740158364,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499615208964848,
      "loss": 5.9347,
      "mean_token_accuracy": 0.16165445894002914,
      "num_tokens": 5891582.0,
      "step": 2640
    },
    {
      "entropy": 5.978861856460571,
      "epoch": 0.2970741843095412,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996128583400077,
      "loss": 5.932,
      "mean_token_accuracy": 0.150858586281538,
      "num_tokens": 5903543.0,
      "step": 2645
    },
    {
      "entropy": 5.955687618255615,
      "epoch": 0.29763576121749874,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996105005634066,
      "loss": 5.9003,
      "mean_token_accuracy": 0.1639521986246109,
      "num_tokens": 5913140.0,
      "step": 2650
    },
    {
      "entropy": 5.951510906219482,
      "epoch": 0.2981973381254563,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996081356351202,
      "loss": 5.87,
      "mean_token_accuracy": 0.15955302566289903,
      "num_tokens": 5924010.0,
      "step": 2655
    },
    {
      "entropy": 6.063926649093628,
      "epoch": 0.29875891503341384,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996057635552237,
      "loss": 6.0587,
      "mean_token_accuracy": 0.15360975861549378,
      "num_tokens": 5936246.0,
      "step": 2660
    },
    {
      "entropy": 6.02715482711792,
      "epoch": 0.2993204919413714,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996033843237925,
      "loss": 5.9615,
      "mean_token_accuracy": 0.1463753491640091,
      "num_tokens": 5947594.0,
      "step": 2665
    },
    {
      "entropy": 6.025208616256714,
      "epoch": 0.29988206884932894,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996009979409025,
      "loss": 5.8914,
      "mean_token_accuracy": 0.1480000451207161,
      "num_tokens": 5957660.0,
      "step": 2670
    },
    {
      "entropy": 5.800934886932373,
      "epoch": 0.30044364575728644,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995986044066295,
      "loss": 5.7274,
      "mean_token_accuracy": 0.16546161770820617,
      "num_tokens": 5968881.0,
      "step": 2675
    },
    {
      "entropy": 5.921378183364868,
      "epoch": 0.301005222665244,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00049959620372105,
      "loss": 5.9065,
      "mean_token_accuracy": 0.1579037681221962,
      "num_tokens": 5980032.0,
      "step": 2680
    },
    {
      "entropy": 5.919540977478027,
      "epoch": 0.30156679957320154,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004995937958842401,
      "loss": 5.8937,
      "mean_token_accuracy": 0.15591801702976227,
      "num_tokens": 5991781.0,
      "step": 2685
    },
    {
      "entropy": 5.953354835510254,
      "epoch": 0.3021283764811591,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995913808962767,
      "loss": 5.948,
      "mean_token_accuracy": 0.15558021366596222,
      "num_tokens": 6002303.0,
      "step": 2690
    },
    {
      "entropy": 6.10141167640686,
      "epoch": 0.30268995338911664,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995889587572366,
      "loss": 6.0053,
      "mean_token_accuracy": 0.1546380952000618,
      "num_tokens": 6011523.0,
      "step": 2695
    },
    {
      "entropy": 5.984244871139526,
      "epoch": 0.3032515302970742,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499586529467197,
      "loss": 5.8236,
      "mean_token_accuracy": 0.15790919214487076,
      "num_tokens": 6022169.0,
      "step": 2700
    },
    {
      "entropy": 5.809388446807861,
      "epoch": 0.30381310720503174,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995840930262351,
      "loss": 5.8721,
      "mean_token_accuracy": 0.1557815633714199,
      "num_tokens": 6033393.0,
      "step": 2705
    },
    {
      "entropy": 6.070313596725464,
      "epoch": 0.3043746841129893,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004995816494344287,
      "loss": 5.9486,
      "mean_token_accuracy": 0.15179309844970704,
      "num_tokens": 6045365.0,
      "step": 2710
    },
    {
      "entropy": 5.969070339202881,
      "epoch": 0.30493626102094684,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995791986918554,
      "loss": 5.9637,
      "mean_token_accuracy": 0.15632378160953522,
      "num_tokens": 6057482.0,
      "step": 2715
    },
    {
      "entropy": 5.966453695297242,
      "epoch": 0.30549783792890434,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995767407985933,
      "loss": 5.8717,
      "mean_token_accuracy": 0.16056556701660157,
      "num_tokens": 6068194.0,
      "step": 2720
    },
    {
      "entropy": 5.998284339904785,
      "epoch": 0.3060594148368619,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995742757547207,
      "loss": 5.9356,
      "mean_token_accuracy": 0.1577038884162903,
      "num_tokens": 6080540.0,
      "step": 2725
    },
    {
      "entropy": 5.920602607727051,
      "epoch": 0.30662099174481944,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000499571803560316,
      "loss": 5.9674,
      "mean_token_accuracy": 0.15675391405820846,
      "num_tokens": 6092464.0,
      "step": 2730
    },
    {
      "entropy": 6.051038122177124,
      "epoch": 0.307182568652777,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004995693242154581,
      "loss": 5.9221,
      "mean_token_accuracy": 0.1568444013595581,
      "num_tokens": 6103242.0,
      "step": 2735
    },
    {
      "entropy": 5.945160531997681,
      "epoch": 0.30774414556073454,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004995668377202258,
      "loss": 5.9509,
      "mean_token_accuracy": 0.15690646022558213,
      "num_tokens": 6116383.0,
      "step": 2740
    },
    {
      "entropy": 5.832088613510132,
      "epoch": 0.3083057224686921,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995643440746982,
      "loss": 5.7578,
      "mean_token_accuracy": 0.1716798350214958,
      "num_tokens": 6126883.0,
      "step": 2745
    },
    {
      "entropy": 6.002315855026245,
      "epoch": 0.30886729937664964,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995618432789548,
      "loss": 5.9508,
      "mean_token_accuracy": 0.15082325488328935,
      "num_tokens": 6139613.0,
      "step": 2750
    },
    {
      "entropy": 6.019475936889648,
      "epoch": 0.3094288762846072,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004995593353330753,
      "loss": 6.0145,
      "mean_token_accuracy": 0.14947013407945633,
      "num_tokens": 6151950.0,
      "step": 2755
    },
    {
      "entropy": 5.9253801822662355,
      "epoch": 0.30999045319256474,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995568202371393,
      "loss": 5.798,
      "mean_token_accuracy": 0.1644776478409767,
      "num_tokens": 6161563.0,
      "step": 2760
    },
    {
      "entropy": 5.855195665359497,
      "epoch": 0.3105520301005223,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995542979912272,
      "loss": 5.8162,
      "mean_token_accuracy": 0.16404712200164795,
      "num_tokens": 6170966.0,
      "step": 2765
    },
    {
      "entropy": 5.913083553314209,
      "epoch": 0.3111136070084798,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995517685954192,
      "loss": 5.888,
      "mean_token_accuracy": 0.15659398585557938,
      "num_tokens": 6182563.0,
      "step": 2770
    },
    {
      "entropy": 6.013282346725464,
      "epoch": 0.31167518391643734,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004995492320497958,
      "loss": 5.9971,
      "mean_token_accuracy": 0.15299334228038788,
      "num_tokens": 6194495.0,
      "step": 2775
    },
    {
      "entropy": 6.030122756958008,
      "epoch": 0.3122367608243949,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995466883544376,
      "loss": 5.895,
      "mean_token_accuracy": 0.1562565177679062,
      "num_tokens": 6206594.0,
      "step": 2780
    },
    {
      "entropy": 5.946122312545777,
      "epoch": 0.31279833773235244,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004995441375094258,
      "loss": 5.9247,
      "mean_token_accuracy": 0.15526296347379684,
      "num_tokens": 6218371.0,
      "step": 2785
    },
    {
      "entropy": 6.001246643066406,
      "epoch": 0.31335991464031,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004995415795148417,
      "loss": 5.8648,
      "mean_token_accuracy": 0.15969813764095306,
      "num_tokens": 6228723.0,
      "step": 2790
    },
    {
      "entropy": 5.9897809505462645,
      "epoch": 0.31392149154826754,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995390143707665,
      "loss": 6.0495,
      "mean_token_accuracy": 0.15685173273086547,
      "num_tokens": 6240535.0,
      "step": 2795
    },
    {
      "entropy": 5.88944444656372,
      "epoch": 0.3144830684562251,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995364420772821,
      "loss": 5.7423,
      "mean_token_accuracy": 0.16546466201543808,
      "num_tokens": 6250643.0,
      "step": 2800
    },
    {
      "entropy": 5.891298627853393,
      "epoch": 0.31504464536418264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995338626344702,
      "loss": 5.8873,
      "mean_token_accuracy": 0.16037205457687378,
      "num_tokens": 6259569.0,
      "step": 2805
    },
    {
      "entropy": 6.019705200195313,
      "epoch": 0.3156062222721402,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499531276042413,
      "loss": 6.0154,
      "mean_token_accuracy": 0.16337641030550004,
      "num_tokens": 6271108.0,
      "step": 2810
    },
    {
      "entropy": 6.061544942855835,
      "epoch": 0.31616779918009774,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995286823011929,
      "loss": 5.8578,
      "mean_token_accuracy": 0.15389910191297532,
      "num_tokens": 6282352.0,
      "step": 2815
    },
    {
      "entropy": 5.868814754486084,
      "epoch": 0.31672937608805524,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004995260814108926,
      "loss": 5.9603,
      "mean_token_accuracy": 0.157982175052166,
      "num_tokens": 6294946.0,
      "step": 2820
    },
    {
      "entropy": 5.909223985671997,
      "epoch": 0.3172909529960128,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995234733715947,
      "loss": 5.7985,
      "mean_token_accuracy": 0.1622636839747429,
      "num_tokens": 6305552.0,
      "step": 2825
    },
    {
      "entropy": 5.955548286437988,
      "epoch": 0.31785252990397034,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004995208581833824,
      "loss": 5.8637,
      "mean_token_accuracy": 0.1647718846797943,
      "num_tokens": 6315815.0,
      "step": 2830
    },
    {
      "entropy": 5.898680114746094,
      "epoch": 0.3184141068119279,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995182358463388,
      "loss": 5.8973,
      "mean_token_accuracy": 0.1575847841799259,
      "num_tokens": 6327384.0,
      "step": 2835
    },
    {
      "entropy": 5.998651695251465,
      "epoch": 0.31897568371988544,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004995156063605475,
      "loss": 5.8787,
      "mean_token_accuracy": 0.15362707823514937,
      "num_tokens": 6338083.0,
      "step": 2840
    },
    {
      "entropy": 6.001203966140747,
      "epoch": 0.319537260627843,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004995129697260922,
      "loss": 5.8944,
      "mean_token_accuracy": 0.16161568313837052,
      "num_tokens": 6349058.0,
      "step": 2845
    },
    {
      "entropy": 5.875576114654541,
      "epoch": 0.32009883753580054,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004995103259430569,
      "loss": 5.9006,
      "mean_token_accuracy": 0.16204904466867448,
      "num_tokens": 6359655.0,
      "step": 2850
    },
    {
      "entropy": 6.017832040786743,
      "epoch": 0.3206604144437581,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995076750115258,
      "loss": 6.027,
      "mean_token_accuracy": 0.15353279262781144,
      "num_tokens": 6372088.0,
      "step": 2855
    },
    {
      "entropy": 5.99152102470398,
      "epoch": 0.32122199135171564,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995050169315832,
      "loss": 5.9362,
      "mean_token_accuracy": 0.15284417122602462,
      "num_tokens": 6382946.0,
      "step": 2860
    },
    {
      "entropy": 6.00714955329895,
      "epoch": 0.32178356825967314,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995023517033138,
      "loss": 5.9206,
      "mean_token_accuracy": 0.15483998730778695,
      "num_tokens": 6392591.0,
      "step": 2865
    },
    {
      "entropy": 5.8733357906341555,
      "epoch": 0.3223451451676307,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994996793268024,
      "loss": 5.8586,
      "mean_token_accuracy": 0.16367374062538148,
      "num_tokens": 6403159.0,
      "step": 2870
    },
    {
      "entropy": 5.974071979522705,
      "epoch": 0.32290672207558824,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994969998021342,
      "loss": 5.88,
      "mean_token_accuracy": 0.15994496941566466,
      "num_tokens": 6414232.0,
      "step": 2875
    },
    {
      "entropy": 5.964216232299805,
      "epoch": 0.3234682989835458,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004994943131293944,
      "loss": 5.8534,
      "mean_token_accuracy": 0.16927893757820128,
      "num_tokens": 6425762.0,
      "step": 2880
    },
    {
      "entropy": 5.935434436798095,
      "epoch": 0.32402987589150334,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994916193086687,
      "loss": 5.9197,
      "mean_token_accuracy": 0.1512233465909958,
      "num_tokens": 6439077.0,
      "step": 2885
    },
    {
      "entropy": 6.02995548248291,
      "epoch": 0.3245914527994609,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994889183400428,
      "loss": 5.9628,
      "mean_token_accuracy": 0.1587122306227684,
      "num_tokens": 6450425.0,
      "step": 2890
    },
    {
      "entropy": 5.9784949779510494,
      "epoch": 0.32515302970741844,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004994862102236025,
      "loss": 5.9359,
      "mean_token_accuracy": 0.15752865448594094,
      "num_tokens": 6461347.0,
      "step": 2895
    },
    {
      "entropy": 6.020418548583985,
      "epoch": 0.325714606615376,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994834949594342,
      "loss": 5.924,
      "mean_token_accuracy": 0.1578624352812767,
      "num_tokens": 6472428.0,
      "step": 2900
    },
    {
      "entropy": 5.879127836227417,
      "epoch": 0.32627618352333354,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994807725476246,
      "loss": 5.8904,
      "mean_token_accuracy": 0.16676969081163406,
      "num_tokens": 6483393.0,
      "step": 2905
    },
    {
      "entropy": 5.935239315032959,
      "epoch": 0.3268377604312911,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994780429882599,
      "loss": 5.8648,
      "mean_token_accuracy": 0.16207789927721022,
      "num_tokens": 6493520.0,
      "step": 2910
    },
    {
      "entropy": 5.898339509963989,
      "epoch": 0.3273993373392486,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004994753062814274,
      "loss": 5.7977,
      "mean_token_accuracy": 0.16786165684461593,
      "num_tokens": 6505282.0,
      "step": 2915
    },
    {
      "entropy": 5.955675268173218,
      "epoch": 0.32796091424720614,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004994725624272139,
      "loss": 5.971,
      "mean_token_accuracy": 0.16191720515489577,
      "num_tokens": 6516808.0,
      "step": 2920
    },
    {
      "entropy": 5.991767835617066,
      "epoch": 0.3285224911551637,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004994698114257071,
      "loss": 5.9314,
      "mean_token_accuracy": 0.1532451778650284,
      "num_tokens": 6529184.0,
      "step": 2925
    },
    {
      "entropy": 5.945623350143433,
      "epoch": 0.32908406806312124,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994670532769944,
      "loss": 5.7807,
      "mean_token_accuracy": 0.1646365851163864,
      "num_tokens": 6539909.0,
      "step": 2930
    },
    {
      "entropy": 5.8691150665283205,
      "epoch": 0.3296456449710788,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994642879811637,
      "loss": 5.8382,
      "mean_token_accuracy": 0.16271614283323288,
      "num_tokens": 6550870.0,
      "step": 2935
    },
    {
      "entropy": 5.8914587020874025,
      "epoch": 0.33020722187903634,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499461515538303,
      "loss": 5.8833,
      "mean_token_accuracy": 0.1637248530983925,
      "num_tokens": 6561784.0,
      "step": 2940
    },
    {
      "entropy": 5.8711829662323,
      "epoch": 0.3307687987869939,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994587359485006,
      "loss": 5.8381,
      "mean_token_accuracy": 0.15968684256076812,
      "num_tokens": 6573422.0,
      "step": 2945
    },
    {
      "entropy": 6.02776026725769,
      "epoch": 0.33133037569495144,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499455949211845,
      "loss": 5.9274,
      "mean_token_accuracy": 0.16189566254615784,
      "num_tokens": 6584565.0,
      "step": 2950
    },
    {
      "entropy": 5.9975940704345705,
      "epoch": 0.331891952602909,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994531553284248,
      "loss": 5.823,
      "mean_token_accuracy": 0.16222766041755676,
      "num_tokens": 6594238.0,
      "step": 2955
    },
    {
      "entropy": 5.858269691467285,
      "epoch": 0.3324535295108665,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994503542983293,
      "loss": 5.8909,
      "mean_token_accuracy": 0.15825014412403107,
      "num_tokens": 6606309.0,
      "step": 2960
    },
    {
      "entropy": 6.004867458343506,
      "epoch": 0.33301510641882404,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004994475461216472,
      "loss": 5.987,
      "mean_token_accuracy": 0.15539744198322297,
      "num_tokens": 6617916.0,
      "step": 2965
    },
    {
      "entropy": 6.02716212272644,
      "epoch": 0.3335766833267816,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994447307984684,
      "loss": 5.8425,
      "mean_token_accuracy": 0.160085965692997,
      "num_tokens": 6627951.0,
      "step": 2970
    },
    {
      "entropy": 5.868495655059815,
      "epoch": 0.33413826023473914,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004994419083288823,
      "loss": 5.9095,
      "mean_token_accuracy": 0.1606067419052124,
      "num_tokens": 6639285.0,
      "step": 2975
    },
    {
      "entropy": 5.955172061920166,
      "epoch": 0.3346998371426967,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994390787129787,
      "loss": 5.7658,
      "mean_token_accuracy": 0.17116498202085495,
      "num_tokens": 6649973.0,
      "step": 2980
    },
    {
      "entropy": 5.75142731666565,
      "epoch": 0.33526141405065424,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004994362419508478,
      "loss": 5.8012,
      "mean_token_accuracy": 0.16292239129543304,
      "num_tokens": 6660625.0,
      "step": 2985
    },
    {
      "entropy": 5.8972403049469,
      "epoch": 0.3358229909586118,
      "grad_norm": 0.875,
      "learning_rate": 0.00049943339804258,
      "loss": 5.8733,
      "mean_token_accuracy": 0.1556589461863041,
      "num_tokens": 6673191.0,
      "step": 2990
    },
    {
      "entropy": 5.964262771606445,
      "epoch": 0.33638456786656934,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994305469882657,
      "loss": 5.8682,
      "mean_token_accuracy": 0.1654741033911705,
      "num_tokens": 6683993.0,
      "step": 2995
    },
    {
      "entropy": 5.850085067749023,
      "epoch": 0.3369461447745269,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994276887879957,
      "loss": 5.8926,
      "mean_token_accuracy": 0.16476777344942092,
      "num_tokens": 6694881.0,
      "step": 3000
    },
    {
      "epoch": 0.3369461447745269,
      "eval_entropy": 5.856326704514774,
      "eval_loss": 5.906490325927734,
      "eval_mean_token_accuracy": 0.1660596932999661,
      "eval_num_tokens": 6694881.0,
      "eval_runtime": 60.6974,
      "eval_samples_per_second": 510.944,
      "eval_steps_per_second": 63.874,
      "step": 3000
    },
    {
      "entropy": 6.009792947769165,
      "epoch": 0.33750772168248444,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994248234418611,
      "loss": 5.8779,
      "mean_token_accuracy": 0.16287034898996353,
      "num_tokens": 6706648.0,
      "step": 3005
    },
    {
      "entropy": 5.946700239181519,
      "epoch": 0.33806929859044194,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499421950949953,
      "loss": 5.8992,
      "mean_token_accuracy": 0.16301757097244263,
      "num_tokens": 6716858.0,
      "step": 3010
    },
    {
      "entropy": 5.883094263076782,
      "epoch": 0.3386308754983995,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499419071312363,
      "loss": 5.8546,
      "mean_token_accuracy": 0.1635548561811447,
      "num_tokens": 6726558.0,
      "step": 3015
    },
    {
      "entropy": 5.962727642059326,
      "epoch": 0.33919245240635704,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994161845291828,
      "loss": 5.855,
      "mean_token_accuracy": 0.15484548956155778,
      "num_tokens": 6737819.0,
      "step": 3020
    },
    {
      "entropy": 5.8366251468658445,
      "epoch": 0.3397540293143146,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994132906005041,
      "loss": 5.8073,
      "mean_token_accuracy": 0.16233350038528443,
      "num_tokens": 6748593.0,
      "step": 3025
    },
    {
      "entropy": 5.829312515258789,
      "epoch": 0.34031560622227214,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004994103895264193,
      "loss": 5.9023,
      "mean_token_accuracy": 0.15177217721939087,
      "num_tokens": 6760631.0,
      "step": 3030
    },
    {
      "entropy": 6.040865039825439,
      "epoch": 0.3408771831302297,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004994074813070207,
      "loss": 5.8109,
      "mean_token_accuracy": 0.1586655080318451,
      "num_tokens": 6772318.0,
      "step": 3035
    },
    {
      "entropy": 5.788278579711914,
      "epoch": 0.34143876003818724,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004994045659424007,
      "loss": 5.8562,
      "mean_token_accuracy": 0.16249461621046066,
      "num_tokens": 6782684.0,
      "step": 3040
    },
    {
      "entropy": 5.948060512542725,
      "epoch": 0.3420003369461448,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004994016434326524,
      "loss": 5.7292,
      "mean_token_accuracy": 0.16573673039674758,
      "num_tokens": 6793895.0,
      "step": 3045
    },
    {
      "entropy": 5.8175944805145265,
      "epoch": 0.34256191385410234,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004993987137778687,
      "loss": 5.7933,
      "mean_token_accuracy": 0.16275266855955123,
      "num_tokens": 6804794.0,
      "step": 3050
    },
    {
      "entropy": 5.914697074890137,
      "epoch": 0.3431234907620599,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499395776978143,
      "loss": 5.9175,
      "mean_token_accuracy": 0.15746153593063356,
      "num_tokens": 6815976.0,
      "step": 3055
    },
    {
      "entropy": 5.967100524902344,
      "epoch": 0.3436850676700174,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993928330335686,
      "loss": 5.8987,
      "mean_token_accuracy": 0.15376825481653214,
      "num_tokens": 6828027.0,
      "step": 3060
    },
    {
      "entropy": 5.991644477844238,
      "epoch": 0.34424664457797494,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004993898819442395,
      "loss": 5.8724,
      "mean_token_accuracy": 0.15755766332149507,
      "num_tokens": 6839195.0,
      "step": 3065
    },
    {
      "entropy": 5.703225994110108,
      "epoch": 0.3448082214859325,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004993869237102494,
      "loss": 5.7587,
      "mean_token_accuracy": 0.1565806671977043,
      "num_tokens": 6851731.0,
      "step": 3070
    },
    {
      "entropy": 5.916368103027343,
      "epoch": 0.34536979839389004,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993839583316927,
      "loss": 5.7981,
      "mean_token_accuracy": 0.16409124583005905,
      "num_tokens": 6862916.0,
      "step": 3075
    },
    {
      "entropy": 6.025053215026856,
      "epoch": 0.3459313753018476,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004993809858086638,
      "loss": 5.9562,
      "mean_token_accuracy": 0.15552304387092591,
      "num_tokens": 6876003.0,
      "step": 3080
    },
    {
      "entropy": 5.906529664993286,
      "epoch": 0.34649295220980514,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993780061412571,
      "loss": 5.946,
      "mean_token_accuracy": 0.15784678012132644,
      "num_tokens": 6886040.0,
      "step": 3085
    },
    {
      "entropy": 5.927360343933105,
      "epoch": 0.3470545291177627,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993750193295678,
      "loss": 5.841,
      "mean_token_accuracy": 0.16224057227373123,
      "num_tokens": 6896758.0,
      "step": 3090
    },
    {
      "entropy": 5.953919267654419,
      "epoch": 0.34761610602572024,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004993720253736909,
      "loss": 5.9075,
      "mean_token_accuracy": 0.1569637894630432,
      "num_tokens": 6907287.0,
      "step": 3095
    },
    {
      "entropy": 5.9171771049499515,
      "epoch": 0.3481776829336778,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993690242737216,
      "loss": 5.9252,
      "mean_token_accuracy": 0.1543675921857357,
      "num_tokens": 6918992.0,
      "step": 3100
    },
    {
      "entropy": 6.060032367706299,
      "epoch": 0.3487392598416353,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004993660160297555,
      "loss": 5.9307,
      "mean_token_accuracy": 0.15680571645498276,
      "num_tokens": 6931493.0,
      "step": 3105
    },
    {
      "entropy": 5.858033657073975,
      "epoch": 0.34930083674959284,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004993630006418885,
      "loss": 5.7713,
      "mean_token_accuracy": 0.1642995670437813,
      "num_tokens": 6941807.0,
      "step": 3110
    },
    {
      "entropy": 5.926855850219726,
      "epoch": 0.3498624136575504,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004993599781102166,
      "loss": 5.9636,
      "mean_token_accuracy": 0.15738050788640975,
      "num_tokens": 6953436.0,
      "step": 3115
    },
    {
      "entropy": 6.154252576828003,
      "epoch": 0.35042399056550794,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993569484348359,
      "loss": 6.0503,
      "mean_token_accuracy": 0.15361388698220252,
      "num_tokens": 6965247.0,
      "step": 3120
    },
    {
      "entropy": 5.882022333145142,
      "epoch": 0.3509855674734655,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004993539116158429,
      "loss": 5.8606,
      "mean_token_accuracy": 0.1635250687599182,
      "num_tokens": 6976605.0,
      "step": 3125
    },
    {
      "entropy": 5.892963695526123,
      "epoch": 0.35154714438142304,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993508676533346,
      "loss": 5.7997,
      "mean_token_accuracy": 0.1708419293165207,
      "num_tokens": 6986205.0,
      "step": 3130
    },
    {
      "entropy": 5.826607275009155,
      "epoch": 0.3521087212893806,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993478165474074,
      "loss": 5.7705,
      "mean_token_accuracy": 0.1646653413772583,
      "num_tokens": 6997699.0,
      "step": 3135
    },
    {
      "entropy": 5.784408044815064,
      "epoch": 0.35267029819733814,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004993447582981588,
      "loss": 5.7973,
      "mean_token_accuracy": 0.15687063187360764,
      "num_tokens": 7008968.0,
      "step": 3140
    },
    {
      "entropy": 6.000455713272094,
      "epoch": 0.3532318751052957,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499341692905686,
      "loss": 5.8224,
      "mean_token_accuracy": 0.1572865605354309,
      "num_tokens": 7020633.0,
      "step": 3145
    },
    {
      "entropy": 5.800573825836182,
      "epoch": 0.35379345201325324,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993386203700868,
      "loss": 5.7457,
      "mean_token_accuracy": 0.1712852880358696,
      "num_tokens": 7031187.0,
      "step": 3150
    },
    {
      "entropy": 5.866760349273681,
      "epoch": 0.35435502892121074,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004993355406914589,
      "loss": 5.9662,
      "mean_token_accuracy": 0.15580730140209198,
      "num_tokens": 7042662.0,
      "step": 3155
    },
    {
      "entropy": 6.075199604034424,
      "epoch": 0.3549166058291683,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004993324538699003,
      "loss": 5.9474,
      "mean_token_accuracy": 0.16323336064815522,
      "num_tokens": 7054297.0,
      "step": 3160
    },
    {
      "entropy": 5.99321403503418,
      "epoch": 0.35547818273712584,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004993293599055094,
      "loss": 5.8053,
      "mean_token_accuracy": 0.1664621964097023,
      "num_tokens": 7064961.0,
      "step": 3165
    },
    {
      "entropy": 5.883630561828613,
      "epoch": 0.3560397596450834,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004993262587983848,
      "loss": 5.9654,
      "mean_token_accuracy": 0.15905895978212356,
      "num_tokens": 7076857.0,
      "step": 3170
    },
    {
      "entropy": 5.92643666267395,
      "epoch": 0.35660133655304094,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499323150548625,
      "loss": 5.8288,
      "mean_token_accuracy": 0.1658812627196312,
      "num_tokens": 7088926.0,
      "step": 3175
    },
    {
      "entropy": 5.85989089012146,
      "epoch": 0.3571629134609985,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993200351563291,
      "loss": 5.7505,
      "mean_token_accuracy": 0.16490727439522743,
      "num_tokens": 7099478.0,
      "step": 3180
    },
    {
      "entropy": 5.790369987487793,
      "epoch": 0.35772449036895604,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004993169126215962,
      "loss": 5.7955,
      "mean_token_accuracy": 0.16942433565855025,
      "num_tokens": 7111776.0,
      "step": 3185
    },
    {
      "entropy": 6.009975481033325,
      "epoch": 0.3582860672769136,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993137829445259,
      "loss": 5.8808,
      "mean_token_accuracy": 0.16010644435882568,
      "num_tokens": 7123084.0,
      "step": 3190
    },
    {
      "entropy": 5.8817391872406,
      "epoch": 0.35884764418487114,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993106461252177,
      "loss": 5.8452,
      "mean_token_accuracy": 0.16542601436376572,
      "num_tokens": 7134973.0,
      "step": 3195
    },
    {
      "entropy": 5.892574882507324,
      "epoch": 0.3594092210928287,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004993075021637716,
      "loss": 5.8804,
      "mean_token_accuracy": 0.15410309433937072,
      "num_tokens": 7147227.0,
      "step": 3200
    },
    {
      "entropy": 5.97821683883667,
      "epoch": 0.3599707980007862,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004993043510602875,
      "loss": 5.864,
      "mean_token_accuracy": 0.16147272139787675,
      "num_tokens": 7158201.0,
      "step": 3205
    },
    {
      "entropy": 5.785186910629273,
      "epoch": 0.36053237490874374,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499301192814866,
      "loss": 5.7684,
      "mean_token_accuracy": 0.1673009842634201,
      "num_tokens": 7168909.0,
      "step": 3210
    },
    {
      "entropy": 5.995763444900513,
      "epoch": 0.3610939518167013,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992980274276074,
      "loss": 5.9293,
      "mean_token_accuracy": 0.1569584757089615,
      "num_tokens": 7179877.0,
      "step": 3215
    },
    {
      "entropy": 5.913835334777832,
      "epoch": 0.36165552872465884,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992948548986128,
      "loss": 5.8438,
      "mean_token_accuracy": 0.16550108194351196,
      "num_tokens": 7190823.0,
      "step": 3220
    },
    {
      "entropy": 5.784553384780883,
      "epoch": 0.3622171056326164,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992916752279829,
      "loss": 5.7717,
      "mean_token_accuracy": 0.16480892449617385,
      "num_tokens": 7201700.0,
      "step": 3225
    },
    {
      "entropy": 5.910831880569458,
      "epoch": 0.36277868254057394,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499288488415819,
      "loss": 5.8078,
      "mean_token_accuracy": 0.1658710226416588,
      "num_tokens": 7213618.0,
      "step": 3230
    },
    {
      "entropy": 6.051048278808594,
      "epoch": 0.3633402594485315,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004992852944622229,
      "loss": 5.9842,
      "mean_token_accuracy": 0.1599672853946686,
      "num_tokens": 7226207.0,
      "step": 3235
    },
    {
      "entropy": 5.871135187149048,
      "epoch": 0.36390183635648904,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004992820933672958,
      "loss": 5.8453,
      "mean_token_accuracy": 0.16316219717264174,
      "num_tokens": 7237545.0,
      "step": 3240
    },
    {
      "entropy": 5.8658225536346436,
      "epoch": 0.3644634132644466,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004992788851311399,
      "loss": 5.8152,
      "mean_token_accuracy": 0.16614115238189697,
      "num_tokens": 7247799.0,
      "step": 3245
    },
    {
      "entropy": 5.896589946746826,
      "epoch": 0.3650249901724041,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004992756697538574,
      "loss": 5.7922,
      "mean_token_accuracy": 0.1614817500114441,
      "num_tokens": 7260229.0,
      "step": 3250
    },
    {
      "entropy": 5.829941320419311,
      "epoch": 0.36558656708036164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992724472355505,
      "loss": 5.9007,
      "mean_token_accuracy": 0.1580626793205738,
      "num_tokens": 7271457.0,
      "step": 3255
    },
    {
      "entropy": 5.933812761306763,
      "epoch": 0.3661481439883192,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992692175763219,
      "loss": 5.8052,
      "mean_token_accuracy": 0.16287915259599686,
      "num_tokens": 7282124.0,
      "step": 3260
    },
    {
      "entropy": 5.927642107009888,
      "epoch": 0.36670972089627674,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992659807762745,
      "loss": 5.9091,
      "mean_token_accuracy": 0.16330223083496093,
      "num_tokens": 7292981.0,
      "step": 3265
    },
    {
      "entropy": 6.152715539932251,
      "epoch": 0.3672712978042343,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992627368355112,
      "loss": 6.0666,
      "mean_token_accuracy": 0.15554819852113724,
      "num_tokens": 7304591.0,
      "step": 3270
    },
    {
      "entropy": 5.956924867630005,
      "epoch": 0.36783287471219184,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992594857541355,
      "loss": 5.797,
      "mean_token_accuracy": 0.1661202237010002,
      "num_tokens": 7315918.0,
      "step": 3275
    },
    {
      "entropy": 5.8168786525726315,
      "epoch": 0.3683944516201494,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004992562275322506,
      "loss": 5.9015,
      "mean_token_accuracy": 0.16148915886878967,
      "num_tokens": 7327725.0,
      "step": 3280
    },
    {
      "entropy": 6.069124317169189,
      "epoch": 0.36895602852810694,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992529621699606,
      "loss": 5.9152,
      "mean_token_accuracy": 0.15593092143535614,
      "num_tokens": 7339787.0,
      "step": 3285
    },
    {
      "entropy": 5.9365582942962645,
      "epoch": 0.3695176054360645,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992496896673692,
      "loss": 5.8939,
      "mean_token_accuracy": 0.16256406754255295,
      "num_tokens": 7351284.0,
      "step": 3290
    },
    {
      "entropy": 5.742411994934082,
      "epoch": 0.37007918234402204,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992464100245809,
      "loss": 5.7132,
      "mean_token_accuracy": 0.1695790097117424,
      "num_tokens": 7362105.0,
      "step": 3295
    },
    {
      "entropy": 5.893797731399536,
      "epoch": 0.37064075925197953,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004992431232416997,
      "loss": 5.816,
      "mean_token_accuracy": 0.16402172893285752,
      "num_tokens": 7375204.0,
      "step": 3300
    },
    {
      "entropy": 6.095653533935547,
      "epoch": 0.3712023361599371,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992398293188307,
      "loss": 5.98,
      "mean_token_accuracy": 0.1561136357486248,
      "num_tokens": 7385853.0,
      "step": 3305
    },
    {
      "entropy": 5.926312065124511,
      "epoch": 0.37176391306789464,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992365282560784,
      "loss": 5.8333,
      "mean_token_accuracy": 0.1528450667858124,
      "num_tokens": 7397459.0,
      "step": 3310
    },
    {
      "entropy": 5.815288496017456,
      "epoch": 0.3723254899758522,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992332200535483,
      "loss": 5.8029,
      "mean_token_accuracy": 0.15823303014039994,
      "num_tokens": 7408996.0,
      "step": 3315
    },
    {
      "entropy": 5.959876823425293,
      "epoch": 0.37288706688380974,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004992299047113452,
      "loss": 5.9063,
      "mean_token_accuracy": 0.15332845151424407,
      "num_tokens": 7419938.0,
      "step": 3320
    },
    {
      "entropy": 5.944911193847656,
      "epoch": 0.3734486437917673,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004992265822295752,
      "loss": 5.912,
      "mean_token_accuracy": 0.1548202320933342,
      "num_tokens": 7430181.0,
      "step": 3325
    },
    {
      "entropy": 5.948005962371826,
      "epoch": 0.37401022069972484,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992232526083438,
      "loss": 5.8674,
      "mean_token_accuracy": 0.15860404074192047,
      "num_tokens": 7440774.0,
      "step": 3330
    },
    {
      "entropy": 5.933913946151733,
      "epoch": 0.3745717976076824,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499219915847757,
      "loss": 5.7852,
      "mean_token_accuracy": 0.16459606140851973,
      "num_tokens": 7451744.0,
      "step": 3335
    },
    {
      "entropy": 5.72104172706604,
      "epoch": 0.37513337451563994,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004992165719479211,
      "loss": 5.8251,
      "mean_token_accuracy": 0.16036779880523683,
      "num_tokens": 7463731.0,
      "step": 3340
    },
    {
      "entropy": 5.943351984024048,
      "epoch": 0.37569495142359743,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004992132209089427,
      "loss": 5.8207,
      "mean_token_accuracy": 0.16885581910610198,
      "num_tokens": 7474392.0,
      "step": 3345
    },
    {
      "entropy": 5.8482115268707275,
      "epoch": 0.376256528331555,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992098627309283,
      "loss": 5.7597,
      "mean_token_accuracy": 0.17417206317186357,
      "num_tokens": 7484860.0,
      "step": 3350
    },
    {
      "entropy": 5.800925064086914,
      "epoch": 0.37681810523951254,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499206497413985,
      "loss": 5.8007,
      "mean_token_accuracy": 0.16721806675195694,
      "num_tokens": 7495877.0,
      "step": 3355
    },
    {
      "entropy": 5.954871892929077,
      "epoch": 0.3773796821474701,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992031249582197,
      "loss": 5.8626,
      "mean_token_accuracy": 0.1659214660525322,
      "num_tokens": 7506879.0,
      "step": 3360
    },
    {
      "entropy": 5.829053115844727,
      "epoch": 0.37794125905542764,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991997453637401,
      "loss": 5.7166,
      "mean_token_accuracy": 0.16946493536233903,
      "num_tokens": 7517405.0,
      "step": 3365
    },
    {
      "entropy": 5.846356964111328,
      "epoch": 0.3785028359633852,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004991963586306536,
      "loss": 5.7839,
      "mean_token_accuracy": 0.16438688188791276,
      "num_tokens": 7528832.0,
      "step": 3370
    },
    {
      "entropy": 5.828200101852417,
      "epoch": 0.37906441287134274,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004991929647590682,
      "loss": 5.7811,
      "mean_token_accuracy": 0.15837691128253936,
      "num_tokens": 7541048.0,
      "step": 3375
    },
    {
      "entropy": 5.882894372940063,
      "epoch": 0.3796259897793003,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991895637490917,
      "loss": 5.7892,
      "mean_token_accuracy": 0.16829562336206436,
      "num_tokens": 7551849.0,
      "step": 3380
    },
    {
      "entropy": 5.801930713653564,
      "epoch": 0.38018756668725784,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991861556008326,
      "loss": 5.8084,
      "mean_token_accuracy": 0.16716636419296266,
      "num_tokens": 7563148.0,
      "step": 3385
    },
    {
      "entropy": 6.005253744125366,
      "epoch": 0.3807491435952154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991827403143992,
      "loss": 5.9857,
      "mean_token_accuracy": 0.15619939416646958,
      "num_tokens": 7575119.0,
      "step": 3390
    },
    {
      "entropy": 5.920757913589478,
      "epoch": 0.3813107205031729,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004991793178899006,
      "loss": 5.8477,
      "mean_token_accuracy": 0.16746386140584946,
      "num_tokens": 7585899.0,
      "step": 3395
    },
    {
      "entropy": 5.927013492584228,
      "epoch": 0.38187229741113043,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004991758883274455,
      "loss": 5.7408,
      "mean_token_accuracy": 0.17119719982147216,
      "num_tokens": 7598938.0,
      "step": 3400
    },
    {
      "entropy": 5.718728494644165,
      "epoch": 0.382433874319088,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991724516271432,
      "loss": 5.8176,
      "mean_token_accuracy": 0.15823707431554795,
      "num_tokens": 7610087.0,
      "step": 3405
    },
    {
      "entropy": 6.031653833389282,
      "epoch": 0.38299545122704554,
      "grad_norm": 1.0,
      "learning_rate": 0.0004991690077891032,
      "loss": 5.9304,
      "mean_token_accuracy": 0.15817293301224708,
      "num_tokens": 7622186.0,
      "step": 3410
    },
    {
      "entropy": 5.95524001121521,
      "epoch": 0.3835570281350031,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004991655568134348,
      "loss": 5.8724,
      "mean_token_accuracy": 0.15889675095677375,
      "num_tokens": 7633501.0,
      "step": 3415
    },
    {
      "entropy": 5.811369943618774,
      "epoch": 0.38411860504296064,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991620987002483,
      "loss": 5.8081,
      "mean_token_accuracy": 0.1626387983560562,
      "num_tokens": 7643656.0,
      "step": 3420
    },
    {
      "entropy": 5.975064659118653,
      "epoch": 0.3846801819509182,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004991586334496536,
      "loss": 5.8101,
      "mean_token_accuracy": 0.16192732900381088,
      "num_tokens": 7654565.0,
      "step": 3425
    },
    {
      "entropy": 5.829861116409302,
      "epoch": 0.38524175885887574,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991551610617611,
      "loss": 5.7368,
      "mean_token_accuracy": 0.17527267187833787,
      "num_tokens": 7664725.0,
      "step": 3430
    },
    {
      "entropy": 5.78968071937561,
      "epoch": 0.3858033357668333,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991516815366812,
      "loss": 5.8136,
      "mean_token_accuracy": 0.1644575446844101,
      "num_tokens": 7674772.0,
      "step": 3435
    },
    {
      "entropy": 5.963439416885376,
      "epoch": 0.38636491267479084,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499148194874525,
      "loss": 5.8914,
      "mean_token_accuracy": 0.15519467890262603,
      "num_tokens": 7686488.0,
      "step": 3440
    },
    {
      "entropy": 5.9735828876495365,
      "epoch": 0.38692648958274833,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991447010754032,
      "loss": 5.8796,
      "mean_token_accuracy": 0.16299588978290558,
      "num_tokens": 7698842.0,
      "step": 3445
    },
    {
      "entropy": 5.723750066757202,
      "epoch": 0.3874880664907059,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991412001394273,
      "loss": 5.6652,
      "mean_token_accuracy": 0.16502486169338226,
      "num_tokens": 7709255.0,
      "step": 3450
    },
    {
      "entropy": 5.712912368774414,
      "epoch": 0.38804964339866344,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991376920667086,
      "loss": 5.7061,
      "mean_token_accuracy": 0.1652544140815735,
      "num_tokens": 7720182.0,
      "step": 3455
    },
    {
      "entropy": 5.820716857910156,
      "epoch": 0.388611220306621,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991341768573589,
      "loss": 5.7476,
      "mean_token_accuracy": 0.17566618472337722,
      "num_tokens": 7730944.0,
      "step": 3460
    },
    {
      "entropy": 5.849359560012817,
      "epoch": 0.38917279721457854,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00049913065451149,
      "loss": 5.8219,
      "mean_token_accuracy": 0.16147851198911667,
      "num_tokens": 7741686.0,
      "step": 3465
    },
    {
      "entropy": 5.922025728225708,
      "epoch": 0.3897343741225361,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004991271250292141,
      "loss": 5.807,
      "mean_token_accuracy": 0.16342652589082718,
      "num_tokens": 7752245.0,
      "step": 3470
    },
    {
      "entropy": 5.757428932189941,
      "epoch": 0.39029595103049364,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004991235884106437,
      "loss": 5.7788,
      "mean_token_accuracy": 0.159398677945137,
      "num_tokens": 7762622.0,
      "step": 3475
    },
    {
      "entropy": 5.963356447219849,
      "epoch": 0.3908575279384512,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004991200446558913,
      "loss": 5.9069,
      "mean_token_accuracy": 0.16293678879737855,
      "num_tokens": 7773550.0,
      "step": 3480
    },
    {
      "entropy": 5.894080686569214,
      "epoch": 0.39141910484640874,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004991164937650698,
      "loss": 5.789,
      "mean_token_accuracy": 0.16059745252132415,
      "num_tokens": 7784988.0,
      "step": 3485
    },
    {
      "entropy": 5.940509033203125,
      "epoch": 0.39198068175436623,
      "grad_norm": 0.875,
      "learning_rate": 0.0004991129357382922,
      "loss": 5.8767,
      "mean_token_accuracy": 0.16270290464162826,
      "num_tokens": 7797706.0,
      "step": 3490
    },
    {
      "entropy": 5.870716524124146,
      "epoch": 0.3925422586623238,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004991093705756718,
      "loss": 5.7541,
      "mean_token_accuracy": 0.1708513543009758,
      "num_tokens": 7808960.0,
      "step": 3495
    },
    {
      "entropy": 5.8290143489837645,
      "epoch": 0.39310383557028133,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004991057982773221,
      "loss": 5.7225,
      "mean_token_accuracy": 0.17448564767837524,
      "num_tokens": 7819271.0,
      "step": 3500
    },
    {
      "entropy": 5.827299976348877,
      "epoch": 0.3936654124782389,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004991022188433569,
      "loss": 5.8587,
      "mean_token_accuracy": 0.16258958131074905,
      "num_tokens": 7832257.0,
      "step": 3505
    },
    {
      "entropy": 5.969977903366089,
      "epoch": 0.39422698938619644,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00049909863227389,
      "loss": 5.8743,
      "mean_token_accuracy": 0.16018556654453278,
      "num_tokens": 7842684.0,
      "step": 3510
    },
    {
      "entropy": 5.831171321868896,
      "epoch": 0.394788566294154,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004990950385690358,
      "loss": 5.878,
      "mean_token_accuracy": 0.1610353022813797,
      "num_tokens": 7855031.0,
      "step": 3515
    },
    {
      "entropy": 5.840944480895996,
      "epoch": 0.39535014320211154,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990914377289088,
      "loss": 5.6858,
      "mean_token_accuracy": 0.16333473473787308,
      "num_tokens": 7865291.0,
      "step": 3520
    },
    {
      "entropy": 5.744883012771607,
      "epoch": 0.3959117201100691,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990878297536233,
      "loss": 5.6451,
      "mean_token_accuracy": 0.16607558876276016,
      "num_tokens": 7876401.0,
      "step": 3525
    },
    {
      "entropy": 5.718347787857056,
      "epoch": 0.39647329701802664,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990842146432946,
      "loss": 5.6944,
      "mean_token_accuracy": 0.16728755831718445,
      "num_tokens": 7887787.0,
      "step": 3530
    },
    {
      "entropy": 5.763777160644532,
      "epoch": 0.3970348739259842,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990805923980375,
      "loss": 5.6914,
      "mean_token_accuracy": 0.16545438915491104,
      "num_tokens": 7898550.0,
      "step": 3535
    },
    {
      "entropy": 5.8369975090026855,
      "epoch": 0.3975964508339417,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990769630179673,
      "loss": 5.8377,
      "mean_token_accuracy": 0.16232438683509826,
      "num_tokens": 7909254.0,
      "step": 3540
    },
    {
      "entropy": 6.022816181182861,
      "epoch": 0.39815802774189923,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990733265032,
      "loss": 5.8797,
      "mean_token_accuracy": 0.16756701320409775,
      "num_tokens": 7920164.0,
      "step": 3545
    },
    {
      "entropy": 5.7947046756744385,
      "epoch": 0.3987196046498568,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990696828538508,
      "loss": 5.7971,
      "mean_token_accuracy": 0.16098637133836746,
      "num_tokens": 7931390.0,
      "step": 3550
    },
    {
      "entropy": 5.927447414398193,
      "epoch": 0.39928118155781434,
      "grad_norm": 0.890625,
      "learning_rate": 0.000499066032070036,
      "loss": 5.8199,
      "mean_token_accuracy": 0.16026633083820344,
      "num_tokens": 7943765.0,
      "step": 3555
    },
    {
      "entropy": 5.817334890365601,
      "epoch": 0.3998427584657719,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990623741518719,
      "loss": 5.6543,
      "mean_token_accuracy": 0.16805292963981627,
      "num_tokens": 7954621.0,
      "step": 3560
    },
    {
      "entropy": 5.7308307647705075,
      "epoch": 0.40040433537372944,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499058709099475,
      "loss": 5.8617,
      "mean_token_accuracy": 0.16110717505216599,
      "num_tokens": 7966063.0,
      "step": 3565
    },
    {
      "entropy": 5.984078693389892,
      "epoch": 0.400965912281687,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990550369129618,
      "loss": 5.7822,
      "mean_token_accuracy": 0.16557336151599883,
      "num_tokens": 7977391.0,
      "step": 3570
    },
    {
      "entropy": 5.949883031845093,
      "epoch": 0.40152748918964454,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990513575924494,
      "loss": 5.9017,
      "mean_token_accuracy": 0.1615771010518074,
      "num_tokens": 7988786.0,
      "step": 3575
    },
    {
      "entropy": 5.823601341247558,
      "epoch": 0.4020890660976021,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004990476711380548,
      "loss": 5.7066,
      "mean_token_accuracy": 0.16304796189069748,
      "num_tokens": 8000848.0,
      "step": 3580
    },
    {
      "entropy": 5.806692171096802,
      "epoch": 0.40265064300555964,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990439775498954,
      "loss": 5.8641,
      "mean_token_accuracy": 0.16117497086524962,
      "num_tokens": 8012621.0,
      "step": 3585
    },
    {
      "entropy": 6.056632232666016,
      "epoch": 0.40321221991351713,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499040276828089,
      "loss": 5.9032,
      "mean_token_accuracy": 0.16068349182605743,
      "num_tokens": 8023599.0,
      "step": 3590
    },
    {
      "entropy": 5.734525108337403,
      "epoch": 0.4037737968214747,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990365689727533,
      "loss": 5.7073,
      "mean_token_accuracy": 0.16968922764062883,
      "num_tokens": 8037151.0,
      "step": 3595
    },
    {
      "entropy": 5.866984462738037,
      "epoch": 0.40433537372943223,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004990328539840062,
      "loss": 5.9032,
      "mean_token_accuracy": 0.15894825309514998,
      "num_tokens": 8048606.0,
      "step": 3600
    },
    {
      "entropy": 5.814764881134034,
      "epoch": 0.4048969506373898,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990291318619663,
      "loss": 5.6844,
      "mean_token_accuracy": 0.1796335518360138,
      "num_tokens": 8058430.0,
      "step": 3605
    },
    {
      "entropy": 5.724921464920044,
      "epoch": 0.40545852754534734,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990254026067518,
      "loss": 5.7002,
      "mean_token_accuracy": 0.16601582020521163,
      "num_tokens": 8069554.0,
      "step": 3610
    },
    {
      "entropy": 5.946437072753906,
      "epoch": 0.4060201044533049,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004990216662184817,
      "loss": 5.8534,
      "mean_token_accuracy": 0.15956933349370955,
      "num_tokens": 8082131.0,
      "step": 3615
    },
    {
      "entropy": 5.800406789779663,
      "epoch": 0.40658168136126244,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990179226972749,
      "loss": 5.6988,
      "mean_token_accuracy": 0.16990892142057418,
      "num_tokens": 8093454.0,
      "step": 3620
    },
    {
      "entropy": 5.98328971862793,
      "epoch": 0.40714325826922,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990141720432504,
      "loss": 5.9571,
      "mean_token_accuracy": 0.1504925012588501,
      "num_tokens": 8105502.0,
      "step": 3625
    },
    {
      "entropy": 5.859578943252563,
      "epoch": 0.40770483517717754,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990104142565279,
      "loss": 5.8145,
      "mean_token_accuracy": 0.1626613110303879,
      "num_tokens": 8116524.0,
      "step": 3630
    },
    {
      "entropy": 5.894063568115234,
      "epoch": 0.40826641208513503,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499006649337227,
      "loss": 5.7959,
      "mean_token_accuracy": 0.17154594510793686,
      "num_tokens": 8126949.0,
      "step": 3635
    },
    {
      "entropy": 5.830201005935669,
      "epoch": 0.4088279889930926,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004990028772854674,
      "loss": 5.763,
      "mean_token_accuracy": 0.16976310908794404,
      "num_tokens": 8137346.0,
      "step": 3640
    },
    {
      "entropy": 5.788245964050293,
      "epoch": 0.40938956590105013,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989990981013693,
      "loss": 5.7978,
      "mean_token_accuracy": 0.16071729511022567,
      "num_tokens": 8148499.0,
      "step": 3645
    },
    {
      "entropy": 5.900784349441528,
      "epoch": 0.4099511428090077,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498995311785053,
      "loss": 5.8276,
      "mean_token_accuracy": 0.16623455286026,
      "num_tokens": 8159645.0,
      "step": 3650
    },
    {
      "entropy": 5.830116033554077,
      "epoch": 0.41051271971696524,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004989915183366392,
      "loss": 5.6977,
      "mean_token_accuracy": 0.16828417778015137,
      "num_tokens": 8171033.0,
      "step": 3655
    },
    {
      "entropy": 5.790131378173828,
      "epoch": 0.4110742966249228,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989877177562485,
      "loss": 5.792,
      "mean_token_accuracy": 0.16119424402713775,
      "num_tokens": 8181582.0,
      "step": 3660
    },
    {
      "entropy": 5.773313522338867,
      "epoch": 0.41163587353288034,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498983910044002,
      "loss": 5.6858,
      "mean_token_accuracy": 0.16789519041776657,
      "num_tokens": 8192432.0,
      "step": 3665
    },
    {
      "entropy": 5.975875806808472,
      "epoch": 0.4121974504408379,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989800952000209,
      "loss": 5.867,
      "mean_token_accuracy": 0.16418623179197311,
      "num_tokens": 8203995.0,
      "step": 3670
    },
    {
      "entropy": 5.781466293334961,
      "epoch": 0.41275902734879544,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989762732244268,
      "loss": 5.7005,
      "mean_token_accuracy": 0.16910858303308487,
      "num_tokens": 8214323.0,
      "step": 3675
    },
    {
      "entropy": 5.71599268913269,
      "epoch": 0.413320604256753,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989724441173413,
      "loss": 5.7628,
      "mean_token_accuracy": 0.15963411182165146,
      "num_tokens": 8226285.0,
      "step": 3680
    },
    {
      "entropy": 6.009037590026855,
      "epoch": 0.4138821811647105,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989686078788862,
      "loss": 5.9395,
      "mean_token_accuracy": 0.15572211444377898,
      "num_tokens": 8237961.0,
      "step": 3685
    },
    {
      "entropy": 5.771542167663574,
      "epoch": 0.41444375807266803,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004989647645091839,
      "loss": 5.6755,
      "mean_token_accuracy": 0.17040250152349473,
      "num_tokens": 8248463.0,
      "step": 3690
    },
    {
      "entropy": 5.865232706069946,
      "epoch": 0.4150053349806256,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989609140083565,
      "loss": 5.815,
      "mean_token_accuracy": 0.16708862632513047,
      "num_tokens": 8260342.0,
      "step": 3695
    },
    {
      "entropy": 5.873272800445557,
      "epoch": 0.41556691188858313,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004989570563765269,
      "loss": 5.9115,
      "mean_token_accuracy": 0.16665390282869338,
      "num_tokens": 8272055.0,
      "step": 3700
    },
    {
      "entropy": 6.020225572586059,
      "epoch": 0.4161284887965407,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989531916138176,
      "loss": 5.876,
      "mean_token_accuracy": 0.16337255090475084,
      "num_tokens": 8282129.0,
      "step": 3705
    },
    {
      "entropy": 5.9397468090057375,
      "epoch": 0.41669006570449824,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004989493197203519,
      "loss": 5.8663,
      "mean_token_accuracy": 0.16446712464094163,
      "num_tokens": 8294663.0,
      "step": 3710
    },
    {
      "entropy": 5.813158178329468,
      "epoch": 0.4172516426124558,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000498945440696253,
      "loss": 5.6755,
      "mean_token_accuracy": 0.17049953639507293,
      "num_tokens": 8305882.0,
      "step": 3715
    },
    {
      "entropy": 5.772831439971924,
      "epoch": 0.41781321952041334,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989415545416444,
      "loss": 5.7867,
      "mean_token_accuracy": 0.16710352897644043,
      "num_tokens": 8316604.0,
      "step": 3720
    },
    {
      "entropy": 5.902248382568359,
      "epoch": 0.4183747964283709,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004989376612566498,
      "loss": 5.8747,
      "mean_token_accuracy": 0.16121989041566848,
      "num_tokens": 8328126.0,
      "step": 3725
    },
    {
      "entropy": 5.865182256698608,
      "epoch": 0.4189363733363284,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989337608413933,
      "loss": 5.799,
      "mean_token_accuracy": 0.1652168944478035,
      "num_tokens": 8339050.0,
      "step": 3730
    },
    {
      "entropy": 5.732286357879639,
      "epoch": 0.41949795024428593,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989298532959989,
      "loss": 5.6212,
      "mean_token_accuracy": 0.17028117328882217,
      "num_tokens": 8350216.0,
      "step": 3735
    },
    {
      "entropy": 5.8672754764556885,
      "epoch": 0.4200595271522435,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004989259386205912,
      "loss": 5.7858,
      "mean_token_accuracy": 0.16723376661539077,
      "num_tokens": 8362311.0,
      "step": 3740
    },
    {
      "entropy": 5.879049682617188,
      "epoch": 0.42062110406020103,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989220168152948,
      "loss": 5.9504,
      "mean_token_accuracy": 0.16739826500415803,
      "num_tokens": 8373751.0,
      "step": 3745
    },
    {
      "entropy": 5.871591663360595,
      "epoch": 0.4211826809681586,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989180878802346,
      "loss": 5.7839,
      "mean_token_accuracy": 0.1667262703180313,
      "num_tokens": 8384932.0,
      "step": 3750
    },
    {
      "entropy": 5.870432186126709,
      "epoch": 0.42174425787611614,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989141518155354,
      "loss": 5.7206,
      "mean_token_accuracy": 0.172009739279747,
      "num_tokens": 8394973.0,
      "step": 3755
    },
    {
      "entropy": 5.825662612915039,
      "epoch": 0.4223058347840737,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004989102086213229,
      "loss": 5.8181,
      "mean_token_accuracy": 0.1665082648396492,
      "num_tokens": 8406590.0,
      "step": 3760
    },
    {
      "entropy": 5.801623344421387,
      "epoch": 0.42286741169203124,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004989062582977225,
      "loss": 5.7343,
      "mean_token_accuracy": 0.1675838053226471,
      "num_tokens": 8417127.0,
      "step": 3765
    },
    {
      "entropy": 5.918402051925659,
      "epoch": 0.4234289885999888,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989023008448599,
      "loss": 5.8453,
      "mean_token_accuracy": 0.1659783571958542,
      "num_tokens": 8429442.0,
      "step": 3770
    },
    {
      "entropy": 5.811355018615723,
      "epoch": 0.42399056550794634,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004988983362628614,
      "loss": 5.7607,
      "mean_token_accuracy": 0.16928921788930892,
      "num_tokens": 8439375.0,
      "step": 3775
    },
    {
      "entropy": 5.832507801055908,
      "epoch": 0.42455214241590383,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004988943645518528,
      "loss": 5.7695,
      "mean_token_accuracy": 0.1644688591361046,
      "num_tokens": 8449817.0,
      "step": 3780
    },
    {
      "entropy": 5.78357048034668,
      "epoch": 0.4251137193238614,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988903857119611,
      "loss": 5.653,
      "mean_token_accuracy": 0.17269241362810134,
      "num_tokens": 8460096.0,
      "step": 3785
    },
    {
      "entropy": 5.856718921661377,
      "epoch": 0.42567529623181893,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988863997433125,
      "loss": 5.783,
      "mean_token_accuracy": 0.1635545700788498,
      "num_tokens": 8470241.0,
      "step": 3790
    },
    {
      "entropy": 5.866709804534912,
      "epoch": 0.4262368731397765,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004988824066460341,
      "loss": 5.7959,
      "mean_token_accuracy": 0.15800238102674485,
      "num_tokens": 8481438.0,
      "step": 3795
    },
    {
      "entropy": 5.85970401763916,
      "epoch": 0.42679845004773403,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498878406420253,
      "loss": 5.8495,
      "mean_token_accuracy": 0.16143330335617065,
      "num_tokens": 8492588.0,
      "step": 3800
    },
    {
      "entropy": 5.779029798507691,
      "epoch": 0.4273600269556916,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988743990660967,
      "loss": 5.7801,
      "mean_token_accuracy": 0.1666340261697769,
      "num_tokens": 8502298.0,
      "step": 3805
    },
    {
      "entropy": 5.804246425628662,
      "epoch": 0.42792160386364914,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988703845836927,
      "loss": 5.6317,
      "mean_token_accuracy": 0.1682993531227112,
      "num_tokens": 8513507.0,
      "step": 3810
    },
    {
      "entropy": 5.814224147796631,
      "epoch": 0.4284831807716067,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988663629731687,
      "loss": 5.7407,
      "mean_token_accuracy": 0.17205402106046677,
      "num_tokens": 8523888.0,
      "step": 3815
    },
    {
      "entropy": 5.748360204696655,
      "epoch": 0.42904475767956424,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498862334234653,
      "loss": 5.6829,
      "mean_token_accuracy": 0.17145948857069016,
      "num_tokens": 8534032.0,
      "step": 3820
    },
    {
      "entropy": 5.761185073852539,
      "epoch": 0.4296063345875218,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988582983682737,
      "loss": 5.819,
      "mean_token_accuracy": 0.16719216257333755,
      "num_tokens": 8545544.0,
      "step": 3825
    },
    {
      "entropy": 5.888425254821778,
      "epoch": 0.4301679114954793,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988542553741594,
      "loss": 5.7701,
      "mean_token_accuracy": 0.167030468583107,
      "num_tokens": 8555428.0,
      "step": 3830
    },
    {
      "entropy": 5.7430023670196535,
      "epoch": 0.43072948840343683,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004988502052524388,
      "loss": 5.6721,
      "mean_token_accuracy": 0.16957473009824753,
      "num_tokens": 8566456.0,
      "step": 3835
    },
    {
      "entropy": 5.826628875732422,
      "epoch": 0.4312910653113944,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004988461480032408,
      "loss": 5.6896,
      "mean_token_accuracy": 0.1679348900914192,
      "num_tokens": 8578125.0,
      "step": 3840
    },
    {
      "entropy": 5.783590936660767,
      "epoch": 0.43185264221935193,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004988420836266946,
      "loss": 5.7499,
      "mean_token_accuracy": 0.1610526993870735,
      "num_tokens": 8589530.0,
      "step": 3845
    },
    {
      "entropy": 5.828071880340576,
      "epoch": 0.4324142191273095,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004988380121229298,
      "loss": 5.8504,
      "mean_token_accuracy": 0.16466037034988404,
      "num_tokens": 8601655.0,
      "step": 3850
    },
    {
      "entropy": 5.8106616020202635,
      "epoch": 0.43297579603526704,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988339334920757,
      "loss": 5.763,
      "mean_token_accuracy": 0.1670367658138275,
      "num_tokens": 8613355.0,
      "step": 3855
    },
    {
      "entropy": 5.823573637008667,
      "epoch": 0.4335373729432246,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004988298477342625,
      "loss": 5.7118,
      "mean_token_accuracy": 0.16885844767093658,
      "num_tokens": 8624154.0,
      "step": 3860
    },
    {
      "entropy": 5.815149307250977,
      "epoch": 0.43409894985118214,
      "grad_norm": 0.953125,
      "learning_rate": 0.00049882575484962,
      "loss": 5.7858,
      "mean_token_accuracy": 0.16578275412321092,
      "num_tokens": 8633888.0,
      "step": 3865
    },
    {
      "entropy": 5.878628587722778,
      "epoch": 0.4346605267591397,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004988216548382787,
      "loss": 5.7605,
      "mean_token_accuracy": 0.16726396083831788,
      "num_tokens": 8644965.0,
      "step": 3870
    },
    {
      "entropy": 5.801642656326294,
      "epoch": 0.4352221036670972,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004988175477003691,
      "loss": 5.7145,
      "mean_token_accuracy": 0.17267416566610336,
      "num_tokens": 8655483.0,
      "step": 3875
    },
    {
      "entropy": 5.81993408203125,
      "epoch": 0.43578368057505473,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498813433436022,
      "loss": 5.6916,
      "mean_token_accuracy": 0.16909216344356537,
      "num_tokens": 8666443.0,
      "step": 3880
    },
    {
      "entropy": 5.816588592529297,
      "epoch": 0.4363452574830123,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988093120453684,
      "loss": 5.7545,
      "mean_token_accuracy": 0.17053562998771668,
      "num_tokens": 8677198.0,
      "step": 3885
    },
    {
      "entropy": 5.756594896316528,
      "epoch": 0.43690683439096983,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988051835285394,
      "loss": 5.7977,
      "mean_token_accuracy": 0.16542441099882127,
      "num_tokens": 8688303.0,
      "step": 3890
    },
    {
      "entropy": 5.910586500167847,
      "epoch": 0.4374684112989274,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004988010478856667,
      "loss": 5.7561,
      "mean_token_accuracy": 0.16605307906866074,
      "num_tokens": 8700326.0,
      "step": 3895
    },
    {
      "entropy": 5.762407350540161,
      "epoch": 0.43802998820688493,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004987969051168816,
      "loss": 5.721,
      "mean_token_accuracy": 0.1687159538269043,
      "num_tokens": 8711101.0,
      "step": 3900
    },
    {
      "entropy": 5.769402265548706,
      "epoch": 0.4385915651148425,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987927552223165,
      "loss": 5.8423,
      "mean_token_accuracy": 0.16267140656709672,
      "num_tokens": 8721751.0,
      "step": 3905
    },
    {
      "entropy": 5.954661417007446,
      "epoch": 0.43915314202280004,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987885982021032,
      "loss": 5.9352,
      "mean_token_accuracy": 0.15480559915304185,
      "num_tokens": 8733550.0,
      "step": 3910
    },
    {
      "entropy": 5.869749975204468,
      "epoch": 0.4397147189307576,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987844340563741,
      "loss": 5.7293,
      "mean_token_accuracy": 0.16600938737392426,
      "num_tokens": 8744218.0,
      "step": 3915
    },
    {
      "entropy": 5.753007507324218,
      "epoch": 0.44027629583871514,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004987802627852619,
      "loss": 5.7763,
      "mean_token_accuracy": 0.16374737173318862,
      "num_tokens": 8756531.0,
      "step": 3920
    },
    {
      "entropy": 6.002922201156617,
      "epoch": 0.44083787274667263,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987760843888992,
      "loss": 5.8171,
      "mean_token_accuracy": 0.16575318574905396,
      "num_tokens": 8768114.0,
      "step": 3925
    },
    {
      "entropy": 5.745939636230469,
      "epoch": 0.4413994496546302,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987718988674193,
      "loss": 5.805,
      "mean_token_accuracy": 0.1609507381916046,
      "num_tokens": 8780010.0,
      "step": 3930
    },
    {
      "entropy": 5.886521244049073,
      "epoch": 0.44196102656258773,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987677062209552,
      "loss": 5.7653,
      "mean_token_accuracy": 0.16114584356546402,
      "num_tokens": 8792183.0,
      "step": 3935
    },
    {
      "entropy": 5.888902807235718,
      "epoch": 0.4425226034705453,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004987635064496407,
      "loss": 5.8566,
      "mean_token_accuracy": 0.16036604791879655,
      "num_tokens": 8804513.0,
      "step": 3940
    },
    {
      "entropy": 5.777865171432495,
      "epoch": 0.44308418037850283,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987592995536093,
      "loss": 5.7553,
      "mean_token_accuracy": 0.17000913769006729,
      "num_tokens": 8814918.0,
      "step": 3945
    },
    {
      "entropy": 5.8164514064788815,
      "epoch": 0.4436457572864604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987550855329951,
      "loss": 5.704,
      "mean_token_accuracy": 0.16863586157560348,
      "num_tokens": 8826176.0,
      "step": 3950
    },
    {
      "entropy": 5.743463325500488,
      "epoch": 0.44420733419441794,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987508643879322,
      "loss": 5.7513,
      "mean_token_accuracy": 0.16684129387140273,
      "num_tokens": 8837242.0,
      "step": 3955
    },
    {
      "entropy": 5.9041355609893795,
      "epoch": 0.4447689111023755,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004987466361185549,
      "loss": 5.7139,
      "mean_token_accuracy": 0.1682710364460945,
      "num_tokens": 8848972.0,
      "step": 3960
    },
    {
      "entropy": 5.703074598312378,
      "epoch": 0.44533048801033304,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004987424007249979,
      "loss": 5.674,
      "mean_token_accuracy": 0.1637897565960884,
      "num_tokens": 8861108.0,
      "step": 3965
    },
    {
      "entropy": 5.782131719589233,
      "epoch": 0.4458920649182906,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987381582073962,
      "loss": 5.782,
      "mean_token_accuracy": 0.16714437305927277,
      "num_tokens": 8872161.0,
      "step": 3970
    },
    {
      "entropy": 6.058572435379029,
      "epoch": 0.4464536418262481,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987339085658848,
      "loss": 5.8653,
      "mean_token_accuracy": 0.16646484285593033,
      "num_tokens": 8882497.0,
      "step": 3975
    },
    {
      "entropy": 5.645896625518799,
      "epoch": 0.44701521873420563,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987296518005988,
      "loss": 5.5833,
      "mean_token_accuracy": 0.17363847494125367,
      "num_tokens": 8893155.0,
      "step": 3980
    },
    {
      "entropy": 5.686550903320312,
      "epoch": 0.4475767956421632,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498725387911674,
      "loss": 5.6662,
      "mean_token_accuracy": 0.1726548358798027,
      "num_tokens": 8904478.0,
      "step": 3985
    },
    {
      "entropy": 5.747406625747681,
      "epoch": 0.44813837255012073,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987211168992461,
      "loss": 5.7828,
      "mean_token_accuracy": 0.15897889584302902,
      "num_tokens": 8915470.0,
      "step": 3990
    },
    {
      "entropy": 5.84643874168396,
      "epoch": 0.4486999494580783,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498716838763451,
      "loss": 5.718,
      "mean_token_accuracy": 0.1680638760328293,
      "num_tokens": 8925644.0,
      "step": 3995
    },
    {
      "entropy": 5.809060144424438,
      "epoch": 0.44926152636603583,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498712553504425,
      "loss": 5.7545,
      "mean_token_accuracy": 0.16400753408670427,
      "num_tokens": 8937000.0,
      "step": 4000
    },
    {
      "entropy": 5.759871149063111,
      "epoch": 0.4498231032739934,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004987082611223046,
      "loss": 5.7301,
      "mean_token_accuracy": 0.16376389265060426,
      "num_tokens": 8948258.0,
      "step": 4005
    },
    {
      "entropy": 5.8029169082641605,
      "epoch": 0.45038468018195094,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987039616172265,
      "loss": 5.746,
      "mean_token_accuracy": 0.16891608238220215,
      "num_tokens": 8960107.0,
      "step": 4010
    },
    {
      "entropy": 5.846726274490356,
      "epoch": 0.4509462570899085,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986996549893273,
      "loss": 5.7237,
      "mean_token_accuracy": 0.16962783485651017,
      "num_tokens": 8969612.0,
      "step": 4015
    },
    {
      "entropy": 5.89162974357605,
      "epoch": 0.451507833997866,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004986953412387445,
      "loss": 5.8475,
      "mean_token_accuracy": 0.16363618075847625,
      "num_tokens": 8980970.0,
      "step": 4020
    },
    {
      "entropy": 5.861903858184815,
      "epoch": 0.45206941090582353,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004986910203656151,
      "loss": 5.771,
      "mean_token_accuracy": 0.16609935760498046,
      "num_tokens": 8992218.0,
      "step": 4025
    },
    {
      "entropy": 5.727895593643188,
      "epoch": 0.4526309878137811,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004986866923700769,
      "loss": 5.6925,
      "mean_token_accuracy": 0.17172871232032777,
      "num_tokens": 9003737.0,
      "step": 4030
    },
    {
      "entropy": 5.871471118927002,
      "epoch": 0.45319256472173863,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004986823572522677,
      "loss": 5.7609,
      "mean_token_accuracy": 0.16709625869989395,
      "num_tokens": 9015427.0,
      "step": 4035
    },
    {
      "entropy": 5.7899058818817135,
      "epoch": 0.4537541416296962,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986780150123254,
      "loss": 5.7803,
      "mean_token_accuracy": 0.16394250839948654,
      "num_tokens": 9025003.0,
      "step": 4040
    },
    {
      "entropy": 5.775137424468994,
      "epoch": 0.45431571853765373,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004986736656503885,
      "loss": 5.726,
      "mean_token_accuracy": 0.16647830456495286,
      "num_tokens": 9036074.0,
      "step": 4045
    },
    {
      "entropy": 5.936799144744873,
      "epoch": 0.4548772954456113,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004986693091665952,
      "loss": 5.738,
      "mean_token_accuracy": 0.16546613425016404,
      "num_tokens": 9046878.0,
      "step": 4050
    },
    {
      "entropy": 5.783922004699707,
      "epoch": 0.45543887235356884,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004986649455610845,
      "loss": 5.8007,
      "mean_token_accuracy": 0.16725005060434342,
      "num_tokens": 9058366.0,
      "step": 4055
    },
    {
      "entropy": 5.899766874313355,
      "epoch": 0.4560004492615264,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498660574833995,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16507331877946854,
      "num_tokens": 9069542.0,
      "step": 4060
    },
    {
      "entropy": 5.914427661895752,
      "epoch": 0.45656202616948394,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004986561969854661,
      "loss": 5.9796,
      "mean_token_accuracy": 0.15912373960018159,
      "num_tokens": 9081219.0,
      "step": 4065
    },
    {
      "entropy": 5.8226522445678714,
      "epoch": 0.45712360307744143,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004986518120156373,
      "loss": 5.6785,
      "mean_token_accuracy": 0.17707761228084565,
      "num_tokens": 9092101.0,
      "step": 4070
    },
    {
      "entropy": 5.781389141082764,
      "epoch": 0.457685179985399,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000498647419924648,
      "loss": 5.7367,
      "mean_token_accuracy": 0.16802326887845992,
      "num_tokens": 9101900.0,
      "step": 4075
    },
    {
      "entropy": 5.965834522247315,
      "epoch": 0.45824675689335653,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498643020712638,
      "loss": 5.867,
      "mean_token_accuracy": 0.1598948821425438,
      "num_tokens": 9114369.0,
      "step": 4080
    },
    {
      "entropy": 5.796750783920288,
      "epoch": 0.4588083338013141,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004986386143797475,
      "loss": 5.6732,
      "mean_token_accuracy": 0.16585982143878936,
      "num_tokens": 9126548.0,
      "step": 4085
    },
    {
      "entropy": 5.825146055221557,
      "epoch": 0.45936991070927163,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004986342009261169,
      "loss": 5.8567,
      "mean_token_accuracy": 0.16830722838640214,
      "num_tokens": 9138070.0,
      "step": 4090
    },
    {
      "entropy": 5.863235330581665,
      "epoch": 0.4599314876172292,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004986297803518865,
      "loss": 5.759,
      "mean_token_accuracy": 0.16653492152690888,
      "num_tokens": 9149489.0,
      "step": 4095
    },
    {
      "entropy": 5.8261312484741214,
      "epoch": 0.46049306452518673,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986253526571971,
      "loss": 5.7561,
      "mean_token_accuracy": 0.1725111186504364,
      "num_tokens": 9160363.0,
      "step": 4100
    },
    {
      "entropy": 5.8416753768920895,
      "epoch": 0.4610546414331443,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004986209178421899,
      "loss": 5.8144,
      "mean_token_accuracy": 0.16463642865419387,
      "num_tokens": 9171619.0,
      "step": 4105
    },
    {
      "entropy": 5.7307634353637695,
      "epoch": 0.46161621834110184,
      "grad_norm": 1.0,
      "learning_rate": 0.000498616475907006,
      "loss": 5.6512,
      "mean_token_accuracy": 0.17227994948625563,
      "num_tokens": 9182828.0,
      "step": 4110
    },
    {
      "entropy": 5.877857303619384,
      "epoch": 0.46217779524905933,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004986120268517865,
      "loss": 5.8204,
      "mean_token_accuracy": 0.17060580402612685,
      "num_tokens": 9195249.0,
      "step": 4115
    },
    {
      "entropy": 5.769808769226074,
      "epoch": 0.4627393721570169,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986075706766736,
      "loss": 5.6353,
      "mean_token_accuracy": 0.1722049430012703,
      "num_tokens": 9205770.0,
      "step": 4120
    },
    {
      "entropy": 5.683967733383179,
      "epoch": 0.46330094906497443,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004986031073818088,
      "loss": 5.6459,
      "mean_token_accuracy": 0.1706199139356613,
      "num_tokens": 9216367.0,
      "step": 4125
    },
    {
      "entropy": 5.781439018249512,
      "epoch": 0.463862525972932,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004985986369673343,
      "loss": 5.7368,
      "mean_token_accuracy": 0.1613730251789093,
      "num_tokens": 9228524.0,
      "step": 4130
    },
    {
      "entropy": 5.818389558792115,
      "epoch": 0.46442410288088953,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985941594333926,
      "loss": 5.7121,
      "mean_token_accuracy": 0.17004109174013138,
      "num_tokens": 9238907.0,
      "step": 4135
    },
    {
      "entropy": 5.794554662704468,
      "epoch": 0.4649856797888471,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498589674780126,
      "loss": 5.8502,
      "mean_token_accuracy": 0.16455847769975662,
      "num_tokens": 9250491.0,
      "step": 4140
    },
    {
      "entropy": 5.864607095718384,
      "epoch": 0.46554725669680463,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004985851830076775,
      "loss": 5.6501,
      "mean_token_accuracy": 0.17331554889678955,
      "num_tokens": 9260984.0,
      "step": 4145
    },
    {
      "entropy": 5.840168809890747,
      "epoch": 0.4661088336047622,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004985806841161901,
      "loss": 5.808,
      "mean_token_accuracy": 0.1647871918976307,
      "num_tokens": 9272416.0,
      "step": 4150
    },
    {
      "entropy": 5.794215488433838,
      "epoch": 0.46667041051271974,
      "grad_norm": 0.90625,
      "learning_rate": 0.000498576178105807,
      "loss": 5.7851,
      "mean_token_accuracy": 0.1661459356546402,
      "num_tokens": 9284513.0,
      "step": 4155
    },
    {
      "entropy": 5.86452431678772,
      "epoch": 0.4672319874206773,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004985716649766717,
      "loss": 5.6809,
      "mean_token_accuracy": 0.17553289085626603,
      "num_tokens": 9293422.0,
      "step": 4160
    },
    {
      "entropy": 5.720726060867309,
      "epoch": 0.4677935643286348,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985671447289277,
      "loss": 5.6794,
      "mean_token_accuracy": 0.16760085970163346,
      "num_tokens": 9304735.0,
      "step": 4165
    },
    {
      "entropy": 5.788778066635132,
      "epoch": 0.46835514123659233,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985626173627194,
      "loss": 5.7873,
      "mean_token_accuracy": 0.16755305379629135,
      "num_tokens": 9315659.0,
      "step": 4170
    },
    {
      "entropy": 5.810567426681518,
      "epoch": 0.4689167181445499,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004985580828781904,
      "loss": 5.8001,
      "mean_token_accuracy": 0.1698588401079178,
      "num_tokens": 9327106.0,
      "step": 4175
    },
    {
      "entropy": 5.9411545753479,
      "epoch": 0.46947829505250743,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004985535412754855,
      "loss": 5.8433,
      "mean_token_accuracy": 0.16779095083475112,
      "num_tokens": 9338762.0,
      "step": 4180
    },
    {
      "entropy": 5.756154012680054,
      "epoch": 0.470039871960465,
      "grad_norm": 0.953125,
      "learning_rate": 0.000498548992554749,
      "loss": 5.7029,
      "mean_token_accuracy": 0.17006189972162247,
      "num_tokens": 9349495.0,
      "step": 4185
    },
    {
      "entropy": 5.910188484191894,
      "epoch": 0.47060144886842253,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004985444367161259,
      "loss": 5.8079,
      "mean_token_accuracy": 0.17130580097436904,
      "num_tokens": 9360830.0,
      "step": 4190
    },
    {
      "entropy": 5.705734586715698,
      "epoch": 0.4711630257763801,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985398737597614,
      "loss": 5.6845,
      "mean_token_accuracy": 0.16564561575651168,
      "num_tokens": 9371668.0,
      "step": 4195
    },
    {
      "entropy": 5.817333698272705,
      "epoch": 0.47172460268433763,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004985353036858004,
      "loss": 5.8105,
      "mean_token_accuracy": 0.16629635244607927,
      "num_tokens": 9383437.0,
      "step": 4200
    },
    {
      "entropy": 5.768382501602173,
      "epoch": 0.4722861795922952,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004985307264943887,
      "loss": 5.6339,
      "mean_token_accuracy": 0.17444221079349517,
      "num_tokens": 9394597.0,
      "step": 4205
    },
    {
      "entropy": 5.6198773860931395,
      "epoch": 0.47284775650025274,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004985261421856721,
      "loss": 5.577,
      "mean_token_accuracy": 0.17500848025083543,
      "num_tokens": 9404322.0,
      "step": 4210
    },
    {
      "entropy": 5.8928163051605225,
      "epoch": 0.47340933340821023,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985215507597963,
      "loss": 5.9324,
      "mean_token_accuracy": 0.15648610070347785,
      "num_tokens": 9416740.0,
      "step": 4215
    },
    {
      "entropy": 5.868355560302734,
      "epoch": 0.4739709103161678,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985169522169076,
      "loss": 5.7415,
      "mean_token_accuracy": 0.17201738208532333,
      "num_tokens": 9428617.0,
      "step": 4220
    },
    {
      "entropy": 5.704996824264526,
      "epoch": 0.47453248722412533,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004985123465571526,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1724829375743866,
      "num_tokens": 9439686.0,
      "step": 4225
    },
    {
      "entropy": 5.8679687023162845,
      "epoch": 0.4750940641320829,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985077337806777,
      "loss": 5.6878,
      "mean_token_accuracy": 0.17257601916790008,
      "num_tokens": 9450378.0,
      "step": 4230
    },
    {
      "entropy": 5.819633293151855,
      "epoch": 0.47565564104004043,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985031138876299,
      "loss": 5.7126,
      "mean_token_accuracy": 0.1686561957001686,
      "num_tokens": 9461659.0,
      "step": 4235
    },
    {
      "entropy": 5.743727445602417,
      "epoch": 0.476217217947998,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004984984868781563,
      "loss": 5.7547,
      "mean_token_accuracy": 0.17106645703315734,
      "num_tokens": 9472528.0,
      "step": 4240
    },
    {
      "entropy": 5.846922206878662,
      "epoch": 0.47677879485595553,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984938527524041,
      "loss": 5.6776,
      "mean_token_accuracy": 0.17797594964504243,
      "num_tokens": 9483388.0,
      "step": 4245
    },
    {
      "entropy": 5.644798994064331,
      "epoch": 0.4773403717639131,
      "grad_norm": 1.0,
      "learning_rate": 0.000498489211510521,
      "loss": 5.5383,
      "mean_token_accuracy": 0.1787099227309227,
      "num_tokens": 9495178.0,
      "step": 4250
    },
    {
      "entropy": 5.699044275283813,
      "epoch": 0.47790194867187064,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004984845631526547,
      "loss": 5.7797,
      "mean_token_accuracy": 0.16378788501024247,
      "num_tokens": 9506226.0,
      "step": 4255
    },
    {
      "entropy": 5.863605213165283,
      "epoch": 0.47846352557982813,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004984799076789534,
      "loss": 5.6729,
      "mean_token_accuracy": 0.16951856166124343,
      "num_tokens": 9517671.0,
      "step": 4260
    },
    {
      "entropy": 5.764629888534546,
      "epoch": 0.4790251024877857,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004984752450895649,
      "loss": 5.7478,
      "mean_token_accuracy": 0.1722684383392334,
      "num_tokens": 9527637.0,
      "step": 4265
    },
    {
      "entropy": 5.832716703414917,
      "epoch": 0.47958667939574323,
      "grad_norm": 0.96875,
      "learning_rate": 0.000498470575384638,
      "loss": 5.8458,
      "mean_token_accuracy": 0.16255225837230683,
      "num_tokens": 9539686.0,
      "step": 4270
    },
    {
      "entropy": 5.893230867385864,
      "epoch": 0.4801482563037008,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984658985643214,
      "loss": 5.8008,
      "mean_token_accuracy": 0.17014207541942597,
      "num_tokens": 9550775.0,
      "step": 4275
    },
    {
      "entropy": 5.86106767654419,
      "epoch": 0.48070983321165833,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004984612146287638,
      "loss": 5.7186,
      "mean_token_accuracy": 0.1644670069217682,
      "num_tokens": 9561760.0,
      "step": 4280
    },
    {
      "entropy": 5.694846820831299,
      "epoch": 0.4812714101196159,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004984565235781147,
      "loss": 5.7358,
      "mean_token_accuracy": 0.1661587029695511,
      "num_tokens": 9573899.0,
      "step": 4285
    },
    {
      "entropy": 5.78567066192627,
      "epoch": 0.48183298702757343,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498451825412523,
      "loss": 5.7017,
      "mean_token_accuracy": 0.16853241473436356,
      "num_tokens": 9584336.0,
      "step": 4290
    },
    {
      "entropy": 5.927900266647339,
      "epoch": 0.482394563935531,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004984471201321386,
      "loss": 5.88,
      "mean_token_accuracy": 0.162874473631382,
      "num_tokens": 9596614.0,
      "step": 4295
    },
    {
      "entropy": 5.835678482055664,
      "epoch": 0.48295614084348854,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004984424077371112,
      "loss": 5.7325,
      "mean_token_accuracy": 0.17014361917972565,
      "num_tokens": 9606796.0,
      "step": 4300
    },
    {
      "entropy": 5.778892517089844,
      "epoch": 0.4835177177514461,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984376882275909,
      "loss": 5.7093,
      "mean_token_accuracy": 0.16853998601436615,
      "num_tokens": 9617705.0,
      "step": 4305
    },
    {
      "entropy": 5.704960536956787,
      "epoch": 0.4840792946594036,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004984329616037279,
      "loss": 5.7265,
      "mean_token_accuracy": 0.17152929455041885,
      "num_tokens": 9628098.0,
      "step": 4310
    },
    {
      "entropy": 5.815963077545166,
      "epoch": 0.48464087156736113,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004984282278656727,
      "loss": 5.6878,
      "mean_token_accuracy": 0.17872087508440018,
      "num_tokens": 9637697.0,
      "step": 4315
    },
    {
      "entropy": 5.674087190628052,
      "epoch": 0.4852024484753187,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984234870135762,
      "loss": 5.5793,
      "mean_token_accuracy": 0.17939777821302413,
      "num_tokens": 9648221.0,
      "step": 4320
    },
    {
      "entropy": 5.805170249938965,
      "epoch": 0.48576402538327623,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004984187390475891,
      "loss": 5.7692,
      "mean_token_accuracy": 0.1637537583708763,
      "num_tokens": 9660182.0,
      "step": 4325
    },
    {
      "entropy": 5.825180864334106,
      "epoch": 0.4863256022912338,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004984139839678628,
      "loss": 5.777,
      "mean_token_accuracy": 0.16494843363761902,
      "num_tokens": 9672121.0,
      "step": 4330
    },
    {
      "entropy": 5.823313140869141,
      "epoch": 0.48688717919919133,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984092217745486,
      "loss": 5.6781,
      "mean_token_accuracy": 0.17020544558763503,
      "num_tokens": 9682651.0,
      "step": 4335
    },
    {
      "entropy": 5.806669759750366,
      "epoch": 0.4874487561071489,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004984044524677981,
      "loss": 5.8241,
      "mean_token_accuracy": 0.16280035227537154,
      "num_tokens": 9693989.0,
      "step": 4340
    },
    {
      "entropy": 5.678175973892212,
      "epoch": 0.48801033301510643,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004983996760477632,
      "loss": 5.6732,
      "mean_token_accuracy": 0.17800934761762618,
      "num_tokens": 9703802.0,
      "step": 4345
    },
    {
      "entropy": 5.73551115989685,
      "epoch": 0.488571909923064,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498394892514596,
      "loss": 5.6309,
      "mean_token_accuracy": 0.17766664773225785,
      "num_tokens": 9714287.0,
      "step": 4350
    },
    {
      "entropy": 5.779881143569947,
      "epoch": 0.48913348683102154,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004983901018684486,
      "loss": 5.655,
      "mean_token_accuracy": 0.17147285342216492,
      "num_tokens": 9724819.0,
      "step": 4355
    },
    {
      "entropy": 5.801750993728637,
      "epoch": 0.48969506373897903,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004983853041094739,
      "loss": 5.8086,
      "mean_token_accuracy": 0.16176895052194595,
      "num_tokens": 9736326.0,
      "step": 4360
    },
    {
      "entropy": 5.879187440872192,
      "epoch": 0.4902566406469366,
      "grad_norm": 0.875,
      "learning_rate": 0.0004983804992378243,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16969723701477052,
      "num_tokens": 9747883.0,
      "step": 4365
    },
    {
      "entropy": 5.767481184005737,
      "epoch": 0.49081821755489413,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004983756872536532,
      "loss": 5.6812,
      "mean_token_accuracy": 0.17253028899431228,
      "num_tokens": 9758191.0,
      "step": 4370
    },
    {
      "entropy": 5.8729065418243405,
      "epoch": 0.4913797944628517,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983708681571134,
      "loss": 5.7807,
      "mean_token_accuracy": 0.16696447283029556,
      "num_tokens": 9768852.0,
      "step": 4375
    },
    {
      "entropy": 5.849040746688843,
      "epoch": 0.49194137137080923,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004983660419483586,
      "loss": 5.7301,
      "mean_token_accuracy": 0.16603904068470002,
      "num_tokens": 9779033.0,
      "step": 4380
    },
    {
      "entropy": 5.716362190246582,
      "epoch": 0.4925029482787668,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004983612086275424,
      "loss": 5.6373,
      "mean_token_accuracy": 0.17526944130659103,
      "num_tokens": 9790777.0,
      "step": 4385
    },
    {
      "entropy": 5.915458679199219,
      "epoch": 0.49306452518672433,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983563681948187,
      "loss": 5.909,
      "mean_token_accuracy": 0.15452502816915512,
      "num_tokens": 9802305.0,
      "step": 4390
    },
    {
      "entropy": 5.837260770797729,
      "epoch": 0.4936261020946819,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983515206503416,
      "loss": 5.7088,
      "mean_token_accuracy": 0.16492092460393906,
      "num_tokens": 9813590.0,
      "step": 4395
    },
    {
      "entropy": 5.762714004516601,
      "epoch": 0.49418767900263944,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983466659942655,
      "loss": 5.78,
      "mean_token_accuracy": 0.1626877874135971,
      "num_tokens": 9824080.0,
      "step": 4400
    },
    {
      "entropy": 5.770403957366943,
      "epoch": 0.49474925591059693,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983418042267449,
      "loss": 5.6743,
      "mean_token_accuracy": 0.16747995167970658,
      "num_tokens": 9835441.0,
      "step": 4405
    },
    {
      "entropy": 5.74335503578186,
      "epoch": 0.4953108328185545,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004983369353479345,
      "loss": 5.659,
      "mean_token_accuracy": 0.1735517680644989,
      "num_tokens": 9846750.0,
      "step": 4410
    },
    {
      "entropy": 5.842858934402466,
      "epoch": 0.49587240972651203,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004983320593579895,
      "loss": 5.81,
      "mean_token_accuracy": 0.16727619916200637,
      "num_tokens": 9857987.0,
      "step": 4415
    },
    {
      "entropy": 5.785632085800171,
      "epoch": 0.4964339866344696,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004983271762570653,
      "loss": 5.7654,
      "mean_token_accuracy": 0.16815820336341858,
      "num_tokens": 9869736.0,
      "step": 4420
    },
    {
      "entropy": 5.8420435905456545,
      "epoch": 0.49699556354242713,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000498322286045317,
      "loss": 5.7281,
      "mean_token_accuracy": 0.16822068840265275,
      "num_tokens": 9881610.0,
      "step": 4425
    },
    {
      "entropy": 5.783283519744873,
      "epoch": 0.4975571404503847,
      "grad_norm": 1.0,
      "learning_rate": 0.0004983173887229006,
      "loss": 5.7289,
      "mean_token_accuracy": 0.16912795081734658,
      "num_tokens": 9893203.0,
      "step": 4430
    },
    {
      "entropy": 5.775019788742066,
      "epoch": 0.49811871735834223,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004983124842899719,
      "loss": 5.6411,
      "mean_token_accuracy": 0.16674574017524718,
      "num_tokens": 9904392.0,
      "step": 4435
    },
    {
      "entropy": 5.601576042175293,
      "epoch": 0.4986802942662998,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004983075727466871,
      "loss": 5.6408,
      "mean_token_accuracy": 0.17083830386400223,
      "num_tokens": 9916184.0,
      "step": 4440
    },
    {
      "entropy": 5.817759895324707,
      "epoch": 0.49924187117425733,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004983026540932026,
      "loss": 5.7422,
      "mean_token_accuracy": 0.17348867505788804,
      "num_tokens": 9927516.0,
      "step": 4445
    },
    {
      "entropy": 5.817559862136841,
      "epoch": 0.4998034480822149,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004982977283296749,
      "loss": 5.7749,
      "mean_token_accuracy": 0.16315798461437225,
      "num_tokens": 9939182.0,
      "step": 4450
    },
    {
      "entropy": 5.855092334747314,
      "epoch": 0.5003650249901724,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498292795456261,
      "loss": 5.7898,
      "mean_token_accuracy": 0.16669366359710694,
      "num_tokens": 9950382.0,
      "step": 4455
    },
    {
      "entropy": 5.796814346313477,
      "epoch": 0.50092660189813,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004982878554731178,
      "loss": 5.7932,
      "mean_token_accuracy": 0.17010375410318374,
      "num_tokens": 9961053.0,
      "step": 4460
    },
    {
      "entropy": 5.765320682525635,
      "epoch": 0.5014881788060875,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004982829083804028,
      "loss": 5.6947,
      "mean_token_accuracy": 0.17564453780651093,
      "num_tokens": 9971313.0,
      "step": 4465
    },
    {
      "entropy": 5.784207010269165,
      "epoch": 0.5020497557140451,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982779541782733,
      "loss": 5.7036,
      "mean_token_accuracy": 0.17723664939403533,
      "num_tokens": 9981257.0,
      "step": 4470
    },
    {
      "entropy": 5.7435914993286135,
      "epoch": 0.5026113326220026,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004982729928668872,
      "loss": 5.6139,
      "mean_token_accuracy": 0.17061744183301925,
      "num_tokens": 9991860.0,
      "step": 4475
    },
    {
      "entropy": 5.721191549301148,
      "epoch": 0.5031729095299601,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004982680244464024,
      "loss": 5.7319,
      "mean_token_accuracy": 0.170281982421875,
      "num_tokens": 10002320.0,
      "step": 4480
    },
    {
      "entropy": 5.869790172576904,
      "epoch": 0.5037344864379176,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004982630489169771,
      "loss": 5.8147,
      "mean_token_accuracy": 0.16252802163362504,
      "num_tokens": 10013187.0,
      "step": 4485
    },
    {
      "entropy": 5.748865604400635,
      "epoch": 0.5042960633458752,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004982580662787697,
      "loss": 5.551,
      "mean_token_accuracy": 0.18117449134588243,
      "num_tokens": 10024268.0,
      "step": 4490
    },
    {
      "entropy": 5.773206615447998,
      "epoch": 0.5048576402538327,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982530765319389,
      "loss": 5.8156,
      "mean_token_accuracy": 0.16586428880691528,
      "num_tokens": 10034738.0,
      "step": 4495
    },
    {
      "entropy": 5.867972803115845,
      "epoch": 0.5054192171617903,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004982480796766436,
      "loss": 5.7633,
      "mean_token_accuracy": 0.16974856257438659,
      "num_tokens": 10045666.0,
      "step": 4500
    },
    {
      "entropy": 5.755939483642578,
      "epoch": 0.5059807940697478,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004982430757130429,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17190644890069962,
      "num_tokens": 10057084.0,
      "step": 4505
    },
    {
      "entropy": 5.667339277267456,
      "epoch": 0.5065423709777054,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982380646412959,
      "loss": 5.6199,
      "mean_token_accuracy": 0.17011208087205887,
      "num_tokens": 10066754.0,
      "step": 4510
    },
    {
      "entropy": 5.678653812408447,
      "epoch": 0.5071039478856629,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004982330464615625,
      "loss": 5.6282,
      "mean_token_accuracy": 0.17506971210241318,
      "num_tokens": 10078139.0,
      "step": 4515
    },
    {
      "entropy": 5.80544900894165,
      "epoch": 0.5076655247936205,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004982280211740023,
      "loss": 5.6884,
      "mean_token_accuracy": 0.16866966784000398,
      "num_tokens": 10089652.0,
      "step": 4520
    },
    {
      "entropy": 5.6051665306091305,
      "epoch": 0.508227101701578,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004982229887787753,
      "loss": 5.5358,
      "mean_token_accuracy": 0.17593766152858734,
      "num_tokens": 10100642.0,
      "step": 4525
    },
    {
      "entropy": 5.782035064697266,
      "epoch": 0.5087886786095356,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004982179492760418,
      "loss": 5.7448,
      "mean_token_accuracy": 0.16809683442115783,
      "num_tokens": 10112036.0,
      "step": 4530
    },
    {
      "entropy": 5.784066915512085,
      "epoch": 0.5093502555174931,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004982129026659621,
      "loss": 5.6198,
      "mean_token_accuracy": 0.17079970091581345,
      "num_tokens": 10121653.0,
      "step": 4535
    },
    {
      "entropy": 5.767913055419922,
      "epoch": 0.5099118324254507,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982078489486972,
      "loss": 5.7242,
      "mean_token_accuracy": 0.1688558965921402,
      "num_tokens": 10132271.0,
      "step": 4540
    },
    {
      "entropy": 5.658374261856079,
      "epoch": 0.5104734093334082,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982027881244076,
      "loss": 5.606,
      "mean_token_accuracy": 0.17197940349578858,
      "num_tokens": 10143411.0,
      "step": 4545
    },
    {
      "entropy": 5.847708702087402,
      "epoch": 0.5110349862413658,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981977201932549,
      "loss": 5.7242,
      "mean_token_accuracy": 0.1733257830142975,
      "num_tokens": 10154338.0,
      "step": 4550
    },
    {
      "entropy": 5.802937316894531,
      "epoch": 0.5115965631493233,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004981926451554001,
      "loss": 5.7516,
      "mean_token_accuracy": 0.17324941903352736,
      "num_tokens": 10164810.0,
      "step": 4555
    },
    {
      "entropy": 5.7534888744354244,
      "epoch": 0.5121581400572809,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981875630110048,
      "loss": 5.6731,
      "mean_token_accuracy": 0.17414593547582627,
      "num_tokens": 10176323.0,
      "step": 4560
    },
    {
      "entropy": 5.736021661758423,
      "epoch": 0.5127197169652384,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981824737602311,
      "loss": 5.7073,
      "mean_token_accuracy": 0.17113523185253143,
      "num_tokens": 10187110.0,
      "step": 4565
    },
    {
      "entropy": 5.761826086044311,
      "epoch": 0.513281293873196,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004981773774032409,
      "loss": 5.6717,
      "mean_token_accuracy": 0.16766324490308762,
      "num_tokens": 10199099.0,
      "step": 4570
    },
    {
      "entropy": 5.785593271255493,
      "epoch": 0.5138428707811534,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004981722739401962,
      "loss": 5.6747,
      "mean_token_accuracy": 0.1726672261953354,
      "num_tokens": 10209726.0,
      "step": 4575
    },
    {
      "entropy": 5.755669784545899,
      "epoch": 0.514404447689111,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00049816716337126,
      "loss": 5.7604,
      "mean_token_accuracy": 0.16705486178398132,
      "num_tokens": 10220858.0,
      "step": 4580
    },
    {
      "entropy": 5.644749641418457,
      "epoch": 0.5149660245970685,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004981620456965946,
      "loss": 5.577,
      "mean_token_accuracy": 0.1802625373005867,
      "num_tokens": 10232250.0,
      "step": 4585
    },
    {
      "entropy": 5.893209981918335,
      "epoch": 0.5155276015050261,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004981569209163633,
      "loss": 5.7647,
      "mean_token_accuracy": 0.1636040985584259,
      "num_tokens": 10244584.0,
      "step": 4590
    },
    {
      "entropy": 5.823698377609253,
      "epoch": 0.5160891784129836,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004981517890307288,
      "loss": 5.6569,
      "mean_token_accuracy": 0.16793317794799806,
      "num_tokens": 10255926.0,
      "step": 4595
    },
    {
      "entropy": 5.751343584060669,
      "epoch": 0.5166507553209412,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498146650039855,
      "loss": 5.7326,
      "mean_token_accuracy": 0.17038354724645616,
      "num_tokens": 10266691.0,
      "step": 4600
    },
    {
      "entropy": 5.818839836120605,
      "epoch": 0.5172123322288987,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981415039439054,
      "loss": 5.7986,
      "mean_token_accuracy": 0.1667577788233757,
      "num_tokens": 10277744.0,
      "step": 4605
    },
    {
      "entropy": 5.718520307540894,
      "epoch": 0.5177739091368563,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004981363507430436,
      "loss": 5.6473,
      "mean_token_accuracy": 0.17276567965745926,
      "num_tokens": 10288792.0,
      "step": 4610
    },
    {
      "entropy": 5.818396329879761,
      "epoch": 0.5183354860448138,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004981311904374339,
      "loss": 5.6795,
      "mean_token_accuracy": 0.16973294615745543,
      "num_tokens": 10299569.0,
      "step": 4615
    },
    {
      "entropy": 5.722053909301758,
      "epoch": 0.5188970629527714,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004981260230272405,
      "loss": 5.6642,
      "mean_token_accuracy": 0.17573922574520112,
      "num_tokens": 10309742.0,
      "step": 4620
    },
    {
      "entropy": 5.8336011409759525,
      "epoch": 0.5194586398607289,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004981208485126281,
      "loss": 5.7048,
      "mean_token_accuracy": 0.1665973499417305,
      "num_tokens": 10320603.0,
      "step": 4625
    },
    {
      "entropy": 5.736460113525391,
      "epoch": 0.5200202167686865,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981156668937613,
      "loss": 5.6645,
      "mean_token_accuracy": 0.17377568930387496,
      "num_tokens": 10331498.0,
      "step": 4630
    },
    {
      "entropy": 5.715865516662598,
      "epoch": 0.520581793676644,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004981104781708051,
      "loss": 5.6988,
      "mean_token_accuracy": 0.17312800139188766,
      "num_tokens": 10342002.0,
      "step": 4635
    },
    {
      "entropy": 5.749444532394409,
      "epoch": 0.5211433705846016,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004981052823439247,
      "loss": 5.6599,
      "mean_token_accuracy": 0.16870316565036775,
      "num_tokens": 10351797.0,
      "step": 4640
    },
    {
      "entropy": 5.719133377075195,
      "epoch": 0.5217049474925591,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004981000794132858,
      "loss": 5.6073,
      "mean_token_accuracy": 0.17236316353082656,
      "num_tokens": 10362538.0,
      "step": 4645
    },
    {
      "entropy": 5.737236261367798,
      "epoch": 0.5222665244005167,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004980948693790536,
      "loss": 5.6399,
      "mean_token_accuracy": 0.17452512383461,
      "num_tokens": 10373365.0,
      "step": 4650
    },
    {
      "entropy": 5.641112422943115,
      "epoch": 0.5228281013084742,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004980896522413943,
      "loss": 5.6009,
      "mean_token_accuracy": 0.17220466434955597,
      "num_tokens": 10385087.0,
      "step": 4655
    },
    {
      "entropy": 5.7947471141815186,
      "epoch": 0.5233896782164318,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004980844280004739,
      "loss": 5.7315,
      "mean_token_accuracy": 0.1702256292104721,
      "num_tokens": 10396014.0,
      "step": 4660
    },
    {
      "entropy": 5.765877151489258,
      "epoch": 0.5239512551243893,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004980791966564589,
      "loss": 5.6451,
      "mean_token_accuracy": 0.1674633339047432,
      "num_tokens": 10407844.0,
      "step": 4665
    },
    {
      "entropy": 5.6722806930542,
      "epoch": 0.5245128320323468,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004980739582095158,
      "loss": 5.7118,
      "mean_token_accuracy": 0.16796951591968537,
      "num_tokens": 10419892.0,
      "step": 4670
    },
    {
      "entropy": 5.8044257164001465,
      "epoch": 0.5250744089403043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004980687126598111,
      "loss": 5.6947,
      "mean_token_accuracy": 0.16804990619421006,
      "num_tokens": 10430686.0,
      "step": 4675
    },
    {
      "entropy": 5.7502185821533205,
      "epoch": 0.5256359858482619,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004980634600075123,
      "loss": 5.7058,
      "mean_token_accuracy": 0.16406390219926834,
      "num_tokens": 10443785.0,
      "step": 4680
    },
    {
      "entropy": 5.69890947341919,
      "epoch": 0.5261975627562194,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980582002527863,
      "loss": 5.6032,
      "mean_token_accuracy": 0.17852845937013626,
      "num_tokens": 10454889.0,
      "step": 4685
    },
    {
      "entropy": 5.756927967071533,
      "epoch": 0.526759139664177,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004980529333958008,
      "loss": 5.7589,
      "mean_token_accuracy": 0.1650112196803093,
      "num_tokens": 10466525.0,
      "step": 4690
    },
    {
      "entropy": 5.854841518402099,
      "epoch": 0.5273207165721345,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004980476594367233,
      "loss": 5.7132,
      "mean_token_accuracy": 0.16974902898073196,
      "num_tokens": 10476329.0,
      "step": 4695
    },
    {
      "entropy": 5.827983236312866,
      "epoch": 0.5278822934800921,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980423783757219,
      "loss": 5.7684,
      "mean_token_accuracy": 0.16705012321472168,
      "num_tokens": 10487178.0,
      "step": 4700
    },
    {
      "entropy": 5.820889663696289,
      "epoch": 0.5284438703880496,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004980370902129646,
      "loss": 5.7131,
      "mean_token_accuracy": 0.1603904828429222,
      "num_tokens": 10499293.0,
      "step": 4705
    },
    {
      "entropy": 5.8819056987762455,
      "epoch": 0.5290054472960072,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049803179494862,
      "loss": 5.8989,
      "mean_token_accuracy": 0.1597862184047699,
      "num_tokens": 10511762.0,
      "step": 4710
    },
    {
      "entropy": 5.775479793548584,
      "epoch": 0.5295670242039647,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980264925828563,
      "loss": 5.6241,
      "mean_token_accuracy": 0.17840833216905594,
      "num_tokens": 10521725.0,
      "step": 4715
    },
    {
      "entropy": 5.639372968673706,
      "epoch": 0.5301286011119223,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004980211831158427,
      "loss": 5.5844,
      "mean_token_accuracy": 0.17198170274496077,
      "num_tokens": 10531764.0,
      "step": 4720
    },
    {
      "entropy": 5.737982130050659,
      "epoch": 0.5306901780198798,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004980158665477481,
      "loss": 5.6481,
      "mean_token_accuracy": 0.17444745302200318,
      "num_tokens": 10542127.0,
      "step": 4725
    },
    {
      "entropy": 5.851061010360718,
      "epoch": 0.5312517549278374,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004980105428787419,
      "loss": 5.7351,
      "mean_token_accuracy": 0.16558413803577424,
      "num_tokens": 10552580.0,
      "step": 4730
    },
    {
      "entropy": 5.777568435668945,
      "epoch": 0.5318133318357949,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980052121089934,
      "loss": 5.6917,
      "mean_token_accuracy": 0.1689352035522461,
      "num_tokens": 10562961.0,
      "step": 4735
    },
    {
      "entropy": 5.761757183074951,
      "epoch": 0.5323749087437525,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979998742386725,
      "loss": 5.6857,
      "mean_token_accuracy": 0.174189792573452,
      "num_tokens": 10573210.0,
      "step": 4740
    },
    {
      "entropy": 5.648348140716553,
      "epoch": 0.53293648565171,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004979945292679492,
      "loss": 5.6351,
      "mean_token_accuracy": 0.16878298968076705,
      "num_tokens": 10583878.0,
      "step": 4745
    },
    {
      "entropy": 5.798595285415649,
      "epoch": 0.5334980625596676,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004979891771969935,
      "loss": 5.7171,
      "mean_token_accuracy": 0.16758286952972412,
      "num_tokens": 10593420.0,
      "step": 4750
    },
    {
      "entropy": 5.799276733398438,
      "epoch": 0.5340596394676251,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497983818025976,
      "loss": 5.7368,
      "mean_token_accuracy": 0.16662465855479242,
      "num_tokens": 10604213.0,
      "step": 4755
    },
    {
      "entropy": 5.797427368164063,
      "epoch": 0.5346212163755827,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979784517550671,
      "loss": 5.7054,
      "mean_token_accuracy": 0.1650031179189682,
      "num_tokens": 10615502.0,
      "step": 4760
    },
    {
      "entropy": 5.812700986862183,
      "epoch": 0.5351827932835401,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004979730783844379,
      "loss": 5.7668,
      "mean_token_accuracy": 0.1636280044913292,
      "num_tokens": 10627589.0,
      "step": 4765
    },
    {
      "entropy": 5.721961879730225,
      "epoch": 0.5357443701914977,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979676979142595,
      "loss": 5.6092,
      "mean_token_accuracy": 0.17420833557844162,
      "num_tokens": 10638365.0,
      "step": 4770
    },
    {
      "entropy": 5.775761985778809,
      "epoch": 0.5363059470994552,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497962310344703,
      "loss": 5.6992,
      "mean_token_accuracy": 0.1731460452079773,
      "num_tokens": 10650664.0,
      "step": 4775
    },
    {
      "entropy": 5.7953284740447994,
      "epoch": 0.5368675240074128,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00049795691567594,
      "loss": 5.7564,
      "mean_token_accuracy": 0.17478667497634887,
      "num_tokens": 10662418.0,
      "step": 4780
    },
    {
      "entropy": 5.774968767166138,
      "epoch": 0.5374291009153703,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004979515139081426,
      "loss": 5.6265,
      "mean_token_accuracy": 0.1747649058699608,
      "num_tokens": 10672793.0,
      "step": 4785
    },
    {
      "entropy": 5.7399228572845455,
      "epoch": 0.5379906778233279,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004979461050414822,
      "loss": 5.6728,
      "mean_token_accuracy": 0.1787702426314354,
      "num_tokens": 10684082.0,
      "step": 4790
    },
    {
      "entropy": 5.687348508834839,
      "epoch": 0.5385522547312854,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979406890761315,
      "loss": 5.6123,
      "mean_token_accuracy": 0.17388249784708024,
      "num_tokens": 10694490.0,
      "step": 4795
    },
    {
      "entropy": 5.682422208786011,
      "epoch": 0.539113831639243,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004979352660122629,
      "loss": 5.6102,
      "mean_token_accuracy": 0.1730710655450821,
      "num_tokens": 10704731.0,
      "step": 4800
    },
    {
      "entropy": 5.7284330368042,
      "epoch": 0.5396754085472005,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004979298358500488,
      "loss": 5.6458,
      "mean_token_accuracy": 0.17584740072488786,
      "num_tokens": 10715735.0,
      "step": 4805
    },
    {
      "entropy": 5.70305609703064,
      "epoch": 0.5402369854551581,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004979243985896624,
      "loss": 5.6098,
      "mean_token_accuracy": 0.1710106059908867,
      "num_tokens": 10726951.0,
      "step": 4810
    },
    {
      "entropy": 5.645564556121826,
      "epoch": 0.5407985623631156,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004979189542312765,
      "loss": 5.6851,
      "mean_token_accuracy": 0.16763894110918046,
      "num_tokens": 10737609.0,
      "step": 4815
    },
    {
      "entropy": 5.723374080657959,
      "epoch": 0.5413601392710732,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004979135027750649,
      "loss": 5.5759,
      "mean_token_accuracy": 0.17617718130350113,
      "num_tokens": 10748957.0,
      "step": 4820
    },
    {
      "entropy": 5.7252906322479244,
      "epoch": 0.5419217161790307,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979080442212007,
      "loss": 5.7624,
      "mean_token_accuracy": 0.16397164314985274,
      "num_tokens": 10759398.0,
      "step": 4825
    },
    {
      "entropy": 5.8732177734375,
      "epoch": 0.5424832930869883,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979025785698581,
      "loss": 5.7446,
      "mean_token_accuracy": 0.16531347930431367,
      "num_tokens": 10770436.0,
      "step": 4830
    },
    {
      "entropy": 5.633301305770874,
      "epoch": 0.5430448699949458,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004978971058212108,
      "loss": 5.5752,
      "mean_token_accuracy": 0.1801727831363678,
      "num_tokens": 10781160.0,
      "step": 4835
    },
    {
      "entropy": 5.753065013885498,
      "epoch": 0.5436064469029034,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004978916259754331,
      "loss": 5.7015,
      "mean_token_accuracy": 0.1770429030060768,
      "num_tokens": 10793170.0,
      "step": 4840
    },
    {
      "entropy": 5.803515291213989,
      "epoch": 0.5441680238108609,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004978861390326998,
      "loss": 5.6435,
      "mean_token_accuracy": 0.1693033382296562,
      "num_tokens": 10803515.0,
      "step": 4845
    },
    {
      "entropy": 5.7509010314941404,
      "epoch": 0.5447296007188185,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004978806449931854,
      "loss": 5.6651,
      "mean_token_accuracy": 0.1700579084455967,
      "num_tokens": 10814508.0,
      "step": 4850
    },
    {
      "entropy": 5.723728084564209,
      "epoch": 0.545291177626776,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978751438570647,
      "loss": 5.7722,
      "mean_token_accuracy": 0.16889193505048752,
      "num_tokens": 10825924.0,
      "step": 4855
    },
    {
      "entropy": 5.863556575775147,
      "epoch": 0.5458527545347336,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978696356245129,
      "loss": 5.7463,
      "mean_token_accuracy": 0.16716790348291397,
      "num_tokens": 10837854.0,
      "step": 4860
    },
    {
      "entropy": 5.762776231765747,
      "epoch": 0.546414331442691,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004978641202957057,
      "loss": 5.6418,
      "mean_token_accuracy": 0.1727205768227577,
      "num_tokens": 10848279.0,
      "step": 4865
    },
    {
      "entropy": 5.611614179611206,
      "epoch": 0.5469759083506486,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004978585978708183,
      "loss": 5.654,
      "mean_token_accuracy": 0.1656600594520569,
      "num_tokens": 10859615.0,
      "step": 4870
    },
    {
      "entropy": 5.882994222640991,
      "epoch": 0.5475374852586061,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004978530683500267,
      "loss": 5.6263,
      "mean_token_accuracy": 0.17195343524217604,
      "num_tokens": 10871388.0,
      "step": 4875
    },
    {
      "entropy": 5.682742977142334,
      "epoch": 0.5480990621665637,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004978475317335071,
      "loss": 5.6919,
      "mean_token_accuracy": 0.1701396718621254,
      "num_tokens": 10882411.0,
      "step": 4880
    },
    {
      "entropy": 5.731553316116333,
      "epoch": 0.5486606390745212,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004978419880214356,
      "loss": 5.7924,
      "mean_token_accuracy": 0.16876837164163588,
      "num_tokens": 10894116.0,
      "step": 4885
    },
    {
      "entropy": 5.752131938934326,
      "epoch": 0.5492222159824788,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978364372139887,
      "loss": 5.5682,
      "mean_token_accuracy": 0.17538228034973144,
      "num_tokens": 10904514.0,
      "step": 4890
    },
    {
      "entropy": 5.689201021194458,
      "epoch": 0.5497837928904363,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004978308793113434,
      "loss": 5.6134,
      "mean_token_accuracy": 0.17261481434106826,
      "num_tokens": 10915602.0,
      "step": 4895
    },
    {
      "entropy": 5.682745933532715,
      "epoch": 0.5503453697983939,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004978253143136762,
      "loss": 5.6033,
      "mean_token_accuracy": 0.17805052399635315,
      "num_tokens": 10926172.0,
      "step": 4900
    },
    {
      "entropy": 5.778084802627563,
      "epoch": 0.5509069467063514,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004978197422211647,
      "loss": 5.7268,
      "mean_token_accuracy": 0.16874644458293914,
      "num_tokens": 10938435.0,
      "step": 4905
    },
    {
      "entropy": 5.671151447296142,
      "epoch": 0.551468523614309,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004978141630339863,
      "loss": 5.6205,
      "mean_token_accuracy": 0.17252751886844636,
      "num_tokens": 10949118.0,
      "step": 4910
    },
    {
      "entropy": 5.777276611328125,
      "epoch": 0.5520301005222665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978085767523184,
      "loss": 5.657,
      "mean_token_accuracy": 0.17057308554649353,
      "num_tokens": 10960120.0,
      "step": 4915
    },
    {
      "entropy": 5.742342805862426,
      "epoch": 0.5525916774302241,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497802983376339,
      "loss": 5.6397,
      "mean_token_accuracy": 0.17520381212234498,
      "num_tokens": 10971210.0,
      "step": 4920
    },
    {
      "entropy": 5.655767631530762,
      "epoch": 0.5531532543381816,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004977973829062263,
      "loss": 5.6121,
      "mean_token_accuracy": 0.1807950958609581,
      "num_tokens": 10983446.0,
      "step": 4925
    },
    {
      "entropy": 5.819978809356689,
      "epoch": 0.5537148312461392,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004977917753421585,
      "loss": 5.7389,
      "mean_token_accuracy": 0.16750801503658294,
      "num_tokens": 10994776.0,
      "step": 4930
    },
    {
      "entropy": 5.725039482116699,
      "epoch": 0.5542764081540967,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000497786160684314,
      "loss": 5.6497,
      "mean_token_accuracy": 0.16738180369138717,
      "num_tokens": 11005280.0,
      "step": 4935
    },
    {
      "entropy": 5.634877109527588,
      "epoch": 0.5548379850620543,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004977805389328719,
      "loss": 5.6471,
      "mean_token_accuracy": 0.1671821042895317,
      "num_tokens": 11017509.0,
      "step": 4940
    },
    {
      "entropy": 5.773461437225341,
      "epoch": 0.5553995619700118,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000497774910088011,
      "loss": 5.6786,
      "mean_token_accuracy": 0.17146782875061034,
      "num_tokens": 11029087.0,
      "step": 4945
    },
    {
      "entropy": 5.748759794235229,
      "epoch": 0.5559611388779694,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004977692741499106,
      "loss": 5.7066,
      "mean_token_accuracy": 0.16947160810232162,
      "num_tokens": 11041144.0,
      "step": 4950
    },
    {
      "entropy": 5.878700208663941,
      "epoch": 0.5565227157859269,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049776363111875,
      "loss": 5.7887,
      "mean_token_accuracy": 0.16643262207508086,
      "num_tokens": 11054050.0,
      "step": 4955
    },
    {
      "entropy": 5.735003042221069,
      "epoch": 0.5570842926938844,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004977579809947091,
      "loss": 5.6076,
      "mean_token_accuracy": 0.17405114471912383,
      "num_tokens": 11064257.0,
      "step": 4960
    },
    {
      "entropy": 5.790982389450074,
      "epoch": 0.5576458696018419,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004977523237779676,
      "loss": 5.7207,
      "mean_token_accuracy": 0.17392538487911224,
      "num_tokens": 11075568.0,
      "step": 4965
    },
    {
      "entropy": 5.759704113006592,
      "epoch": 0.5582074465097995,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004977466594687057,
      "loss": 5.6607,
      "mean_token_accuracy": 0.17160382568836213,
      "num_tokens": 11087078.0,
      "step": 4970
    },
    {
      "entropy": 5.8532062530517575,
      "epoch": 0.558769023417757,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004977409880671039,
      "loss": 5.8178,
      "mean_token_accuracy": 0.16156835705041886,
      "num_tokens": 11100655.0,
      "step": 4975
    },
    {
      "entropy": 5.802464723587036,
      "epoch": 0.5593306003257146,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004977353095733427,
      "loss": 5.7037,
      "mean_token_accuracy": 0.170867520570755,
      "num_tokens": 11110675.0,
      "step": 4980
    },
    {
      "entropy": 5.717938423156738,
      "epoch": 0.5598921772336721,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004977296239876026,
      "loss": 5.712,
      "mean_token_accuracy": 0.16717755943536758,
      "num_tokens": 11122337.0,
      "step": 4985
    },
    {
      "entropy": 5.822988271713257,
      "epoch": 0.5604537541416297,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000497723931310065,
      "loss": 5.6137,
      "mean_token_accuracy": 0.17525150924921035,
      "num_tokens": 11132672.0,
      "step": 4990
    },
    {
      "entropy": 5.674412488937378,
      "epoch": 0.5610153310495872,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004977182315409111,
      "loss": 5.6591,
      "mean_token_accuracy": 0.1736418917775154,
      "num_tokens": 11143740.0,
      "step": 4995
    },
    {
      "entropy": 5.701447057723999,
      "epoch": 0.5615769079575448,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004977125246803224,
      "loss": 5.6127,
      "mean_token_accuracy": 0.17458404526114463,
      "num_tokens": 11153722.0,
      "step": 5000
    },
    {
      "entropy": 5.794809436798095,
      "epoch": 0.5621384848655023,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004977068107284803,
      "loss": 5.7267,
      "mean_token_accuracy": 0.16622802317142488,
      "num_tokens": 11165042.0,
      "step": 5005
    },
    {
      "entropy": 5.705378532409668,
      "epoch": 0.5627000617734599,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004977010896855672,
      "loss": 5.5937,
      "mean_token_accuracy": 0.1812054619193077,
      "num_tokens": 11176834.0,
      "step": 5010
    },
    {
      "entropy": 5.714614629745483,
      "epoch": 0.5632616386814174,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004976953615517649,
      "loss": 5.6806,
      "mean_token_accuracy": 0.17363234609365463,
      "num_tokens": 11188442.0,
      "step": 5015
    },
    {
      "entropy": 5.6922242641448975,
      "epoch": 0.563823215589375,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000497689626327256,
      "loss": 5.5743,
      "mean_token_accuracy": 0.17609855234622956,
      "num_tokens": 11197855.0,
      "step": 5020
    },
    {
      "entropy": 5.740423107147217,
      "epoch": 0.5643847924973325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497683884012223,
      "loss": 5.6819,
      "mean_token_accuracy": 0.16573509573936462,
      "num_tokens": 11209300.0,
      "step": 5025
    },
    {
      "entropy": 5.729487133026123,
      "epoch": 0.5649463694052901,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004976781346068487,
      "loss": 5.6998,
      "mean_token_accuracy": 0.1707627922296524,
      "num_tokens": 11222175.0,
      "step": 5030
    },
    {
      "entropy": 5.830688619613648,
      "epoch": 0.5655079463132476,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976723781113163,
      "loss": 5.6957,
      "mean_token_accuracy": 0.16623380333185195,
      "num_tokens": 11233277.0,
      "step": 5035
    },
    {
      "entropy": 5.7766283512115475,
      "epoch": 0.5660695232212052,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004976666145258089,
      "loss": 5.7199,
      "mean_token_accuracy": 0.1678628459572792,
      "num_tokens": 11245241.0,
      "step": 5040
    },
    {
      "entropy": 5.7631550312042235,
      "epoch": 0.5666311001291627,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004976608438505103,
      "loss": 5.7385,
      "mean_token_accuracy": 0.17289155423641206,
      "num_tokens": 11257359.0,
      "step": 5045
    },
    {
      "entropy": 5.713629865646363,
      "epoch": 0.5671926770371203,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976550660856041,
      "loss": 5.6593,
      "mean_token_accuracy": 0.17672290354967118,
      "num_tokens": 11270651.0,
      "step": 5050
    },
    {
      "entropy": 5.727716875076294,
      "epoch": 0.5677542539450777,
      "grad_norm": 1.0,
      "learning_rate": 0.000497649281231274,
      "loss": 5.63,
      "mean_token_accuracy": 0.1713034063577652,
      "num_tokens": 11280164.0,
      "step": 5055
    },
    {
      "entropy": 5.755387592315674,
      "epoch": 0.5683158308530353,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004976434892877046,
      "loss": 5.586,
      "mean_token_accuracy": 0.17397502511739732,
      "num_tokens": 11290156.0,
      "step": 5060
    },
    {
      "entropy": 5.699924278259277,
      "epoch": 0.5688774077609928,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004976376902550802,
      "loss": 5.6365,
      "mean_token_accuracy": 0.1798318862915039,
      "num_tokens": 11301929.0,
      "step": 5065
    },
    {
      "entropy": 5.655545091629028,
      "epoch": 0.5694389846689504,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976318841335854,
      "loss": 5.5666,
      "mean_token_accuracy": 0.1807105928659439,
      "num_tokens": 11312805.0,
      "step": 5070
    },
    {
      "entropy": 5.56551160812378,
      "epoch": 0.5700005615769079,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004976260709234049,
      "loss": 5.4619,
      "mean_token_accuracy": 0.18117846846580504,
      "num_tokens": 11322316.0,
      "step": 5075
    },
    {
      "entropy": 5.671806049346924,
      "epoch": 0.5705621384848655,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497620250624724,
      "loss": 5.6488,
      "mean_token_accuracy": 0.1746806189417839,
      "num_tokens": 11332994.0,
      "step": 5080
    },
    {
      "entropy": 5.757491254806519,
      "epoch": 0.571123715392823,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000497614423237728,
      "loss": 5.6294,
      "mean_token_accuracy": 0.17021082043647767,
      "num_tokens": 11344355.0,
      "step": 5085
    },
    {
      "entropy": 5.715160655975342,
      "epoch": 0.5716852923007806,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004976085887626026,
      "loss": 5.6708,
      "mean_token_accuracy": 0.16981630623340607,
      "num_tokens": 11356275.0,
      "step": 5090
    },
    {
      "entropy": 5.7371093273162845,
      "epoch": 0.5722468692087381,
      "grad_norm": 0.875,
      "learning_rate": 0.0004976027471995331,
      "loss": 5.6543,
      "mean_token_accuracy": 0.1713169738650322,
      "num_tokens": 11368624.0,
      "step": 5095
    },
    {
      "entropy": 5.846412563323975,
      "epoch": 0.5728084461166957,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000497596898548706,
      "loss": 5.7707,
      "mean_token_accuracy": 0.17011373490095139,
      "num_tokens": 11380520.0,
      "step": 5100
    },
    {
      "entropy": 5.677037048339844,
      "epoch": 0.5733700230246532,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975910428103072,
      "loss": 5.6728,
      "mean_token_accuracy": 0.17191081047058104,
      "num_tokens": 11392293.0,
      "step": 5105
    },
    {
      "entropy": 5.776775360107422,
      "epoch": 0.5739315999326108,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004975851799845233,
      "loss": 5.7284,
      "mean_token_accuracy": 0.16902240514755248,
      "num_tokens": 11403866.0,
      "step": 5110
    },
    {
      "entropy": 5.731493425369263,
      "epoch": 0.5744931768405683,
      "grad_norm": 0.875,
      "learning_rate": 0.000497579310071541,
      "loss": 5.5573,
      "mean_token_accuracy": 0.18312533646821977,
      "num_tokens": 11415334.0,
      "step": 5115
    },
    {
      "entropy": 5.651573848724365,
      "epoch": 0.5750547537485259,
      "grad_norm": 0.875,
      "learning_rate": 0.0004975734330715471,
      "loss": 5.6967,
      "mean_token_accuracy": 0.1741857036948204,
      "num_tokens": 11426840.0,
      "step": 5120
    },
    {
      "entropy": 5.840533256530762,
      "epoch": 0.5756163306564834,
      "grad_norm": 0.875,
      "learning_rate": 0.0004975675489847287,
      "loss": 5.8478,
      "mean_token_accuracy": 0.16357955038547517,
      "num_tokens": 11438497.0,
      "step": 5125
    },
    {
      "entropy": 5.727730226516724,
      "epoch": 0.576177907564441,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004975616578112734,
      "loss": 5.5642,
      "mean_token_accuracy": 0.177400603890419,
      "num_tokens": 11448705.0,
      "step": 5130
    },
    {
      "entropy": 5.817451763153076,
      "epoch": 0.5767394844723985,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975557595513684,
      "loss": 5.8326,
      "mean_token_accuracy": 0.1604374922811985,
      "num_tokens": 11460610.0,
      "step": 5135
    },
    {
      "entropy": 5.747437524795532,
      "epoch": 0.5773010613803561,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004975498542052018,
      "loss": 5.5884,
      "mean_token_accuracy": 0.17517827302217484,
      "num_tokens": 11471170.0,
      "step": 5140
    },
    {
      "entropy": 5.702098035812378,
      "epoch": 0.5778626382883136,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004975439417729615,
      "loss": 5.578,
      "mean_token_accuracy": 0.18181243389844895,
      "num_tokens": 11482090.0,
      "step": 5145
    },
    {
      "entropy": 5.636244106292724,
      "epoch": 0.5784242151962711,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004975380222548358,
      "loss": 5.6729,
      "mean_token_accuracy": 0.1727846696972847,
      "num_tokens": 11493650.0,
      "step": 5150
    },
    {
      "entropy": 5.775764989852905,
      "epoch": 0.5789857921042286,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004975320956510131,
      "loss": 5.6583,
      "mean_token_accuracy": 0.16539878398180008,
      "num_tokens": 11505903.0,
      "step": 5155
    },
    {
      "entropy": 5.8385509014129635,
      "epoch": 0.5795473690121862,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004975261619616822,
      "loss": 5.7001,
      "mean_token_accuracy": 0.16813577711582184,
      "num_tokens": 11517194.0,
      "step": 5160
    },
    {
      "entropy": 5.7446431636810305,
      "epoch": 0.5801089459201437,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004975202211870321,
      "loss": 5.7074,
      "mean_token_accuracy": 0.17194776237010956,
      "num_tokens": 11528683.0,
      "step": 5165
    },
    {
      "entropy": 5.736820793151855,
      "epoch": 0.5806705228281013,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004975142733272518,
      "loss": 5.7343,
      "mean_token_accuracy": 0.17425447255373,
      "num_tokens": 11541940.0,
      "step": 5170
    },
    {
      "entropy": 5.717590761184693,
      "epoch": 0.5812320997360588,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004975083183825306,
      "loss": 5.6288,
      "mean_token_accuracy": 0.17126648128032684,
      "num_tokens": 11553045.0,
      "step": 5175
    },
    {
      "entropy": 5.62059440612793,
      "epoch": 0.5817936766440164,
      "grad_norm": 0.875,
      "learning_rate": 0.0004975023563530584,
      "loss": 5.5967,
      "mean_token_accuracy": 0.17394222617149352,
      "num_tokens": 11564401.0,
      "step": 5180
    },
    {
      "entropy": 5.729012060165405,
      "epoch": 0.5823552535519739,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004974963872390249,
      "loss": 5.6396,
      "mean_token_accuracy": 0.17092307806015014,
      "num_tokens": 11576424.0,
      "step": 5185
    },
    {
      "entropy": 5.755575132369995,
      "epoch": 0.5829168304599315,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00049749041104062,
      "loss": 5.6283,
      "mean_token_accuracy": 0.17128757536411285,
      "num_tokens": 11587853.0,
      "step": 5190
    },
    {
      "entropy": 5.693592929840088,
      "epoch": 0.583478407367889,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974844277580343,
      "loss": 5.6439,
      "mean_token_accuracy": 0.17354430258274078,
      "num_tokens": 11597938.0,
      "step": 5195
    },
    {
      "entropy": 5.937219858169556,
      "epoch": 0.5840399842758466,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004974784373914579,
      "loss": 5.796,
      "mean_token_accuracy": 0.17016382068395614,
      "num_tokens": 11610077.0,
      "step": 5200
    },
    {
      "entropy": 5.715516757965088,
      "epoch": 0.5846015611838041,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004974724399410818,
      "loss": 5.6748,
      "mean_token_accuracy": 0.1713270962238312,
      "num_tokens": 11622079.0,
      "step": 5205
    },
    {
      "entropy": 5.604576063156128,
      "epoch": 0.5851631380917617,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004974664354070971,
      "loss": 5.5945,
      "mean_token_accuracy": 0.17965344488620758,
      "num_tokens": 11633047.0,
      "step": 5210
    },
    {
      "entropy": 5.649144887924194,
      "epoch": 0.5857247149997192,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004974604237896947,
      "loss": 5.5514,
      "mean_token_accuracy": 0.17715423107147216,
      "num_tokens": 11643858.0,
      "step": 5215
    },
    {
      "entropy": 5.6603387832641605,
      "epoch": 0.5862862919076768,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974544050890662,
      "loss": 5.5828,
      "mean_token_accuracy": 0.1777600809931755,
      "num_tokens": 11654550.0,
      "step": 5220
    },
    {
      "entropy": 5.6631739139556885,
      "epoch": 0.5868478688156343,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000497448379305403,
      "loss": 5.6081,
      "mean_token_accuracy": 0.17178307473659515,
      "num_tokens": 11665787.0,
      "step": 5225
    },
    {
      "entropy": 5.768941593170166,
      "epoch": 0.5874094457235919,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004974423464388973,
      "loss": 5.7174,
      "mean_token_accuracy": 0.16885657459497452,
      "num_tokens": 11677717.0,
      "step": 5230
    },
    {
      "entropy": 5.720490026473999,
      "epoch": 0.5879710226315494,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497436306489741,
      "loss": 5.6263,
      "mean_token_accuracy": 0.17350273728370666,
      "num_tokens": 11687714.0,
      "step": 5235
    },
    {
      "entropy": 5.643569993972778,
      "epoch": 0.588532599539507,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974302594581264,
      "loss": 5.4938,
      "mean_token_accuracy": 0.18226350098848343,
      "num_tokens": 11698777.0,
      "step": 5240
    },
    {
      "entropy": 5.668469762802124,
      "epoch": 0.5890941764474645,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004974242053442461,
      "loss": 5.607,
      "mean_token_accuracy": 0.1782697170972824,
      "num_tokens": 11710922.0,
      "step": 5245
    },
    {
      "entropy": 5.705522632598877,
      "epoch": 0.589655753355422,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004974181441482927,
      "loss": 5.6433,
      "mean_token_accuracy": 0.17290348261594773,
      "num_tokens": 11721450.0,
      "step": 5250
    },
    {
      "entropy": 5.651781797409058,
      "epoch": 0.5902173302633795,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004974120758704595,
      "loss": 5.5491,
      "mean_token_accuracy": 0.1756184220314026,
      "num_tokens": 11732458.0,
      "step": 5255
    },
    {
      "entropy": 5.618138551712036,
      "epoch": 0.5907789071713371,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004974060005109394,
      "loss": 5.5286,
      "mean_token_accuracy": 0.18139473646879195,
      "num_tokens": 11742649.0,
      "step": 5260
    },
    {
      "entropy": 5.63271131515503,
      "epoch": 0.5913404840792946,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000497399918069926,
      "loss": 5.5522,
      "mean_token_accuracy": 0.18027863204479216,
      "num_tokens": 11753811.0,
      "step": 5265
    },
    {
      "entropy": 5.733545780181885,
      "epoch": 0.5919020609872522,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004973938285476131,
      "loss": 5.6592,
      "mean_token_accuracy": 0.16973895877599715,
      "num_tokens": 11763965.0,
      "step": 5270
    },
    {
      "entropy": 5.8242546081542965,
      "epoch": 0.5924636378952097,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004973877319441943,
      "loss": 5.6572,
      "mean_token_accuracy": 0.18044903129339218,
      "num_tokens": 11774323.0,
      "step": 5275
    },
    {
      "entropy": 5.680837869644165,
      "epoch": 0.5930252148031673,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004973816282598639,
      "loss": 5.6447,
      "mean_token_accuracy": 0.1720179185271263,
      "num_tokens": 11786157.0,
      "step": 5280
    },
    {
      "entropy": 5.722670650482177,
      "epoch": 0.5935867917111248,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004973755174948162,
      "loss": 5.6275,
      "mean_token_accuracy": 0.17616336196660995,
      "num_tokens": 11797344.0,
      "step": 5285
    },
    {
      "entropy": 5.539758396148682,
      "epoch": 0.5941483686190824,
      "grad_norm": 1.0,
      "learning_rate": 0.0004973693996492458,
      "loss": 5.5538,
      "mean_token_accuracy": 0.17561053335666657,
      "num_tokens": 11808149.0,
      "step": 5290
    },
    {
      "entropy": 5.658942747116089,
      "epoch": 0.5947099455270399,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004973632747233475,
      "loss": 5.5717,
      "mean_token_accuracy": 0.17923365384340287,
      "num_tokens": 11820203.0,
      "step": 5295
    },
    {
      "entropy": 5.755408239364624,
      "epoch": 0.5952715224349975,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004973571427173163,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17345385253429413,
      "num_tokens": 11830172.0,
      "step": 5300
    },
    {
      "entropy": 5.731786251068115,
      "epoch": 0.595833099342955,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004973510036313473,
      "loss": 5.6034,
      "mean_token_accuracy": 0.18058293759822847,
      "num_tokens": 11840203.0,
      "step": 5305
    },
    {
      "entropy": 5.706607627868652,
      "epoch": 0.5963946762509126,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004973448574656362,
      "loss": 5.5648,
      "mean_token_accuracy": 0.17356404513120652,
      "num_tokens": 11851002.0,
      "step": 5310
    },
    {
      "entropy": 5.624784660339356,
      "epoch": 0.5969562531588701,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004973387042203786,
      "loss": 5.5772,
      "mean_token_accuracy": 0.17031607925891876,
      "num_tokens": 11861559.0,
      "step": 5315
    },
    {
      "entropy": 5.6799314498901365,
      "epoch": 0.5975178300668277,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004973325438957704,
      "loss": 5.6395,
      "mean_token_accuracy": 0.17312609702348708,
      "num_tokens": 11872535.0,
      "step": 5320
    },
    {
      "entropy": 5.7497875690460205,
      "epoch": 0.5980794069747852,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004973263764920079,
      "loss": 5.659,
      "mean_token_accuracy": 0.1717180833220482,
      "num_tokens": 11884308.0,
      "step": 5325
    },
    {
      "entropy": 5.81054277420044,
      "epoch": 0.5986409838827428,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004973202020092871,
      "loss": 5.7253,
      "mean_token_accuracy": 0.17106300741434097,
      "num_tokens": 11896136.0,
      "step": 5330
    },
    {
      "entropy": 5.704445934295654,
      "epoch": 0.5992025607907003,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497314020447805,
      "loss": 5.555,
      "mean_token_accuracy": 0.18488236963748933,
      "num_tokens": 11906497.0,
      "step": 5335
    },
    {
      "entropy": 5.71569709777832,
      "epoch": 0.5997641376986579,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004973078318077583,
      "loss": 5.6882,
      "mean_token_accuracy": 0.16954067945480347,
      "num_tokens": 11917872.0,
      "step": 5340
    },
    {
      "entropy": 5.770382022857666,
      "epoch": 0.6003257146066153,
      "grad_norm": 1.0,
      "learning_rate": 0.000497301636089344,
      "loss": 5.7133,
      "mean_token_accuracy": 0.16962868273258208,
      "num_tokens": 11929727.0,
      "step": 5345
    },
    {
      "entropy": 5.726130390167237,
      "epoch": 0.6008872915145729,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004972954332927592,
      "loss": 5.4884,
      "mean_token_accuracy": 0.17450084090232848,
      "num_tokens": 11939996.0,
      "step": 5350
    },
    {
      "entropy": 5.598533058166504,
      "epoch": 0.6014488684225304,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004972892234182018,
      "loss": 5.5753,
      "mean_token_accuracy": 0.18218182176351547,
      "num_tokens": 11950683.0,
      "step": 5355
    },
    {
      "entropy": 5.614047431945801,
      "epoch": 0.602010445330488,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004972830064658692,
      "loss": 5.5811,
      "mean_token_accuracy": 0.17111939787864686,
      "num_tokens": 11962447.0,
      "step": 5360
    },
    {
      "entropy": 5.748089361190796,
      "epoch": 0.6025720222384455,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004972767824359596,
      "loss": 5.6663,
      "mean_token_accuracy": 0.17336540073156356,
      "num_tokens": 11972740.0,
      "step": 5365
    },
    {
      "entropy": 5.6867390155792235,
      "epoch": 0.6031335991464031,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004972705513286709,
      "loss": 5.5378,
      "mean_token_accuracy": 0.17462631613016127,
      "num_tokens": 11983292.0,
      "step": 5370
    },
    {
      "entropy": 5.741811800003052,
      "epoch": 0.6036951760543606,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004972643131442017,
      "loss": 5.7533,
      "mean_token_accuracy": 0.16546933352947235,
      "num_tokens": 11996384.0,
      "step": 5375
    },
    {
      "entropy": 5.790655422210693,
      "epoch": 0.6042567529623182,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004972580678827506,
      "loss": 5.6973,
      "mean_token_accuracy": 0.1804477110505104,
      "num_tokens": 12006938.0,
      "step": 5380
    },
    {
      "entropy": 5.586878156661987,
      "epoch": 0.6048183298702757,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972518155445163,
      "loss": 5.4956,
      "mean_token_accuracy": 0.17651916891336442,
      "num_tokens": 12018729.0,
      "step": 5385
    },
    {
      "entropy": 5.913886404037475,
      "epoch": 0.6053799067782333,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004972455561296982,
      "loss": 5.8358,
      "mean_token_accuracy": 0.16637750566005707,
      "num_tokens": 12029625.0,
      "step": 5390
    },
    {
      "entropy": 5.72364821434021,
      "epoch": 0.6059414836861908,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972392896384952,
      "loss": 5.5423,
      "mean_token_accuracy": 0.1724603295326233,
      "num_tokens": 12040243.0,
      "step": 5395
    },
    {
      "entropy": 5.774748659133911,
      "epoch": 0.6065030605941484,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004972330160711071,
      "loss": 5.8269,
      "mean_token_accuracy": 0.16337401419878006,
      "num_tokens": 12051961.0,
      "step": 5400
    },
    {
      "entropy": 5.790932512283325,
      "epoch": 0.6070646375021059,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004972267354277336,
      "loss": 5.6903,
      "mean_token_accuracy": 0.1754745662212372,
      "num_tokens": 12063969.0,
      "step": 5405
    },
    {
      "entropy": 5.711603879928589,
      "epoch": 0.6076262144100635,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004972204477085746,
      "loss": 5.5856,
      "mean_token_accuracy": 0.17822082340717316,
      "num_tokens": 12075682.0,
      "step": 5410
    },
    {
      "entropy": 5.753958749771118,
      "epoch": 0.608187791318021,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004972141529138303,
      "loss": 5.7246,
      "mean_token_accuracy": 0.168320532143116,
      "num_tokens": 12086595.0,
      "step": 5415
    },
    {
      "entropy": 5.753818988800049,
      "epoch": 0.6087493682259786,
      "grad_norm": 1.0,
      "learning_rate": 0.0004972078510437013,
      "loss": 5.5994,
      "mean_token_accuracy": 0.17213478088378906,
      "num_tokens": 12099421.0,
      "step": 5420
    },
    {
      "entropy": 5.753985404968262,
      "epoch": 0.6093109451339361,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004972015420983882,
      "loss": 5.706,
      "mean_token_accuracy": 0.16944072097539903,
      "num_tokens": 12109680.0,
      "step": 5425
    },
    {
      "entropy": 5.648378658294678,
      "epoch": 0.6098725220418937,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004971952260780915,
      "loss": 5.5771,
      "mean_token_accuracy": 0.18387772440910338,
      "num_tokens": 12121196.0,
      "step": 5430
    },
    {
      "entropy": 5.64174075126648,
      "epoch": 0.6104340989498512,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497188902983013,
      "loss": 5.5387,
      "mean_token_accuracy": 0.16641906052827835,
      "num_tokens": 12132262.0,
      "step": 5435
    },
    {
      "entropy": 5.77960057258606,
      "epoch": 0.6109956758578087,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004971825728133535,
      "loss": 5.7221,
      "mean_token_accuracy": 0.1732988551259041,
      "num_tokens": 12143747.0,
      "step": 5440
    },
    {
      "entropy": 5.90015869140625,
      "epoch": 0.6115572527657662,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004971762355693146,
      "loss": 5.7968,
      "mean_token_accuracy": 0.16648507714271546,
      "num_tokens": 12155887.0,
      "step": 5445
    },
    {
      "entropy": 5.683777236938477,
      "epoch": 0.6121188296737238,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004971698912510983,
      "loss": 5.5861,
      "mean_token_accuracy": 0.17328004390001298,
      "num_tokens": 12166718.0,
      "step": 5450
    },
    {
      "entropy": 5.721788024902343,
      "epoch": 0.6126804065816813,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004971635398589066,
      "loss": 5.7248,
      "mean_token_accuracy": 0.17312407493591309,
      "num_tokens": 12178540.0,
      "step": 5455
    },
    {
      "entropy": 5.88222484588623,
      "epoch": 0.6132419834896389,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004971571813929415,
      "loss": 5.7589,
      "mean_token_accuracy": 0.16493648439645767,
      "num_tokens": 12189820.0,
      "step": 5460
    },
    {
      "entropy": 5.733379125595093,
      "epoch": 0.6138035603975964,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004971508158534056,
      "loss": 5.683,
      "mean_token_accuracy": 0.1734658569097519,
      "num_tokens": 12201679.0,
      "step": 5465
    },
    {
      "entropy": 5.737128114700317,
      "epoch": 0.614365137305554,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004971444432405015,
      "loss": 5.6765,
      "mean_token_accuracy": 0.16735841035842897,
      "num_tokens": 12213793.0,
      "step": 5470
    },
    {
      "entropy": 5.807603883743286,
      "epoch": 0.6149267142135115,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004971380635544323,
      "loss": 5.7514,
      "mean_token_accuracy": 0.16945342123508453,
      "num_tokens": 12224519.0,
      "step": 5475
    },
    {
      "entropy": 5.8681245803833,
      "epoch": 0.6154882911214691,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004971316767954009,
      "loss": 5.7608,
      "mean_token_accuracy": 0.1658295512199402,
      "num_tokens": 12236401.0,
      "step": 5480
    },
    {
      "entropy": 5.783669519424438,
      "epoch": 0.6160498680294266,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004971252829636108,
      "loss": 5.7865,
      "mean_token_accuracy": 0.16470268964767457,
      "num_tokens": 12247670.0,
      "step": 5485
    },
    {
      "entropy": 5.696588134765625,
      "epoch": 0.6166114449373842,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004971188820592656,
      "loss": 5.6372,
      "mean_token_accuracy": 0.17362091988325118,
      "num_tokens": 12258877.0,
      "step": 5490
    },
    {
      "entropy": 5.721828269958496,
      "epoch": 0.6171730218453417,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497112474082569,
      "loss": 5.6085,
      "mean_token_accuracy": 0.17151312828063964,
      "num_tokens": 12268507.0,
      "step": 5495
    },
    {
      "entropy": 5.705202722549439,
      "epoch": 0.6177345987532993,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000497106059033725,
      "loss": 5.606,
      "mean_token_accuracy": 0.17059515416622162,
      "num_tokens": 12280420.0,
      "step": 5500
    },
    {
      "entropy": 5.597872352600097,
      "epoch": 0.6182961756612568,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004970996369129381,
      "loss": 5.5981,
      "mean_token_accuracy": 0.17261921167373656,
      "num_tokens": 12290876.0,
      "step": 5505
    },
    {
      "entropy": 5.676345777511597,
      "epoch": 0.6188577525692144,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004970932077204125,
      "loss": 5.5022,
      "mean_token_accuracy": 0.18050920814275742,
      "num_tokens": 12301831.0,
      "step": 5510
    },
    {
      "entropy": 5.628951120376587,
      "epoch": 0.6194193294771719,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497086771456353,
      "loss": 5.5275,
      "mean_token_accuracy": 0.17723134756088257,
      "num_tokens": 12312678.0,
      "step": 5515
    },
    {
      "entropy": 5.774626207351685,
      "epoch": 0.6199809063851295,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004970803281209646,
      "loss": 5.7205,
      "mean_token_accuracy": 0.16646523922681808,
      "num_tokens": 12322905.0,
      "step": 5520
    },
    {
      "entropy": 5.714275932312011,
      "epoch": 0.620542483293087,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004970738777144524,
      "loss": 5.5894,
      "mean_token_accuracy": 0.1710187703371048,
      "num_tokens": 12333091.0,
      "step": 5525
    },
    {
      "entropy": 5.697361612319947,
      "epoch": 0.6211040602010446,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004970674202370218,
      "loss": 5.5844,
      "mean_token_accuracy": 0.17744589745998382,
      "num_tokens": 12346310.0,
      "step": 5530
    },
    {
      "entropy": 5.6302886486053465,
      "epoch": 0.621665637109002,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004970609556888785,
      "loss": 5.5137,
      "mean_token_accuracy": 0.17561311572790145,
      "num_tokens": 12357138.0,
      "step": 5535
    },
    {
      "entropy": 5.731592130661011,
      "epoch": 0.6222272140169596,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004970544840702282,
      "loss": 5.6047,
      "mean_token_accuracy": 0.1761130392551422,
      "num_tokens": 12367382.0,
      "step": 5540
    },
    {
      "entropy": 5.6389649391174315,
      "epoch": 0.6227887909249171,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004970480053812771,
      "loss": 5.5956,
      "mean_token_accuracy": 0.17054942697286607,
      "num_tokens": 12380357.0,
      "step": 5545
    },
    {
      "entropy": 5.804252243041992,
      "epoch": 0.6233503678328747,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004970415196222312,
      "loss": 5.7713,
      "mean_token_accuracy": 0.16149107813835145,
      "num_tokens": 12392187.0,
      "step": 5550
    },
    {
      "entropy": 5.722889518737793,
      "epoch": 0.6239119447408322,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004970350267932973,
      "loss": 5.633,
      "mean_token_accuracy": 0.17469703853130342,
      "num_tokens": 12402924.0,
      "step": 5555
    },
    {
      "entropy": 5.734255170822143,
      "epoch": 0.6244735216487898,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497028526894682,
      "loss": 5.7521,
      "mean_token_accuracy": 0.16999632269144058,
      "num_tokens": 12413554.0,
      "step": 5560
    },
    {
      "entropy": 5.798130083084106,
      "epoch": 0.6250350985567473,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004970220199265923,
      "loss": 5.6273,
      "mean_token_accuracy": 0.1768548682332039,
      "num_tokens": 12424525.0,
      "step": 5565
    },
    {
      "entropy": 5.680708885192871,
      "epoch": 0.6255966754647049,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004970155058892354,
      "loss": 5.6322,
      "mean_token_accuracy": 0.17046258449554444,
      "num_tokens": 12436405.0,
      "step": 5570
    },
    {
      "entropy": 5.718565797805786,
      "epoch": 0.6261582523726624,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004970089847828187,
      "loss": 5.7037,
      "mean_token_accuracy": 0.17012005001306535,
      "num_tokens": 12446909.0,
      "step": 5575
    },
    {
      "entropy": 5.737675189971924,
      "epoch": 0.62671982928062,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004970024566075497,
      "loss": 5.6804,
      "mean_token_accuracy": 0.1764724850654602,
      "num_tokens": 12458739.0,
      "step": 5580
    },
    {
      "entropy": 5.707290410995483,
      "epoch": 0.6272814061885775,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004969959213636365,
      "loss": 5.6001,
      "mean_token_accuracy": 0.17520488798618317,
      "num_tokens": 12469529.0,
      "step": 5585
    },
    {
      "entropy": 5.761801242828369,
      "epoch": 0.6278429830965351,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004969893790512871,
      "loss": 5.6483,
      "mean_token_accuracy": 0.16846525967121123,
      "num_tokens": 12480067.0,
      "step": 5590
    },
    {
      "entropy": 5.673601484298706,
      "epoch": 0.6284045600044926,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004969828296707097,
      "loss": 5.5278,
      "mean_token_accuracy": 0.1837966412305832,
      "num_tokens": 12491043.0,
      "step": 5595
    },
    {
      "entropy": 5.561583518981934,
      "epoch": 0.6289661369124502,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004969762732221129,
      "loss": 5.4627,
      "mean_token_accuracy": 0.1810373306274414,
      "num_tokens": 12501852.0,
      "step": 5600
    },
    {
      "entropy": 5.743855667114258,
      "epoch": 0.6295277138204077,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004969697097057054,
      "loss": 5.7447,
      "mean_token_accuracy": 0.1711232617497444,
      "num_tokens": 12512696.0,
      "step": 5605
    },
    {
      "entropy": 5.74007658958435,
      "epoch": 0.6300892907283653,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004969631391216964,
      "loss": 5.6186,
      "mean_token_accuracy": 0.1756168007850647,
      "num_tokens": 12523643.0,
      "step": 5610
    },
    {
      "entropy": 5.71359224319458,
      "epoch": 0.6306508676363228,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004969565614702948,
      "loss": 5.5883,
      "mean_token_accuracy": 0.1746621236205101,
      "num_tokens": 12534827.0,
      "step": 5615
    },
    {
      "entropy": 5.563238573074341,
      "epoch": 0.6312124445442804,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004969499767517104,
      "loss": 5.4719,
      "mean_token_accuracy": 0.17915562987327577,
      "num_tokens": 12545977.0,
      "step": 5620
    },
    {
      "entropy": 5.544950914382935,
      "epoch": 0.6317740214522379,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004969433849661525,
      "loss": 5.5384,
      "mean_token_accuracy": 0.1715235486626625,
      "num_tokens": 12555890.0,
      "step": 5625
    },
    {
      "entropy": 5.698396301269531,
      "epoch": 0.6323355983601955,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004969367861138313,
      "loss": 5.6064,
      "mean_token_accuracy": 0.17803364992141724,
      "num_tokens": 12566412.0,
      "step": 5630
    },
    {
      "entropy": 5.7760552883148195,
      "epoch": 0.6328971752681529,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004969301801949565,
      "loss": 5.6302,
      "mean_token_accuracy": 0.17108214646577835,
      "num_tokens": 12577175.0,
      "step": 5635
    },
    {
      "entropy": 5.595721292495727,
      "epoch": 0.6334587521761105,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000496923567209739,
      "loss": 5.5375,
      "mean_token_accuracy": 0.17472511678934097,
      "num_tokens": 12588760.0,
      "step": 5640
    },
    {
      "entropy": 5.609034824371338,
      "epoch": 0.634020329084068,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004969169471583889,
      "loss": 5.6121,
      "mean_token_accuracy": 0.1745341196656227,
      "num_tokens": 12599899.0,
      "step": 5645
    },
    {
      "entropy": 5.744676113128662,
      "epoch": 0.6345819059920256,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004969103200411171,
      "loss": 5.6017,
      "mean_token_accuracy": 0.1837252125144005,
      "num_tokens": 12610836.0,
      "step": 5650
    },
    {
      "entropy": 5.61579442024231,
      "epoch": 0.6351434828999831,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004969036858581346,
      "loss": 5.5485,
      "mean_token_accuracy": 0.18136483132839204,
      "num_tokens": 12621855.0,
      "step": 5655
    },
    {
      "entropy": 5.69185848236084,
      "epoch": 0.6357050598079407,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004968970446096527,
      "loss": 5.6491,
      "mean_token_accuracy": 0.17270753532648087,
      "num_tokens": 12632964.0,
      "step": 5660
    },
    {
      "entropy": 5.634363222122192,
      "epoch": 0.6362666367158982,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000496890396295883,
      "loss": 5.4726,
      "mean_token_accuracy": 0.1809499219059944,
      "num_tokens": 12643717.0,
      "step": 5665
    },
    {
      "entropy": 5.70470118522644,
      "epoch": 0.6368282136238558,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004968837409170368,
      "loss": 5.6115,
      "mean_token_accuracy": 0.16996596604585648,
      "num_tokens": 12654491.0,
      "step": 5670
    },
    {
      "entropy": 5.6480659484863285,
      "epoch": 0.6373897905318133,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004968770784733263,
      "loss": 5.645,
      "mean_token_accuracy": 0.17329516261816025,
      "num_tokens": 12666128.0,
      "step": 5675
    },
    {
      "entropy": 5.748110628128051,
      "epoch": 0.6379513674397709,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004968704089649636,
      "loss": 5.7332,
      "mean_token_accuracy": 0.16888405978679658,
      "num_tokens": 12677957.0,
      "step": 5680
    },
    {
      "entropy": 5.756485652923584,
      "epoch": 0.6385129443477284,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004968637323921611,
      "loss": 5.5693,
      "mean_token_accuracy": 0.17527114748954772,
      "num_tokens": 12687964.0,
      "step": 5685
    },
    {
      "entropy": 5.590578317642212,
      "epoch": 0.639074521255686,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004968570487551312,
      "loss": 5.574,
      "mean_token_accuracy": 0.17056965082883835,
      "num_tokens": 12698805.0,
      "step": 5690
    },
    {
      "entropy": 5.7417315483093265,
      "epoch": 0.6396360981636435,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004968503580540869,
      "loss": 5.584,
      "mean_token_accuracy": 0.17162402719259262,
      "num_tokens": 12709942.0,
      "step": 5695
    },
    {
      "entropy": 5.778721570968628,
      "epoch": 0.6401976750716011,
      "grad_norm": 1.0,
      "learning_rate": 0.0004968436602892411,
      "loss": 5.6731,
      "mean_token_accuracy": 0.1761099860072136,
      "num_tokens": 12721856.0,
      "step": 5700
    },
    {
      "entropy": 5.556879568099975,
      "epoch": 0.6407592519795586,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004968369554608071,
      "loss": 5.4929,
      "mean_token_accuracy": 0.17659999579191207,
      "num_tokens": 12732619.0,
      "step": 5705
    },
    {
      "entropy": 5.642232465744018,
      "epoch": 0.6413208288875162,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004968302435689985,
      "loss": 5.578,
      "mean_token_accuracy": 0.1764561727643013,
      "num_tokens": 12743478.0,
      "step": 5710
    },
    {
      "entropy": 5.622849702835083,
      "epoch": 0.6418824057954737,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004968235246140288,
      "loss": 5.5327,
      "mean_token_accuracy": 0.16981768161058425,
      "num_tokens": 12754761.0,
      "step": 5715
    },
    {
      "entropy": 5.762554883956909,
      "epoch": 0.6424439827034313,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004968167985961121,
      "loss": 5.6237,
      "mean_token_accuracy": 0.17977579683065414,
      "num_tokens": 12765803.0,
      "step": 5720
    },
    {
      "entropy": 5.767837190628052,
      "epoch": 0.6430055596113888,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004968100655154625,
      "loss": 5.6577,
      "mean_token_accuracy": 0.17164772003889084,
      "num_tokens": 12777882.0,
      "step": 5725
    },
    {
      "entropy": 5.702736282348633,
      "epoch": 0.6435671365193463,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004968033253722943,
      "loss": 5.6754,
      "mean_token_accuracy": 0.17198235988616944,
      "num_tokens": 12788268.0,
      "step": 5730
    },
    {
      "entropy": 5.629304265975952,
      "epoch": 0.6441287134273038,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004967965781668223,
      "loss": 5.6715,
      "mean_token_accuracy": 0.17235511243343354,
      "num_tokens": 12800506.0,
      "step": 5735
    },
    {
      "entropy": 5.7705975532531735,
      "epoch": 0.6446902903352614,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004967898238992613,
      "loss": 5.6343,
      "mean_token_accuracy": 0.17663808166980743,
      "num_tokens": 12813040.0,
      "step": 5740
    },
    {
      "entropy": 5.780026578903199,
      "epoch": 0.6452518672432189,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004967830625698262,
      "loss": 5.6714,
      "mean_token_accuracy": 0.1773954689502716,
      "num_tokens": 12825534.0,
      "step": 5745
    },
    {
      "entropy": 5.6190712451934814,
      "epoch": 0.6458134441511765,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004967762941787323,
      "loss": 5.5549,
      "mean_token_accuracy": 0.17475787103176116,
      "num_tokens": 12835882.0,
      "step": 5750
    },
    {
      "entropy": 5.759916305541992,
      "epoch": 0.646375021059134,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004967695187261952,
      "loss": 5.7408,
      "mean_token_accuracy": 0.16839111149311065,
      "num_tokens": 12847663.0,
      "step": 5755
    },
    {
      "entropy": 5.961766767501831,
      "epoch": 0.6469365979670916,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004967627362124306,
      "loss": 5.8788,
      "mean_token_accuracy": 0.16633762419223785,
      "num_tokens": 12860789.0,
      "step": 5760
    },
    {
      "entropy": 5.7413105964660645,
      "epoch": 0.6474981748750491,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004967559466376546,
      "loss": 5.6103,
      "mean_token_accuracy": 0.17463897466659545,
      "num_tokens": 12872509.0,
      "step": 5765
    },
    {
      "entropy": 5.628509473800659,
      "epoch": 0.6480597517830067,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004967491500020831,
      "loss": 5.6059,
      "mean_token_accuracy": 0.17851067781448365,
      "num_tokens": 12884321.0,
      "step": 5770
    },
    {
      "entropy": 5.554707098007202,
      "epoch": 0.6486213286909642,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004967423463059328,
      "loss": 5.5081,
      "mean_token_accuracy": 0.17908329516649246,
      "num_tokens": 12894559.0,
      "step": 5775
    },
    {
      "entropy": 5.679509878158569,
      "epoch": 0.6491829055989218,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004967355355494201,
      "loss": 5.6027,
      "mean_token_accuracy": 0.17926470041275025,
      "num_tokens": 12904934.0,
      "step": 5780
    },
    {
      "entropy": 5.823178243637085,
      "epoch": 0.6497444825068793,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496728717732762,
      "loss": 5.7257,
      "mean_token_accuracy": 0.1665002003312111,
      "num_tokens": 12916551.0,
      "step": 5785
    },
    {
      "entropy": 5.698605918884278,
      "epoch": 0.6503060594148369,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004967218928561755,
      "loss": 5.6256,
      "mean_token_accuracy": 0.1717161476612091,
      "num_tokens": 12928097.0,
      "step": 5790
    },
    {
      "entropy": 5.642832279205322,
      "epoch": 0.6508676363227944,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004967150609198779,
      "loss": 5.4937,
      "mean_token_accuracy": 0.18256949335336686,
      "num_tokens": 12939161.0,
      "step": 5795
    },
    {
      "entropy": 5.546928739547729,
      "epoch": 0.651429213230752,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004967082219240869,
      "loss": 5.5389,
      "mean_token_accuracy": 0.1791675016283989,
      "num_tokens": 12950862.0,
      "step": 5800
    },
    {
      "entropy": 5.664335155487061,
      "epoch": 0.6519907901387095,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004967013758690199,
      "loss": 5.5791,
      "mean_token_accuracy": 0.17750437557697296,
      "num_tokens": 12962333.0,
      "step": 5805
    },
    {
      "entropy": 5.673435211181641,
      "epoch": 0.6525523670466671,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004966945227548953,
      "loss": 5.5719,
      "mean_token_accuracy": 0.17666879445314407,
      "num_tokens": 12973136.0,
      "step": 5810
    },
    {
      "entropy": 5.6725842475891115,
      "epoch": 0.6531139439546246,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004966876625819309,
      "loss": 5.5988,
      "mean_token_accuracy": 0.17476778477430344,
      "num_tokens": 12983602.0,
      "step": 5815
    },
    {
      "entropy": 5.700492525100708,
      "epoch": 0.6536755208625822,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004966807953503456,
      "loss": 5.5205,
      "mean_token_accuracy": 0.18146025538444518,
      "num_tokens": 12994129.0,
      "step": 5820
    },
    {
      "entropy": 5.636948108673096,
      "epoch": 0.6542370977705396,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004966739210603578,
      "loss": 5.5857,
      "mean_token_accuracy": 0.17263583242893218,
      "num_tokens": 13005071.0,
      "step": 5825
    },
    {
      "entropy": 5.795610237121582,
      "epoch": 0.6547986746784972,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004966670397121862,
      "loss": 5.7477,
      "mean_token_accuracy": 0.172401587665081,
      "num_tokens": 13016202.0,
      "step": 5830
    },
    {
      "entropy": 5.8175863265991214,
      "epoch": 0.6553602515864547,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004966601513060502,
      "loss": 5.7552,
      "mean_token_accuracy": 0.16232562959194183,
      "num_tokens": 13028468.0,
      "step": 5835
    },
    {
      "entropy": 5.709358787536621,
      "epoch": 0.6559218284944123,
      "grad_norm": 0.984375,
      "learning_rate": 0.000496653255842169,
      "loss": 5.6296,
      "mean_token_accuracy": 0.1753317266702652,
      "num_tokens": 13039667.0,
      "step": 5840
    },
    {
      "entropy": 5.678438663482666,
      "epoch": 0.6564834054023698,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004966463533207621,
      "loss": 5.6781,
      "mean_token_accuracy": 0.16587844491004944,
      "num_tokens": 13051583.0,
      "step": 5845
    },
    {
      "entropy": 5.732926893234253,
      "epoch": 0.6570449823103274,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004966394437420493,
      "loss": 5.6319,
      "mean_token_accuracy": 0.17361444979906082,
      "num_tokens": 13062401.0,
      "step": 5850
    },
    {
      "entropy": 5.604018449783325,
      "epoch": 0.6576065592182849,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004966325271062508,
      "loss": 5.524,
      "mean_token_accuracy": 0.1782347083091736,
      "num_tokens": 13074219.0,
      "step": 5855
    },
    {
      "entropy": 5.75310869216919,
      "epoch": 0.6581681361262425,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004966256034135866,
      "loss": 5.7095,
      "mean_token_accuracy": 0.17001335322856903,
      "num_tokens": 13086073.0,
      "step": 5860
    },
    {
      "entropy": 5.7198251247406,
      "epoch": 0.6587297130342,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004966186726642772,
      "loss": 5.6264,
      "mean_token_accuracy": 0.1701858565211296,
      "num_tokens": 13096340.0,
      "step": 5865
    },
    {
      "entropy": 5.67186975479126,
      "epoch": 0.6592912899421576,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004966117348585434,
      "loss": 5.5863,
      "mean_token_accuracy": 0.17610248178243637,
      "num_tokens": 13107273.0,
      "step": 5870
    },
    {
      "entropy": 5.664875936508179,
      "epoch": 0.6598528668501151,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004966047899966059,
      "loss": 5.5649,
      "mean_token_accuracy": 0.1737078085541725,
      "num_tokens": 13117388.0,
      "step": 5875
    },
    {
      "entropy": 5.7048064231872555,
      "epoch": 0.6604144437580727,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000496597838078686,
      "loss": 5.6793,
      "mean_token_accuracy": 0.1677414134144783,
      "num_tokens": 13128720.0,
      "step": 5880
    },
    {
      "entropy": 5.75440559387207,
      "epoch": 0.6609760206660302,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496590879105005,
      "loss": 5.5945,
      "mean_token_accuracy": 0.17483343333005905,
      "num_tokens": 13139405.0,
      "step": 5885
    },
    {
      "entropy": 5.713056468963623,
      "epoch": 0.6615375975739878,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004965839130757845,
      "loss": 5.6805,
      "mean_token_accuracy": 0.1753469303250313,
      "num_tokens": 13151408.0,
      "step": 5890
    },
    {
      "entropy": 5.69801516532898,
      "epoch": 0.6620991744819453,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004965769399912462,
      "loss": 5.6001,
      "mean_token_accuracy": 0.1750464230775833,
      "num_tokens": 13162592.0,
      "step": 5895
    },
    {
      "entropy": 5.758023262023926,
      "epoch": 0.6626607513899029,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004965699598516122,
      "loss": 5.6202,
      "mean_token_accuracy": 0.1782073572278023,
      "num_tokens": 13173040.0,
      "step": 5900
    },
    {
      "entropy": 5.64671893119812,
      "epoch": 0.6632223282978604,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004965629726571047,
      "loss": 5.69,
      "mean_token_accuracy": 0.1743127301335335,
      "num_tokens": 13185475.0,
      "step": 5905
    },
    {
      "entropy": 5.667902898788452,
      "epoch": 0.663783905205818,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004965559784079463,
      "loss": 5.5685,
      "mean_token_accuracy": 0.17888267785310746,
      "num_tokens": 13196481.0,
      "step": 5910
    },
    {
      "entropy": 5.656558847427368,
      "epoch": 0.6643454821137755,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004965489771043596,
      "loss": 5.4688,
      "mean_token_accuracy": 0.18201548606157303,
      "num_tokens": 13207947.0,
      "step": 5915
    },
    {
      "entropy": 5.50324239730835,
      "epoch": 0.664907059021733,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004965419687465676,
      "loss": 5.5078,
      "mean_token_accuracy": 0.18712942004203797,
      "num_tokens": 13219392.0,
      "step": 5920
    },
    {
      "entropy": 5.573181676864624,
      "epoch": 0.6654686359296905,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004965349533347934,
      "loss": 5.5411,
      "mean_token_accuracy": 0.18074103444814682,
      "num_tokens": 13231556.0,
      "step": 5925
    },
    {
      "entropy": 5.654805421829224,
      "epoch": 0.6660302128376481,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004965279308692602,
      "loss": 5.4946,
      "mean_token_accuracy": 0.18450129479169847,
      "num_tokens": 13241287.0,
      "step": 5930
    },
    {
      "entropy": 5.568438386917114,
      "epoch": 0.6665917897456056,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004965209013501919,
      "loss": 5.5469,
      "mean_token_accuracy": 0.16648050397634506,
      "num_tokens": 13254120.0,
      "step": 5935
    },
    {
      "entropy": 5.712764930725098,
      "epoch": 0.6671533666535632,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004965138647778122,
      "loss": 5.6151,
      "mean_token_accuracy": 0.17329633831977845,
      "num_tokens": 13265745.0,
      "step": 5940
    },
    {
      "entropy": 5.793698453903199,
      "epoch": 0.6677149435615207,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000496506821152345,
      "loss": 5.6867,
      "mean_token_accuracy": 0.17168214321136474,
      "num_tokens": 13277791.0,
      "step": 5945
    },
    {
      "entropy": 5.608066082000732,
      "epoch": 0.6682765204694783,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004964997704740147,
      "loss": 5.5327,
      "mean_token_accuracy": 0.17143328189849855,
      "num_tokens": 13288078.0,
      "step": 5950
    },
    {
      "entropy": 5.670023727416992,
      "epoch": 0.6688380973774358,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000496492712743046,
      "loss": 5.5634,
      "mean_token_accuracy": 0.18261950612068176,
      "num_tokens": 13300547.0,
      "step": 5955
    },
    {
      "entropy": 5.675726366043091,
      "epoch": 0.6693996742853934,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004964856479596632,
      "loss": 5.6515,
      "mean_token_accuracy": 0.17231980115175247,
      "num_tokens": 13310339.0,
      "step": 5960
    },
    {
      "entropy": 5.7539581775665285,
      "epoch": 0.6699612511933509,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004964785761240915,
      "loss": 5.6133,
      "mean_token_accuracy": 0.1755753353238106,
      "num_tokens": 13321294.0,
      "step": 5965
    },
    {
      "entropy": 5.729451990127563,
      "epoch": 0.6705228281013085,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004964714972365562,
      "loss": 5.646,
      "mean_token_accuracy": 0.174198018014431,
      "num_tokens": 13332608.0,
      "step": 5970
    },
    {
      "entropy": 5.684158277511597,
      "epoch": 0.671084405009266,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004964644112972824,
      "loss": 5.585,
      "mean_token_accuracy": 0.18588267415761947,
      "num_tokens": 13344169.0,
      "step": 5975
    },
    {
      "entropy": 5.595675802230835,
      "epoch": 0.6716459819172236,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004964573183064959,
      "loss": 5.5457,
      "mean_token_accuracy": 0.17660977244377135,
      "num_tokens": 13355713.0,
      "step": 5980
    },
    {
      "entropy": 5.7029187202453615,
      "epoch": 0.6722075588251811,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004964502182644225,
      "loss": 5.6341,
      "mean_token_accuracy": 0.17223501354455947,
      "num_tokens": 13366644.0,
      "step": 5985
    },
    {
      "entropy": 5.672948265075684,
      "epoch": 0.6727691357331387,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004964431111712883,
      "loss": 5.6001,
      "mean_token_accuracy": 0.17075798511505128,
      "num_tokens": 13378125.0,
      "step": 5990
    },
    {
      "entropy": 5.659946870803833,
      "epoch": 0.6733307126410962,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004964359970273195,
      "loss": 5.6462,
      "mean_token_accuracy": 0.17172100096940995,
      "num_tokens": 13389681.0,
      "step": 5995
    },
    {
      "entropy": 5.716019535064698,
      "epoch": 0.6738922895490538,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004964288758327426,
      "loss": 5.6374,
      "mean_token_accuracy": 0.1769365757703781,
      "num_tokens": 13399814.0,
      "step": 6000
    },
    {
      "epoch": 0.6738922895490538,
      "eval_entropy": 5.609915955582629,
      "eval_loss": 5.632114887237549,
      "eval_mean_token_accuracy": 0.18043994106619451,
      "eval_num_tokens": 13399814.0,
      "eval_runtime": 59.1053,
      "eval_samples_per_second": 524.707,
      "eval_steps_per_second": 65.595,
      "step": 6000
    },
    {
      "entropy": 5.779499101638794,
      "epoch": 0.6744538664570113,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004964217475877846,
      "loss": 5.6522,
      "mean_token_accuracy": 0.17724189162254333,
      "num_tokens": 13410614.0,
      "step": 6005
    },
    {
      "entropy": 5.697342586517334,
      "epoch": 0.6750154433649689,
      "grad_norm": 1.03125,
      "learning_rate": 0.000496414612292672,
      "loss": 5.6599,
      "mean_token_accuracy": 0.1712150201201439,
      "num_tokens": 13421392.0,
      "step": 6010
    },
    {
      "entropy": 5.715792560577393,
      "epoch": 0.6755770202729264,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004964074699476325,
      "loss": 5.6966,
      "mean_token_accuracy": 0.17103470414876937,
      "num_tokens": 13432004.0,
      "step": 6015
    },
    {
      "entropy": 5.720300006866455,
      "epoch": 0.6761385971808839,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964003205528933,
      "loss": 5.5345,
      "mean_token_accuracy": 0.18039939254522325,
      "num_tokens": 13442450.0,
      "step": 6020
    },
    {
      "entropy": 5.674740743637085,
      "epoch": 0.6767001740888414,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004963931641086819,
      "loss": 5.6041,
      "mean_token_accuracy": 0.17143985629081726,
      "num_tokens": 13453848.0,
      "step": 6025
    },
    {
      "entropy": 5.611522245407104,
      "epoch": 0.677261750996799,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963860006152262,
      "loss": 5.5994,
      "mean_token_accuracy": 0.17509289979934692,
      "num_tokens": 13465119.0,
      "step": 6030
    },
    {
      "entropy": 5.611239814758301,
      "epoch": 0.6778233279047565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004963788300727546,
      "loss": 5.582,
      "mean_token_accuracy": 0.17205574214458466,
      "num_tokens": 13475937.0,
      "step": 6035
    },
    {
      "entropy": 5.642318058013916,
      "epoch": 0.6783849048127141,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004963716524814951,
      "loss": 5.5671,
      "mean_token_accuracy": 0.17940536588430406,
      "num_tokens": 13487430.0,
      "step": 6040
    },
    {
      "entropy": 5.672739744186401,
      "epoch": 0.6789464817206716,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004963644678416763,
      "loss": 5.5517,
      "mean_token_accuracy": 0.17838966995477676,
      "num_tokens": 13499631.0,
      "step": 6045
    },
    {
      "entropy": 5.693057012557984,
      "epoch": 0.6795080586286292,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000496357276153527,
      "loss": 5.629,
      "mean_token_accuracy": 0.1752404034137726,
      "num_tokens": 13511753.0,
      "step": 6050
    },
    {
      "entropy": 5.735835170745849,
      "epoch": 0.6800696355365867,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004963500774172763,
      "loss": 5.6887,
      "mean_token_accuracy": 0.16904593259096146,
      "num_tokens": 13524228.0,
      "step": 6055
    },
    {
      "entropy": 5.823309230804443,
      "epoch": 0.6806312124445443,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000496342871633153,
      "loss": 5.7107,
      "mean_token_accuracy": 0.1734006255865097,
      "num_tokens": 13535700.0,
      "step": 6060
    },
    {
      "entropy": 5.719172239303589,
      "epoch": 0.6811927893525018,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004963356588013871,
      "loss": 5.6879,
      "mean_token_accuracy": 0.1680354133248329,
      "num_tokens": 13547337.0,
      "step": 6065
    },
    {
      "entropy": 5.625253963470459,
      "epoch": 0.6817543662604594,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004963284389222078,
      "loss": 5.5363,
      "mean_token_accuracy": 0.17368890196084977,
      "num_tokens": 13558350.0,
      "step": 6070
    },
    {
      "entropy": 5.730729627609253,
      "epoch": 0.6823159431684169,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004963212119958453,
      "loss": 5.5879,
      "mean_token_accuracy": 0.1806545540690422,
      "num_tokens": 13569230.0,
      "step": 6075
    },
    {
      "entropy": 5.726511764526367,
      "epoch": 0.6828775200763745,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004963139780225295,
      "loss": 5.6743,
      "mean_token_accuracy": 0.17548759877681733,
      "num_tokens": 13580502.0,
      "step": 6080
    },
    {
      "entropy": 5.736097192764282,
      "epoch": 0.683439096984332,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004963067370024909,
      "loss": 5.5845,
      "mean_token_accuracy": 0.17914303243160248,
      "num_tokens": 13591524.0,
      "step": 6085
    },
    {
      "entropy": 5.7245134830474855,
      "epoch": 0.6840006738922896,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004962994889359598,
      "loss": 5.5997,
      "mean_token_accuracy": 0.1826063796877861,
      "num_tokens": 13602432.0,
      "step": 6090
    },
    {
      "entropy": 5.560599994659424,
      "epoch": 0.6845622508002471,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004962922338231673,
      "loss": 5.5253,
      "mean_token_accuracy": 0.18265436589717865,
      "num_tokens": 13613052.0,
      "step": 6095
    },
    {
      "entropy": 5.642383337020874,
      "epoch": 0.6851238277082047,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962849716643441,
      "loss": 5.5741,
      "mean_token_accuracy": 0.17706613391637802,
      "num_tokens": 13623961.0,
      "step": 6100
    },
    {
      "entropy": 5.657310056686401,
      "epoch": 0.6856854046161622,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004962777024597216,
      "loss": 5.556,
      "mean_token_accuracy": 0.1735036551952362,
      "num_tokens": 13635285.0,
      "step": 6105
    },
    {
      "entropy": 5.779331064224243,
      "epoch": 0.6862469815241198,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004962704262095314,
      "loss": 5.7065,
      "mean_token_accuracy": 0.16852724254131318,
      "num_tokens": 13647720.0,
      "step": 6110
    },
    {
      "entropy": 5.626996564865112,
      "epoch": 0.6868085584320772,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004962631429140048,
      "loss": 5.5526,
      "mean_token_accuracy": 0.17468138933181762,
      "num_tokens": 13659726.0,
      "step": 6115
    },
    {
      "entropy": 5.763138675689698,
      "epoch": 0.6873701353400348,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000496255852573374,
      "loss": 5.7382,
      "mean_token_accuracy": 0.16682715564966202,
      "num_tokens": 13670332.0,
      "step": 6120
    },
    {
      "entropy": 5.783021879196167,
      "epoch": 0.6879317122479923,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004962485551878711,
      "loss": 5.676,
      "mean_token_accuracy": 0.16434723436832427,
      "num_tokens": 13682419.0,
      "step": 6125
    },
    {
      "entropy": 5.578999662399292,
      "epoch": 0.6884932891559499,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004962412507577283,
      "loss": 5.434,
      "mean_token_accuracy": 0.18486667722463607,
      "num_tokens": 13692698.0,
      "step": 6130
    },
    {
      "entropy": 5.5511950016021725,
      "epoch": 0.6890548660639074,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004962339392831783,
      "loss": 5.6314,
      "mean_token_accuracy": 0.17299454361200334,
      "num_tokens": 13704791.0,
      "step": 6135
    },
    {
      "entropy": 5.669870662689209,
      "epoch": 0.689616442971865,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000496226620764454,
      "loss": 5.5595,
      "mean_token_accuracy": 0.18140525370836258,
      "num_tokens": 13715860.0,
      "step": 6140
    },
    {
      "entropy": 5.706946134567261,
      "epoch": 0.6901780198798225,
      "grad_norm": 1.015625,
      "learning_rate": 0.000496219295201788,
      "loss": 5.5431,
      "mean_token_accuracy": 0.18075013756752015,
      "num_tokens": 13727029.0,
      "step": 6145
    },
    {
      "entropy": 5.627535486221314,
      "epoch": 0.6907395967877801,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004962119625954141,
      "loss": 5.5307,
      "mean_token_accuracy": 0.182185660302639,
      "num_tokens": 13737168.0,
      "step": 6150
    },
    {
      "entropy": 5.73567795753479,
      "epoch": 0.6913011736957376,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004962046229455653,
      "loss": 5.6696,
      "mean_token_accuracy": 0.16650990694761275,
      "num_tokens": 13749828.0,
      "step": 6155
    },
    {
      "entropy": 5.654178047180176,
      "epoch": 0.6918627506036952,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004961972762524757,
      "loss": 5.4991,
      "mean_token_accuracy": 0.17278295755386353,
      "num_tokens": 13761617.0,
      "step": 6160
    },
    {
      "entropy": 5.6681290626525875,
      "epoch": 0.6924243275116527,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496189922516379,
      "loss": 5.5524,
      "mean_token_accuracy": 0.18426286280155182,
      "num_tokens": 13772721.0,
      "step": 6165
    },
    {
      "entropy": 5.575658178329467,
      "epoch": 0.6929859044196103,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961825617375093,
      "loss": 5.5412,
      "mean_token_accuracy": 0.1766442507505417,
      "num_tokens": 13783215.0,
      "step": 6170
    },
    {
      "entropy": 5.564092636108398,
      "epoch": 0.6935474813275678,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004961751939161011,
      "loss": 5.4812,
      "mean_token_accuracy": 0.18373349010944368,
      "num_tokens": 13793496.0,
      "step": 6175
    },
    {
      "entropy": 5.509256601333618,
      "epoch": 0.6941090582355254,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496167819052389,
      "loss": 5.4176,
      "mean_token_accuracy": 0.17860936522483825,
      "num_tokens": 13802818.0,
      "step": 6180
    },
    {
      "entropy": 5.770023536682129,
      "epoch": 0.6946706351434829,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004961604371466077,
      "loss": 5.7072,
      "mean_token_accuracy": 0.1726786971092224,
      "num_tokens": 13814449.0,
      "step": 6185
    },
    {
      "entropy": 5.710700941085816,
      "epoch": 0.6952322120514405,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961530481989923,
      "loss": 5.6321,
      "mean_token_accuracy": 0.17531704306602477,
      "num_tokens": 13825369.0,
      "step": 6190
    },
    {
      "entropy": 5.586572647094727,
      "epoch": 0.695793788959398,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004961456522097782,
      "loss": 5.4985,
      "mean_token_accuracy": 0.17635693699121474,
      "num_tokens": 13837992.0,
      "step": 6195
    },
    {
      "entropy": 5.714752912521362,
      "epoch": 0.6963553658673556,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004961382491792006,
      "loss": 5.6089,
      "mean_token_accuracy": 0.17184996008872985,
      "num_tokens": 13849771.0,
      "step": 6200
    },
    {
      "entropy": 5.652769136428833,
      "epoch": 0.6969169427753131,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004961308391074953,
      "loss": 5.587,
      "mean_token_accuracy": 0.17513276189565657,
      "num_tokens": 13861997.0,
      "step": 6205
    },
    {
      "entropy": 5.711466979980469,
      "epoch": 0.6974785196832706,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004961234219948986,
      "loss": 5.6714,
      "mean_token_accuracy": 0.17424398809671401,
      "num_tokens": 13872972.0,
      "step": 6210
    },
    {
      "entropy": 5.5919266700744625,
      "epoch": 0.6980400965912281,
      "grad_norm": 1.0,
      "learning_rate": 0.0004961159978416461,
      "loss": 5.4473,
      "mean_token_accuracy": 0.18001363873481752,
      "num_tokens": 13883139.0,
      "step": 6215
    },
    {
      "entropy": 5.670935344696045,
      "epoch": 0.6986016734991857,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961085666479746,
      "loss": 5.6201,
      "mean_token_accuracy": 0.1752718135714531,
      "num_tokens": 13894326.0,
      "step": 6220
    },
    {
      "entropy": 5.690594577789307,
      "epoch": 0.6991632504071432,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004961011284141206,
      "loss": 5.6457,
      "mean_token_accuracy": 0.1730250060558319,
      "num_tokens": 13905525.0,
      "step": 6225
    },
    {
      "entropy": 5.682399272918701,
      "epoch": 0.6997248273151008,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004960936831403209,
      "loss": 5.5583,
      "mean_token_accuracy": 0.17501617968082428,
      "num_tokens": 13915973.0,
      "step": 6230
    },
    {
      "entropy": 5.584595727920532,
      "epoch": 0.7002864042230583,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960862308268126,
      "loss": 5.3906,
      "mean_token_accuracy": 0.1863795131444931,
      "num_tokens": 13926380.0,
      "step": 6235
    },
    {
      "entropy": 5.582453870773316,
      "epoch": 0.7008479811310159,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004960787714738329,
      "loss": 5.5793,
      "mean_token_accuracy": 0.17046487629413604,
      "num_tokens": 13938523.0,
      "step": 6240
    },
    {
      "entropy": 5.564077138900757,
      "epoch": 0.7014095580389734,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004960713050816194,
      "loss": 5.5593,
      "mean_token_accuracy": 0.1796951949596405,
      "num_tokens": 13949343.0,
      "step": 6245
    },
    {
      "entropy": 5.766108608245849,
      "epoch": 0.701971134946931,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960638316504098,
      "loss": 5.6802,
      "mean_token_accuracy": 0.17632725834846497,
      "num_tokens": 13960056.0,
      "step": 6250
    },
    {
      "entropy": 5.794866371154785,
      "epoch": 0.7025327118548885,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960563511804421,
      "loss": 5.6507,
      "mean_token_accuracy": 0.1752863645553589,
      "num_tokens": 13971932.0,
      "step": 6255
    },
    {
      "entropy": 5.695323705673218,
      "epoch": 0.7030942887628461,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004960488636719545,
      "loss": 5.648,
      "mean_token_accuracy": 0.17090354561805726,
      "num_tokens": 13981992.0,
      "step": 6260
    },
    {
      "entropy": 5.699549341201783,
      "epoch": 0.7036558656708036,
      "grad_norm": 1.0,
      "learning_rate": 0.0004960413691251852,
      "loss": 5.5927,
      "mean_token_accuracy": 0.18272750079631805,
      "num_tokens": 13992215.0,
      "step": 6265
    },
    {
      "entropy": 5.66727876663208,
      "epoch": 0.7042174425787612,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004960338675403731,
      "loss": 5.4959,
      "mean_token_accuracy": 0.18945090025663375,
      "num_tokens": 14002689.0,
      "step": 6270
    },
    {
      "entropy": 5.585068273544311,
      "epoch": 0.7047790194867187,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004960263589177568,
      "loss": 5.615,
      "mean_token_accuracy": 0.17799130380153655,
      "num_tokens": 14014951.0,
      "step": 6275
    },
    {
      "entropy": 5.652612018585205,
      "epoch": 0.7053405963946763,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004960188432575757,
      "loss": 5.5438,
      "mean_token_accuracy": 0.18065126091241837,
      "num_tokens": 14025502.0,
      "step": 6280
    },
    {
      "entropy": 5.7588221549987795,
      "epoch": 0.7059021733026338,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004960113205600688,
      "loss": 5.5959,
      "mean_token_accuracy": 0.18291736096143724,
      "num_tokens": 14036576.0,
      "step": 6285
    },
    {
      "entropy": 5.642173147201538,
      "epoch": 0.7064637502105914,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004960037908254757,
      "loss": 5.6226,
      "mean_token_accuracy": 0.17990546971559523,
      "num_tokens": 14046493.0,
      "step": 6290
    },
    {
      "entropy": 5.751773023605347,
      "epoch": 0.7070253271185489,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004959962540540361,
      "loss": 5.6775,
      "mean_token_accuracy": 0.18058259636163712,
      "num_tokens": 14057297.0,
      "step": 6295
    },
    {
      "entropy": 5.614803981781006,
      "epoch": 0.7075869040265065,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004959887102459903,
      "loss": 5.4618,
      "mean_token_accuracy": 0.18240975141525267,
      "num_tokens": 14068947.0,
      "step": 6300
    },
    {
      "entropy": 5.579032039642334,
      "epoch": 0.7081484809344639,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004959811594015781,
      "loss": 5.5573,
      "mean_token_accuracy": 0.1755152612924576,
      "num_tokens": 14078979.0,
      "step": 6305
    },
    {
      "entropy": 5.620485639572143,
      "epoch": 0.7087100578424215,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004959736015210401,
      "loss": 5.4855,
      "mean_token_accuracy": 0.18234243988990784,
      "num_tokens": 14089785.0,
      "step": 6310
    },
    {
      "entropy": 5.627900648117065,
      "epoch": 0.709271634750379,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959660366046169,
      "loss": 5.5878,
      "mean_token_accuracy": 0.17533970773220062,
      "num_tokens": 14099921.0,
      "step": 6315
    },
    {
      "entropy": 5.62828893661499,
      "epoch": 0.7098332116583366,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004959584646525494,
      "loss": 5.4869,
      "mean_token_accuracy": 0.17739137262105942,
      "num_tokens": 14110965.0,
      "step": 6320
    },
    {
      "entropy": 5.638380813598633,
      "epoch": 0.7103947885662941,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004959508856650788,
      "loss": 5.63,
      "mean_token_accuracy": 0.17418512403964997,
      "num_tokens": 14122409.0,
      "step": 6325
    },
    {
      "entropy": 5.67718539237976,
      "epoch": 0.7109563654742517,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004959432996424463,
      "loss": 5.5665,
      "mean_token_accuracy": 0.1799087792634964,
      "num_tokens": 14134017.0,
      "step": 6330
    },
    {
      "entropy": 5.709336662292481,
      "epoch": 0.7115179423822092,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004959357065848932,
      "loss": 5.6862,
      "mean_token_accuracy": 0.1742307111620903,
      "num_tokens": 14144843.0,
      "step": 6335
    },
    {
      "entropy": 5.6433673858642575,
      "epoch": 0.7120795192901668,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004959281064926618,
      "loss": 5.5399,
      "mean_token_accuracy": 0.1767047733068466,
      "num_tokens": 14155944.0,
      "step": 6340
    },
    {
      "entropy": 5.630907773971558,
      "epoch": 0.7126410961981243,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004959204993659936,
      "loss": 5.5973,
      "mean_token_accuracy": 0.17689424902200698,
      "num_tokens": 14167064.0,
      "step": 6345
    },
    {
      "entropy": 5.731440496444702,
      "epoch": 0.7132026731060819,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004959128852051311,
      "loss": 5.6192,
      "mean_token_accuracy": 0.16884241998195648,
      "num_tokens": 14178583.0,
      "step": 6350
    },
    {
      "entropy": 5.6261039733886715,
      "epoch": 0.7137642500140394,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004959052640103167,
      "loss": 5.5129,
      "mean_token_accuracy": 0.17680273950099945,
      "num_tokens": 14188752.0,
      "step": 6355
    },
    {
      "entropy": 5.608508205413818,
      "epoch": 0.714325826921997,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000495897635781793,
      "loss": 5.4797,
      "mean_token_accuracy": 0.18105533123016357,
      "num_tokens": 14199962.0,
      "step": 6360
    },
    {
      "entropy": 5.588458156585693,
      "epoch": 0.7148874038299545,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958900005198029,
      "loss": 5.5275,
      "mean_token_accuracy": 0.17510385662317277,
      "num_tokens": 14209877.0,
      "step": 6365
    },
    {
      "entropy": 5.5747175216674805,
      "epoch": 0.7154489807379121,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958823582245894,
      "loss": 5.5736,
      "mean_token_accuracy": 0.17937180995941163,
      "num_tokens": 14220672.0,
      "step": 6370
    },
    {
      "entropy": 5.715974426269531,
      "epoch": 0.7160105576458696,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004958747088963961,
      "loss": 5.5765,
      "mean_token_accuracy": 0.18228082805871965,
      "num_tokens": 14232348.0,
      "step": 6375
    },
    {
      "entropy": 5.65572338104248,
      "epoch": 0.7165721345538272,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004958670525354663,
      "loss": 5.5655,
      "mean_token_accuracy": 0.17139022350311278,
      "num_tokens": 14243985.0,
      "step": 6380
    },
    {
      "entropy": 5.591464376449585,
      "epoch": 0.7171337114617847,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000495859389142044,
      "loss": 5.4465,
      "mean_token_accuracy": 0.18223742246627808,
      "num_tokens": 14253548.0,
      "step": 6385
    },
    {
      "entropy": 5.677018022537231,
      "epoch": 0.7176952883697423,
      "grad_norm": 1.046875,
      "learning_rate": 0.000495851718716373,
      "loss": 5.5941,
      "mean_token_accuracy": 0.17394395023584366,
      "num_tokens": 14264757.0,
      "step": 6390
    },
    {
      "entropy": 5.55725588798523,
      "epoch": 0.7182568652776998,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004958440412586977,
      "loss": 5.5014,
      "mean_token_accuracy": 0.19020429253578186,
      "num_tokens": 14275946.0,
      "step": 6395
    },
    {
      "entropy": 5.633136177062989,
      "epoch": 0.7188184421856574,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004958363567692624,
      "loss": 5.5917,
      "mean_token_accuracy": 0.1715858042240143,
      "num_tokens": 14287006.0,
      "step": 6400
    },
    {
      "entropy": 5.769589424133301,
      "epoch": 0.7193800190936148,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004958286652483118,
      "loss": 5.6154,
      "mean_token_accuracy": 0.17177885621786118,
      "num_tokens": 14299328.0,
      "step": 6405
    },
    {
      "entropy": 5.683426475524902,
      "epoch": 0.7199415960015724,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004958209666960909,
      "loss": 5.5995,
      "mean_token_accuracy": 0.17586500495672225,
      "num_tokens": 14310048.0,
      "step": 6410
    },
    {
      "entropy": 5.578950309753418,
      "epoch": 0.7205031729095299,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004958132611128448,
      "loss": 5.5329,
      "mean_token_accuracy": 0.18161720037460327,
      "num_tokens": 14321107.0,
      "step": 6415
    },
    {
      "entropy": 5.588766717910767,
      "epoch": 0.7210647498174875,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004958055484988188,
      "loss": 5.4388,
      "mean_token_accuracy": 0.17925141602754593,
      "num_tokens": 14332561.0,
      "step": 6420
    },
    {
      "entropy": 5.719823837280273,
      "epoch": 0.721626326725445,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004957978288542585,
      "loss": 5.6527,
      "mean_token_accuracy": 0.17188818156719207,
      "num_tokens": 14343238.0,
      "step": 6425
    },
    {
      "entropy": 5.61526780128479,
      "epoch": 0.7221879036334026,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957901021794098,
      "loss": 5.5451,
      "mean_token_accuracy": 0.18082940131425856,
      "num_tokens": 14356067.0,
      "step": 6430
    },
    {
      "entropy": 5.63355016708374,
      "epoch": 0.7227494805413601,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004957823684745184,
      "loss": 5.5694,
      "mean_token_accuracy": 0.17994758635759353,
      "num_tokens": 14367302.0,
      "step": 6435
    },
    {
      "entropy": 5.613122844696045,
      "epoch": 0.7233110574493177,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004957746277398309,
      "loss": 5.5053,
      "mean_token_accuracy": 0.18515312373638154,
      "num_tokens": 14377535.0,
      "step": 6440
    },
    {
      "entropy": 5.712108516693116,
      "epoch": 0.7238726343572752,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004957668799755937,
      "loss": 5.6411,
      "mean_token_accuracy": 0.1715245634317398,
      "num_tokens": 14389493.0,
      "step": 6445
    },
    {
      "entropy": 5.64812970161438,
      "epoch": 0.7244342112652328,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004957591251820532,
      "loss": 5.5475,
      "mean_token_accuracy": 0.17638586610555648,
      "num_tokens": 14401564.0,
      "step": 6450
    },
    {
      "entropy": 5.692028951644898,
      "epoch": 0.7249957881731903,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004957513633594567,
      "loss": 5.6146,
      "mean_token_accuracy": 0.17962251603603363,
      "num_tokens": 14412123.0,
      "step": 6455
    },
    {
      "entropy": 5.654779052734375,
      "epoch": 0.7255573650811479,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957435945080512,
      "loss": 5.5839,
      "mean_token_accuracy": 0.17979903519153595,
      "num_tokens": 14423614.0,
      "step": 6460
    },
    {
      "entropy": 5.698367547988892,
      "epoch": 0.7261189419891054,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004957358186280839,
      "loss": 5.5821,
      "mean_token_accuracy": 0.1727018788456917,
      "num_tokens": 14434830.0,
      "step": 6465
    },
    {
      "entropy": 5.673839664459228,
      "epoch": 0.726680518897063,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004957280357198026,
      "loss": 5.6245,
      "mean_token_accuracy": 0.17713413536548614,
      "num_tokens": 14445015.0,
      "step": 6470
    },
    {
      "entropy": 5.682993030548095,
      "epoch": 0.7272420958050205,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004957202457834551,
      "loss": 5.697,
      "mean_token_accuracy": 0.17095409333705902,
      "num_tokens": 14455809.0,
      "step": 6475
    },
    {
      "entropy": 5.685393333435059,
      "epoch": 0.7278036727129781,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004957124488192893,
      "loss": 5.4851,
      "mean_token_accuracy": 0.184173022210598,
      "num_tokens": 14466565.0,
      "step": 6480
    },
    {
      "entropy": 5.679110431671143,
      "epoch": 0.7283652496209356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004957046448275535,
      "loss": 5.5926,
      "mean_token_accuracy": 0.18195963203907012,
      "num_tokens": 14477383.0,
      "step": 6485
    },
    {
      "entropy": 5.625385284423828,
      "epoch": 0.7289268265288932,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004956968338084961,
      "loss": 5.5654,
      "mean_token_accuracy": 0.1758306458592415,
      "num_tokens": 14489968.0,
      "step": 6490
    },
    {
      "entropy": 5.6930827617645265,
      "epoch": 0.7294884034368507,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004956890157623661,
      "loss": 5.6123,
      "mean_token_accuracy": 0.17990701347589494,
      "num_tokens": 14500808.0,
      "step": 6495
    },
    {
      "entropy": 5.624130773544311,
      "epoch": 0.7300499803448082,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956811906894122,
      "loss": 5.6243,
      "mean_token_accuracy": 0.17792190462350846,
      "num_tokens": 14511444.0,
      "step": 6500
    },
    {
      "entropy": 5.620875692367553,
      "epoch": 0.7306115572527657,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004956733585898836,
      "loss": 5.5315,
      "mean_token_accuracy": 0.1812374547123909,
      "num_tokens": 14522562.0,
      "step": 6505
    },
    {
      "entropy": 5.64646201133728,
      "epoch": 0.7311731341607233,
      "grad_norm": 1.0,
      "learning_rate": 0.0004956655194640297,
      "loss": 5.5665,
      "mean_token_accuracy": 0.17758426070213318,
      "num_tokens": 14533497.0,
      "step": 6510
    },
    {
      "entropy": 5.692796039581299,
      "epoch": 0.7317347110686808,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004956576733121,
      "loss": 5.598,
      "mean_token_accuracy": 0.16916249245405196,
      "num_tokens": 14545637.0,
      "step": 6515
    },
    {
      "entropy": 5.753426265716553,
      "epoch": 0.7322962879766384,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956498201343444,
      "loss": 5.5956,
      "mean_token_accuracy": 0.17224169969558717,
      "num_tokens": 14556826.0,
      "step": 6520
    },
    {
      "entropy": 5.550675916671753,
      "epoch": 0.7328578648845959,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000495641959931013,
      "loss": 5.5334,
      "mean_token_accuracy": 0.17258528470993043,
      "num_tokens": 14567315.0,
      "step": 6525
    },
    {
      "entropy": 5.674978303909302,
      "epoch": 0.7334194417925535,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004956340927023561,
      "loss": 5.5959,
      "mean_token_accuracy": 0.18003043234348298,
      "num_tokens": 14577685.0,
      "step": 6530
    },
    {
      "entropy": 5.717158794403076,
      "epoch": 0.733981018700511,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000495626218448624,
      "loss": 5.5573,
      "mean_token_accuracy": 0.18203723579645156,
      "num_tokens": 14588885.0,
      "step": 6535
    },
    {
      "entropy": 5.646932458877563,
      "epoch": 0.7345425956084686,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004956183371700676,
      "loss": 5.5742,
      "mean_token_accuracy": 0.18059505373239518,
      "num_tokens": 14599774.0,
      "step": 6540
    },
    {
      "entropy": 5.73839750289917,
      "epoch": 0.7351041725164261,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004956104488669377,
      "loss": 5.6226,
      "mean_token_accuracy": 0.1707258015871048,
      "num_tokens": 14612101.0,
      "step": 6545
    },
    {
      "entropy": 5.768576145172119,
      "epoch": 0.7356657494243837,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004956025535394856,
      "loss": 5.6829,
      "mean_token_accuracy": 0.16432776749134065,
      "num_tokens": 14622151.0,
      "step": 6550
    },
    {
      "entropy": 5.67973108291626,
      "epoch": 0.7362273263323412,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004955946511879626,
      "loss": 5.5817,
      "mean_token_accuracy": 0.17310305088758468,
      "num_tokens": 14634631.0,
      "step": 6555
    },
    {
      "entropy": 5.577903175354004,
      "epoch": 0.7367889032402988,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004955867418126205,
      "loss": 5.5306,
      "mean_token_accuracy": 0.1801718667149544,
      "num_tokens": 14645429.0,
      "step": 6560
    },
    {
      "entropy": 5.575721406936646,
      "epoch": 0.7373504801482563,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004955788254137108,
      "loss": 5.5515,
      "mean_token_accuracy": 0.18089888244867325,
      "num_tokens": 14657309.0,
      "step": 6565
    },
    {
      "entropy": 5.642233753204346,
      "epoch": 0.7379120570562139,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004955709019914857,
      "loss": 5.4547,
      "mean_token_accuracy": 0.17947259545326233,
      "num_tokens": 14668578.0,
      "step": 6570
    },
    {
      "entropy": 5.633334541320801,
      "epoch": 0.7384736339641714,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004955629715461977,
      "loss": 5.6982,
      "mean_token_accuracy": 0.17283702045679092,
      "num_tokens": 14679120.0,
      "step": 6575
    },
    {
      "entropy": 5.71691861152649,
      "epoch": 0.739035210872129,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004955550340780991,
      "loss": 5.637,
      "mean_token_accuracy": 0.17072526067495347,
      "num_tokens": 14690849.0,
      "step": 6580
    },
    {
      "entropy": 5.725338459014893,
      "epoch": 0.7395967877800865,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004955470895874427,
      "loss": 5.5853,
      "mean_token_accuracy": 0.17834530174732208,
      "num_tokens": 14702431.0,
      "step": 6585
    },
    {
      "entropy": 5.563197231292724,
      "epoch": 0.7401583646880441,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004955391380744813,
      "loss": 5.4968,
      "mean_token_accuracy": 0.18283657729625702,
      "num_tokens": 14714271.0,
      "step": 6590
    },
    {
      "entropy": 5.62459077835083,
      "epoch": 0.7407199415960015,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004955311795394684,
      "loss": 5.6289,
      "mean_token_accuracy": 0.17922379374504088,
      "num_tokens": 14724796.0,
      "step": 6595
    },
    {
      "entropy": 5.7125084400177,
      "epoch": 0.7412815185039591,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000495523213982657,
      "loss": 5.5526,
      "mean_token_accuracy": 0.1789660319685936,
      "num_tokens": 14735331.0,
      "step": 6600
    },
    {
      "entropy": 5.759558486938476,
      "epoch": 0.7418430954119166,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004955152414043011,
      "loss": 5.6177,
      "mean_token_accuracy": 0.17553897202014923,
      "num_tokens": 14748595.0,
      "step": 6605
    },
    {
      "entropy": 5.696309328079224,
      "epoch": 0.7424046723198742,
      "grad_norm": 1.0,
      "learning_rate": 0.0004955072618046543,
      "loss": 5.5436,
      "mean_token_accuracy": 0.1770661875605583,
      "num_tokens": 14759501.0,
      "step": 6610
    },
    {
      "entropy": 5.581427764892578,
      "epoch": 0.7429662492278317,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004954992751839708,
      "loss": 5.5808,
      "mean_token_accuracy": 0.17926053404808046,
      "num_tokens": 14769752.0,
      "step": 6615
    },
    {
      "entropy": 5.649709987640381,
      "epoch": 0.7435278261357893,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004954912815425048,
      "loss": 5.5369,
      "mean_token_accuracy": 0.1772114470601082,
      "num_tokens": 14781166.0,
      "step": 6620
    },
    {
      "entropy": 5.623387956619263,
      "epoch": 0.7440894030437468,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000495483280880511,
      "loss": 5.4501,
      "mean_token_accuracy": 0.18195586204528807,
      "num_tokens": 14790536.0,
      "step": 6625
    },
    {
      "entropy": 5.708026552200318,
      "epoch": 0.7446509799517044,
      "grad_norm": 1.0625,
      "learning_rate": 0.000495475273198244,
      "loss": 5.633,
      "mean_token_accuracy": 0.17637256681919097,
      "num_tokens": 14801673.0,
      "step": 6630
    },
    {
      "entropy": 5.685859251022339,
      "epoch": 0.7452125568596619,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004954672584959588,
      "loss": 5.534,
      "mean_token_accuracy": 0.17445341795682906,
      "num_tokens": 14812345.0,
      "step": 6635
    },
    {
      "entropy": 5.727948570251465,
      "epoch": 0.7457741337676195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004954592367739104,
      "loss": 5.71,
      "mean_token_accuracy": 0.17131739258766174,
      "num_tokens": 14823620.0,
      "step": 6640
    },
    {
      "entropy": 5.659303665161133,
      "epoch": 0.746335710675577,
      "grad_norm": 1.0,
      "learning_rate": 0.0004954512080323546,
      "loss": 5.5068,
      "mean_token_accuracy": 0.1815517619252205,
      "num_tokens": 14833487.0,
      "step": 6645
    },
    {
      "entropy": 5.528368377685547,
      "epoch": 0.7468972875835346,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004954431722715468,
      "loss": 5.3856,
      "mean_token_accuracy": 0.18170241862535477,
      "num_tokens": 14844421.0,
      "step": 6650
    },
    {
      "entropy": 5.58285117149353,
      "epoch": 0.7474588644914921,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004954351294917428,
      "loss": 5.5516,
      "mean_token_accuracy": 0.18573013246059417,
      "num_tokens": 14855525.0,
      "step": 6655
    },
    {
      "entropy": 5.746717357635498,
      "epoch": 0.7480204413994497,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004954270796931989,
      "loss": 5.6344,
      "mean_token_accuracy": 0.17335387468338012,
      "num_tokens": 14866717.0,
      "step": 6660
    },
    {
      "entropy": 5.617247200012207,
      "epoch": 0.7485820183074072,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004954190228761712,
      "loss": 5.524,
      "mean_token_accuracy": 0.18003176599740983,
      "num_tokens": 14877372.0,
      "step": 6665
    },
    {
      "entropy": 5.609625339508057,
      "epoch": 0.7491435952153648,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004954109590409164,
      "loss": 5.5619,
      "mean_token_accuracy": 0.18102132230997087,
      "num_tokens": 14888618.0,
      "step": 6670
    },
    {
      "entropy": 5.618066883087158,
      "epoch": 0.7497051721233223,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004954028881876911,
      "loss": 5.582,
      "mean_token_accuracy": 0.1763724699616432,
      "num_tokens": 14900421.0,
      "step": 6675
    },
    {
      "entropy": 5.688990259170533,
      "epoch": 0.7502667490312799,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004953948103167524,
      "loss": 5.5426,
      "mean_token_accuracy": 0.17627472281455994,
      "num_tokens": 14910634.0,
      "step": 6680
    },
    {
      "entropy": 5.718801259994507,
      "epoch": 0.7508283259392374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004953867254283575,
      "loss": 5.6102,
      "mean_token_accuracy": 0.16975564509630203,
      "num_tokens": 14921408.0,
      "step": 6685
    },
    {
      "entropy": 5.664679384231567,
      "epoch": 0.7513899028471949,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953786335227636,
      "loss": 5.5914,
      "mean_token_accuracy": 0.1800124704837799,
      "num_tokens": 14933435.0,
      "step": 6690
    },
    {
      "entropy": 5.587392330169678,
      "epoch": 0.7519514797551524,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004953705346002287,
      "loss": 5.6067,
      "mean_token_accuracy": 0.17205231189727782,
      "num_tokens": 14946488.0,
      "step": 6695
    },
    {
      "entropy": 5.611816883087158,
      "epoch": 0.75251305666311,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004953624286610106,
      "loss": 5.4678,
      "mean_token_accuracy": 0.1828654393553734,
      "num_tokens": 14958053.0,
      "step": 6700
    },
    {
      "entropy": 5.687585306167603,
      "epoch": 0.7530746335710675,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004953543157053672,
      "loss": 5.5456,
      "mean_token_accuracy": 0.17251309305429458,
      "num_tokens": 14968922.0,
      "step": 6705
    },
    {
      "entropy": 5.63105788230896,
      "epoch": 0.7536362104790251,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004953461957335569,
      "loss": 5.5565,
      "mean_token_accuracy": 0.18053768426179886,
      "num_tokens": 14979541.0,
      "step": 6710
    },
    {
      "entropy": 5.682077217102051,
      "epoch": 0.7541977873869826,
      "grad_norm": 1.0,
      "learning_rate": 0.0004953380687458382,
      "loss": 5.592,
      "mean_token_accuracy": 0.17432711124420167,
      "num_tokens": 14990009.0,
      "step": 6715
    },
    {
      "entropy": 5.6482017040252686,
      "epoch": 0.7547593642949402,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004953299347424701,
      "loss": 5.5922,
      "mean_token_accuracy": 0.1729342371225357,
      "num_tokens": 15001712.0,
      "step": 6720
    },
    {
      "entropy": 5.6821846008300785,
      "epoch": 0.7553209412028977,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004953217937237114,
      "loss": 5.5338,
      "mean_token_accuracy": 0.18202275931835174,
      "num_tokens": 15012574.0,
      "step": 6725
    },
    {
      "entropy": 5.576630592346191,
      "epoch": 0.7558825181108553,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004953136456898212,
      "loss": 5.536,
      "mean_token_accuracy": 0.18243104964494705,
      "num_tokens": 15023034.0,
      "step": 6730
    },
    {
      "entropy": 5.7357114315032955,
      "epoch": 0.7564440950188128,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004953054906410593,
      "loss": 5.583,
      "mean_token_accuracy": 0.17332218885421752,
      "num_tokens": 15033925.0,
      "step": 6735
    },
    {
      "entropy": 5.629609394073486,
      "epoch": 0.7570056719267704,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004952973285776848,
      "loss": 5.5648,
      "mean_token_accuracy": 0.18474339842796325,
      "num_tokens": 15044947.0,
      "step": 6740
    },
    {
      "entropy": 5.641076469421387,
      "epoch": 0.7575672488347279,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004952891594999582,
      "loss": 5.4805,
      "mean_token_accuracy": 0.18399559259414672,
      "num_tokens": 15054784.0,
      "step": 6745
    },
    {
      "entropy": 5.573967218399048,
      "epoch": 0.7581288257426855,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004952809834081392,
      "loss": 5.557,
      "mean_token_accuracy": 0.17349208146333694,
      "num_tokens": 15066272.0,
      "step": 6750
    },
    {
      "entropy": 5.703629016876221,
      "epoch": 0.758690402650643,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004952728003024883,
      "loss": 5.6326,
      "mean_token_accuracy": 0.17263609766960145,
      "num_tokens": 15079504.0,
      "step": 6755
    },
    {
      "entropy": 5.73172698020935,
      "epoch": 0.7592519795586006,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004952646101832659,
      "loss": 5.5724,
      "mean_token_accuracy": 0.18210249692201613,
      "num_tokens": 15089623.0,
      "step": 6760
    },
    {
      "entropy": 5.607945919036865,
      "epoch": 0.7598135564665581,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004952564130507331,
      "loss": 5.4771,
      "mean_token_accuracy": 0.17305199503898622,
      "num_tokens": 15100816.0,
      "step": 6765
    },
    {
      "entropy": 5.619532680511474,
      "epoch": 0.7603751333745157,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004952482089051506,
      "loss": 5.5681,
      "mean_token_accuracy": 0.18450591266155242,
      "num_tokens": 15111697.0,
      "step": 6770
    },
    {
      "entropy": 5.6809234619140625,
      "epoch": 0.7609367102824732,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004952399977467796,
      "loss": 5.6583,
      "mean_token_accuracy": 0.17277388274669647,
      "num_tokens": 15122869.0,
      "step": 6775
    },
    {
      "entropy": 5.585271644592285,
      "epoch": 0.7614982871904308,
      "grad_norm": 1.0,
      "learning_rate": 0.0004952317795758817,
      "loss": 5.4476,
      "mean_token_accuracy": 0.1824674978852272,
      "num_tokens": 15134709.0,
      "step": 6780
    },
    {
      "entropy": 5.618035459518433,
      "epoch": 0.7620598640983883,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004952235543927185,
      "loss": 5.4912,
      "mean_token_accuracy": 0.1805910736322403,
      "num_tokens": 15145319.0,
      "step": 6785
    },
    {
      "entropy": 5.5981072902679445,
      "epoch": 0.7626214410063458,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000495215322197552,
      "loss": 5.5212,
      "mean_token_accuracy": 0.1760521337389946,
      "num_tokens": 15156463.0,
      "step": 6790
    },
    {
      "entropy": 5.692944288253784,
      "epoch": 0.7631830179143033,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004952070829906442,
      "loss": 5.6445,
      "mean_token_accuracy": 0.16605111286044122,
      "num_tokens": 15167618.0,
      "step": 6795
    },
    {
      "entropy": 5.672136211395264,
      "epoch": 0.7637445948222609,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004951988367722574,
      "loss": 5.5881,
      "mean_token_accuracy": 0.17731750905513763,
      "num_tokens": 15179312.0,
      "step": 6800
    },
    {
      "entropy": 5.575615406036377,
      "epoch": 0.7643061717302184,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004951905835426544,
      "loss": 5.531,
      "mean_token_accuracy": 0.18109908252954482,
      "num_tokens": 15191370.0,
      "step": 6805
    },
    {
      "entropy": 5.690320682525635,
      "epoch": 0.764867748638176,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004951823233020976,
      "loss": 5.4933,
      "mean_token_accuracy": 0.1843721464276314,
      "num_tokens": 15201060.0,
      "step": 6810
    },
    {
      "entropy": 5.567106008529663,
      "epoch": 0.7654293255461335,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004951740560508504,
      "loss": 5.4173,
      "mean_token_accuracy": 0.18041964024305343,
      "num_tokens": 15212186.0,
      "step": 6815
    },
    {
      "entropy": 5.544131898880005,
      "epoch": 0.7659909024540911,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004951657817891758,
      "loss": 5.4382,
      "mean_token_accuracy": 0.18858288377523422,
      "num_tokens": 15223301.0,
      "step": 6820
    },
    {
      "entropy": 5.615245676040649,
      "epoch": 0.7665524793620486,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004951575005173373,
      "loss": 5.5977,
      "mean_token_accuracy": 0.17640408724546433,
      "num_tokens": 15234475.0,
      "step": 6825
    },
    {
      "entropy": 5.6895167350769045,
      "epoch": 0.7671140562700062,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004951492122355986,
      "loss": 5.5712,
      "mean_token_accuracy": 0.17795952558517455,
      "num_tokens": 15245877.0,
      "step": 6830
    },
    {
      "entropy": 5.657568693161011,
      "epoch": 0.7676756331779637,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004951409169442236,
      "loss": 5.5566,
      "mean_token_accuracy": 0.18239046037197112,
      "num_tokens": 15257375.0,
      "step": 6835
    },
    {
      "entropy": 5.628437662124634,
      "epoch": 0.7682372100859213,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951326146434765,
      "loss": 5.5809,
      "mean_token_accuracy": 0.17615204602479934,
      "num_tokens": 15268079.0,
      "step": 6840
    },
    {
      "entropy": 5.630699539184571,
      "epoch": 0.7687987869938788,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004951243053336214,
      "loss": 5.5634,
      "mean_token_accuracy": 0.17514654099941254,
      "num_tokens": 15278678.0,
      "step": 6845
    },
    {
      "entropy": 5.731458759307861,
      "epoch": 0.7693603639018364,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004951159890149232,
      "loss": 5.6736,
      "mean_token_accuracy": 0.17262617200613023,
      "num_tokens": 15290080.0,
      "step": 6850
    },
    {
      "entropy": 5.666080188751221,
      "epoch": 0.7699219408097939,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004951076656876466,
      "loss": 5.596,
      "mean_token_accuracy": 0.17108689546585082,
      "num_tokens": 15300602.0,
      "step": 6855
    },
    {
      "entropy": 5.679002046585083,
      "epoch": 0.7704835177177515,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004950993353520564,
      "loss": 5.5994,
      "mean_token_accuracy": 0.18032609522342682,
      "num_tokens": 15312823.0,
      "step": 6860
    },
    {
      "entropy": 5.705432748794555,
      "epoch": 0.771045094625709,
      "grad_norm": 0.9375,
      "learning_rate": 0.000495090998008418,
      "loss": 5.5971,
      "mean_token_accuracy": 0.17205487936735153,
      "num_tokens": 15323870.0,
      "step": 6865
    },
    {
      "entropy": 5.601526308059692,
      "epoch": 0.7716066715336666,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000495082653656997,
      "loss": 5.4999,
      "mean_token_accuracy": 0.18249612003564836,
      "num_tokens": 15334887.0,
      "step": 6870
    },
    {
      "entropy": 5.6566027164459225,
      "epoch": 0.7721682484416241,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004950743022980587,
      "loss": 5.5961,
      "mean_token_accuracy": 0.17155962884426118,
      "num_tokens": 15345871.0,
      "step": 6875
    },
    {
      "entropy": 5.596802568435669,
      "epoch": 0.7727298253495817,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004950659439318696,
      "loss": 5.4575,
      "mean_token_accuracy": 0.1836513876914978,
      "num_tokens": 15356530.0,
      "step": 6880
    },
    {
      "entropy": 5.604869699478149,
      "epoch": 0.7732914022575391,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004950575785586952,
      "loss": 5.5235,
      "mean_token_accuracy": 0.1817237541079521,
      "num_tokens": 15368756.0,
      "step": 6885
    },
    {
      "entropy": 5.684064817428589,
      "epoch": 0.7738529791654967,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004950492061788023,
      "loss": 5.5977,
      "mean_token_accuracy": 0.18085209876298905,
      "num_tokens": 15379450.0,
      "step": 6890
    },
    {
      "entropy": 5.624396705627442,
      "epoch": 0.7744145560734542,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004950408267924573,
      "loss": 5.5198,
      "mean_token_accuracy": 0.1743107855319977,
      "num_tokens": 15389501.0,
      "step": 6895
    },
    {
      "entropy": 5.5190822124481205,
      "epoch": 0.7749761329814118,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004950324403999271,
      "loss": 5.5206,
      "mean_token_accuracy": 0.18011448979377748,
      "num_tokens": 15400608.0,
      "step": 6900
    },
    {
      "entropy": 5.710084533691406,
      "epoch": 0.7755377098893693,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004950240470014785,
      "loss": 5.603,
      "mean_token_accuracy": 0.17645572125911713,
      "num_tokens": 15411709.0,
      "step": 6905
    },
    {
      "entropy": 5.582209825515747,
      "epoch": 0.7760992867973269,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004950156465973791,
      "loss": 5.4662,
      "mean_token_accuracy": 0.18789805173873902,
      "num_tokens": 15422292.0,
      "step": 6910
    },
    {
      "entropy": 5.6498126029968265,
      "epoch": 0.7766608637052844,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000495007239187896,
      "loss": 5.6158,
      "mean_token_accuracy": 0.17521766871213912,
      "num_tokens": 15434439.0,
      "step": 6915
    },
    {
      "entropy": 5.655129098892212,
      "epoch": 0.777222440613242,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004949988247732972,
      "loss": 5.5481,
      "mean_token_accuracy": 0.17417643517255782,
      "num_tokens": 15445480.0,
      "step": 6920
    },
    {
      "entropy": 5.625287294387817,
      "epoch": 0.7777840175211995,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004949904033538504,
      "loss": 5.5337,
      "mean_token_accuracy": 0.1788464143872261,
      "num_tokens": 15456960.0,
      "step": 6925
    },
    {
      "entropy": 5.583571100234986,
      "epoch": 0.7783455944291571,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000494981974929824,
      "loss": 5.544,
      "mean_token_accuracy": 0.17995237857103347,
      "num_tokens": 15468169.0,
      "step": 6930
    },
    {
      "entropy": 5.6113402366638185,
      "epoch": 0.7789071713371146,
      "grad_norm": 1.09375,
      "learning_rate": 0.000494973539501486,
      "loss": 5.5293,
      "mean_token_accuracy": 0.17978586107492447,
      "num_tokens": 15478740.0,
      "step": 6935
    },
    {
      "entropy": 5.64364652633667,
      "epoch": 0.7794687482450722,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004949650970691052,
      "loss": 5.6259,
      "mean_token_accuracy": 0.1706491157412529,
      "num_tokens": 15489705.0,
      "step": 6940
    },
    {
      "entropy": 5.578392839431762,
      "epoch": 0.7800303251530297,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004949566476329505,
      "loss": 5.4462,
      "mean_token_accuracy": 0.18012455254793167,
      "num_tokens": 15500549.0,
      "step": 6945
    },
    {
      "entropy": 5.718480587005615,
      "epoch": 0.7805919020609873,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004949481911932908,
      "loss": 5.6087,
      "mean_token_accuracy": 0.17435625344514846,
      "num_tokens": 15511124.0,
      "step": 6950
    },
    {
      "entropy": 5.609439659118652,
      "epoch": 0.7811534789689448,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004949397277503954,
      "loss": 5.5027,
      "mean_token_accuracy": 0.18258694410324097,
      "num_tokens": 15521810.0,
      "step": 6955
    },
    {
      "entropy": 5.66283917427063,
      "epoch": 0.7817150558769024,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004949312573045338,
      "loss": 5.5481,
      "mean_token_accuracy": 0.17872778922319413,
      "num_tokens": 15533108.0,
      "step": 6960
    },
    {
      "entropy": 5.627768659591675,
      "epoch": 0.7822766327848599,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004949227798559757,
      "loss": 5.5627,
      "mean_token_accuracy": 0.17870569974184036,
      "num_tokens": 15544841.0,
      "step": 6965
    },
    {
      "entropy": 5.61465892791748,
      "epoch": 0.7828382096928175,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004949142954049909,
      "loss": 5.5608,
      "mean_token_accuracy": 0.17548853754997254,
      "num_tokens": 15556297.0,
      "step": 6970
    },
    {
      "entropy": 5.61036696434021,
      "epoch": 0.783399786600775,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004949058039518496,
      "loss": 5.4589,
      "mean_token_accuracy": 0.18257362693548201,
      "num_tokens": 15566607.0,
      "step": 6975
    },
    {
      "entropy": 5.70405011177063,
      "epoch": 0.7839613635087325,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004948973054968222,
      "loss": 5.6147,
      "mean_token_accuracy": 0.17051844596862792,
      "num_tokens": 15577435.0,
      "step": 6980
    },
    {
      "entropy": 5.639044189453125,
      "epoch": 0.78452294041669,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004948888000401794,
      "loss": 5.5254,
      "mean_token_accuracy": 0.18027030378580094,
      "num_tokens": 15587844.0,
      "step": 6985
    },
    {
      "entropy": 5.669044351577758,
      "epoch": 0.7850845173246476,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004948802875821918,
      "loss": 5.5886,
      "mean_token_accuracy": 0.17704415023326875,
      "num_tokens": 15599586.0,
      "step": 6990
    },
    {
      "entropy": 5.573870754241943,
      "epoch": 0.7856460942326051,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004948717681231308,
      "loss": 5.4768,
      "mean_token_accuracy": 0.18209463059902192,
      "num_tokens": 15610838.0,
      "step": 6995
    },
    {
      "entropy": 5.6168153285980225,
      "epoch": 0.7862076711405627,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004948632416632672,
      "loss": 5.5594,
      "mean_token_accuracy": 0.17618024200201035,
      "num_tokens": 15622652.0,
      "step": 7000
    },
    {
      "entropy": 5.716824579238891,
      "epoch": 0.7867692480485202,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004948547082028728,
      "loss": 5.654,
      "mean_token_accuracy": 0.17388251274824143,
      "num_tokens": 15634280.0,
      "step": 7005
    },
    {
      "entropy": 5.572071170806884,
      "epoch": 0.7873308249564778,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004948461677422193,
      "loss": 5.3767,
      "mean_token_accuracy": 0.18992173969745635,
      "num_tokens": 15644088.0,
      "step": 7010
    },
    {
      "entropy": 5.62109866142273,
      "epoch": 0.7878924018644353,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004948376202815785,
      "loss": 5.6222,
      "mean_token_accuracy": 0.17247122526168823,
      "num_tokens": 15654968.0,
      "step": 7015
    },
    {
      "entropy": 5.619490337371826,
      "epoch": 0.7884539787723929,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004948290658212225,
      "loss": 5.5262,
      "mean_token_accuracy": 0.18326697051525115,
      "num_tokens": 15666758.0,
      "step": 7020
    },
    {
      "entropy": 5.530732345581055,
      "epoch": 0.7890155556803504,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948205043614239,
      "loss": 5.3799,
      "mean_token_accuracy": 0.19031909704208375,
      "num_tokens": 15677295.0,
      "step": 7025
    },
    {
      "entropy": 5.647847747802734,
      "epoch": 0.789577132588308,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004948119359024551,
      "loss": 5.5898,
      "mean_token_accuracy": 0.17960943132638932,
      "num_tokens": 15689508.0,
      "step": 7030
    },
    {
      "entropy": 5.71213321685791,
      "epoch": 0.7901387094962655,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004948033604445891,
      "loss": 5.5478,
      "mean_token_accuracy": 0.17901325672864915,
      "num_tokens": 15701289.0,
      "step": 7035
    },
    {
      "entropy": 5.562152433395386,
      "epoch": 0.7907002864042231,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004947947779880988,
      "loss": 5.4906,
      "mean_token_accuracy": 0.1727774813771248,
      "num_tokens": 15711579.0,
      "step": 7040
    },
    {
      "entropy": 5.611237621307373,
      "epoch": 0.7912618633121806,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004947861885332576,
      "loss": 5.5803,
      "mean_token_accuracy": 0.18268261551856996,
      "num_tokens": 15722629.0,
      "step": 7045
    },
    {
      "entropy": 5.645194244384766,
      "epoch": 0.7918234402201382,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004947775920803388,
      "loss": 5.5212,
      "mean_token_accuracy": 0.17456017136573793,
      "num_tokens": 15735818.0,
      "step": 7050
    },
    {
      "entropy": 5.5891368865966795,
      "epoch": 0.7923850171280957,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947689886296162,
      "loss": 5.5819,
      "mean_token_accuracy": 0.17798920273780822,
      "num_tokens": 15746914.0,
      "step": 7055
    },
    {
      "entropy": 5.586161708831787,
      "epoch": 0.7929465940360533,
      "grad_norm": 1.15625,
      "learning_rate": 0.000494760378181364,
      "loss": 5.4843,
      "mean_token_accuracy": 0.17352930903434755,
      "num_tokens": 15757597.0,
      "step": 7060
    },
    {
      "entropy": 5.605369901657104,
      "epoch": 0.7935081709440108,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004947517607358559,
      "loss": 5.4104,
      "mean_token_accuracy": 0.1861075982451439,
      "num_tokens": 15768860.0,
      "step": 7065
    },
    {
      "entropy": 5.6157104015350345,
      "epoch": 0.7940697478519684,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004947431362933666,
      "loss": 5.5444,
      "mean_token_accuracy": 0.17149641513824462,
      "num_tokens": 15779476.0,
      "step": 7070
    },
    {
      "entropy": 5.578100347518921,
      "epoch": 0.7946313247599258,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947345048541706,
      "loss": 5.5868,
      "mean_token_accuracy": 0.17395118325948716,
      "num_tokens": 15792142.0,
      "step": 7075
    },
    {
      "entropy": 5.61973295211792,
      "epoch": 0.7951929016678834,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947258664185427,
      "loss": 5.5594,
      "mean_token_accuracy": 0.17755297869443892,
      "num_tokens": 15802307.0,
      "step": 7080
    },
    {
      "entropy": 5.578988790512085,
      "epoch": 0.7957544785758409,
      "grad_norm": 1.0,
      "learning_rate": 0.0004947172209867581,
      "loss": 5.5129,
      "mean_token_accuracy": 0.17733030319213866,
      "num_tokens": 15812986.0,
      "step": 7085
    },
    {
      "entropy": 5.567875957489013,
      "epoch": 0.7963160554837985,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004947085685590919,
      "loss": 5.463,
      "mean_token_accuracy": 0.18132586628198624,
      "num_tokens": 15823938.0,
      "step": 7090
    },
    {
      "entropy": 5.565356779098511,
      "epoch": 0.796877632391756,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004946999091358196,
      "loss": 5.4428,
      "mean_token_accuracy": 0.18291499763727187,
      "num_tokens": 15835813.0,
      "step": 7095
    },
    {
      "entropy": 5.620179748535156,
      "epoch": 0.7974392092997136,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000494691242717217,
      "loss": 5.5135,
      "mean_token_accuracy": 0.1817249044775963,
      "num_tokens": 15846203.0,
      "step": 7100
    },
    {
      "entropy": 5.617189073562622,
      "epoch": 0.7980007862076711,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004946825693035601,
      "loss": 5.491,
      "mean_token_accuracy": 0.18329188376665115,
      "num_tokens": 15856919.0,
      "step": 7105
    },
    {
      "entropy": 5.640201139450073,
      "epoch": 0.7985623631156287,
      "grad_norm": 1.03125,
      "learning_rate": 0.000494673888895125,
      "loss": 5.5933,
      "mean_token_accuracy": 0.17678974568843842,
      "num_tokens": 15867149.0,
      "step": 7110
    },
    {
      "entropy": 5.534529066085815,
      "epoch": 0.7991239400235862,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004946652014921881,
      "loss": 5.4821,
      "mean_token_accuracy": 0.17967871576547623,
      "num_tokens": 15877229.0,
      "step": 7115
    },
    {
      "entropy": 5.612982368469238,
      "epoch": 0.7996855169315438,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004946565070950259,
      "loss": 5.4931,
      "mean_token_accuracy": 0.17960063070058824,
      "num_tokens": 15888799.0,
      "step": 7120
    },
    {
      "entropy": 5.567969942092896,
      "epoch": 0.8002470938395013,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004946478057039153,
      "loss": 5.4686,
      "mean_token_accuracy": 0.19139161407947541,
      "num_tokens": 15900144.0,
      "step": 7125
    },
    {
      "entropy": 5.665867805480957,
      "epoch": 0.8008086707474589,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004946390973191335,
      "loss": 5.6336,
      "mean_token_accuracy": 0.17360662966966628,
      "num_tokens": 15912030.0,
      "step": 7130
    },
    {
      "entropy": 5.627162742614746,
      "epoch": 0.8013702476554164,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004946303819409575,
      "loss": 5.5136,
      "mean_token_accuracy": 0.1804039403796196,
      "num_tokens": 15923606.0,
      "step": 7135
    },
    {
      "entropy": 5.525466537475586,
      "epoch": 0.801931824563374,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004946216595696652,
      "loss": 5.5001,
      "mean_token_accuracy": 0.1816282406449318,
      "num_tokens": 15935077.0,
      "step": 7140
    },
    {
      "entropy": 5.528133392333984,
      "epoch": 0.8024934014713315,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004946129302055338,
      "loss": 5.4395,
      "mean_token_accuracy": 0.18267178535461426,
      "num_tokens": 15945331.0,
      "step": 7145
    },
    {
      "entropy": 5.625937223434448,
      "epoch": 0.8030549783792891,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004946041938488417,
      "loss": 5.5317,
      "mean_token_accuracy": 0.177494777739048,
      "num_tokens": 15957031.0,
      "step": 7150
    },
    {
      "entropy": 5.671985340118408,
      "epoch": 0.8036165552872466,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004945954504998668,
      "loss": 5.5756,
      "mean_token_accuracy": 0.17943753749132157,
      "num_tokens": 15967545.0,
      "step": 7155
    },
    {
      "entropy": 5.710970115661621,
      "epoch": 0.8041781321952042,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004945867001588877,
      "loss": 5.614,
      "mean_token_accuracy": 0.1680646523833275,
      "num_tokens": 15978700.0,
      "step": 7160
    },
    {
      "entropy": 5.549652862548828,
      "epoch": 0.8047397091031617,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945779428261827,
      "loss": 5.5004,
      "mean_token_accuracy": 0.18439516574144363,
      "num_tokens": 15989712.0,
      "step": 7165
    },
    {
      "entropy": 5.742206001281739,
      "epoch": 0.8053012860111193,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494569178502031,
      "loss": 5.7094,
      "mean_token_accuracy": 0.1759636864066124,
      "num_tokens": 16002380.0,
      "step": 7170
    },
    {
      "entropy": 5.711532402038574,
      "epoch": 0.8058628629190767,
      "grad_norm": 1.0,
      "learning_rate": 0.0004945604071867114,
      "loss": 5.4894,
      "mean_token_accuracy": 0.1773687332868576,
      "num_tokens": 16013749.0,
      "step": 7175
    },
    {
      "entropy": 5.645091247558594,
      "epoch": 0.8064244398270343,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004945516288805035,
      "loss": 5.6528,
      "mean_token_accuracy": 0.1744043931365013,
      "num_tokens": 16025131.0,
      "step": 7180
    },
    {
      "entropy": 5.638998174667359,
      "epoch": 0.8069860167349918,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945428435836864,
      "loss": 5.5341,
      "mean_token_accuracy": 0.17833992689847947,
      "num_tokens": 16037269.0,
      "step": 7185
    },
    {
      "entropy": 5.7754237174987795,
      "epoch": 0.8075475936429494,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004945340512965401,
      "loss": 5.7265,
      "mean_token_accuracy": 0.17038890421390535,
      "num_tokens": 16050992.0,
      "step": 7190
    },
    {
      "entropy": 5.6251636981964115,
      "epoch": 0.8081091705509069,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945252520193443,
      "loss": 5.5598,
      "mean_token_accuracy": 0.1733875721693039,
      "num_tokens": 16062309.0,
      "step": 7195
    },
    {
      "entropy": 5.756367492675781,
      "epoch": 0.8086707474588645,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004945164457523795,
      "loss": 5.581,
      "mean_token_accuracy": 0.17815395891666413,
      "num_tokens": 16073079.0,
      "step": 7200
    },
    {
      "entropy": 5.636918449401856,
      "epoch": 0.809232324366822,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004945076324959258,
      "loss": 5.4812,
      "mean_token_accuracy": 0.18240586668252945,
      "num_tokens": 16082885.0,
      "step": 7205
    },
    {
      "entropy": 5.44354248046875,
      "epoch": 0.8097939012747796,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004944988122502641,
      "loss": 5.4359,
      "mean_token_accuracy": 0.1906597673892975,
      "num_tokens": 16094173.0,
      "step": 7210
    },
    {
      "entropy": 5.610472202301025,
      "epoch": 0.8103554781827371,
      "grad_norm": 1.015625,
      "learning_rate": 0.000494489985015675,
      "loss": 5.4802,
      "mean_token_accuracy": 0.17765818387269974,
      "num_tokens": 16106351.0,
      "step": 7215
    },
    {
      "entropy": 5.573151445388794,
      "epoch": 0.8109170550906947,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004944811507924396,
      "loss": 5.4836,
      "mean_token_accuracy": 0.17927531152963638,
      "num_tokens": 16116892.0,
      "step": 7220
    },
    {
      "entropy": 5.669474697113037,
      "epoch": 0.8114786319986522,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004944723095808393,
      "loss": 5.6198,
      "mean_token_accuracy": 0.1687384530901909,
      "num_tokens": 16128316.0,
      "step": 7225
    },
    {
      "entropy": 5.626483821868897,
      "epoch": 0.8120402089066098,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004944634613811555,
      "loss": 5.4535,
      "mean_token_accuracy": 0.17954584062099457,
      "num_tokens": 16139245.0,
      "step": 7230
    },
    {
      "entropy": 5.552922773361206,
      "epoch": 0.8126017858145673,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00049445460619367,
      "loss": 5.4844,
      "mean_token_accuracy": 0.1813029646873474,
      "num_tokens": 16150893.0,
      "step": 7235
    },
    {
      "entropy": 5.611425065994263,
      "epoch": 0.8131633627225249,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004944457440186648,
      "loss": 5.6503,
      "mean_token_accuracy": 0.16410815566778184,
      "num_tokens": 16163430.0,
      "step": 7240
    },
    {
      "entropy": 5.757813405990601,
      "epoch": 0.8137249396304824,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004944368748564219,
      "loss": 5.53,
      "mean_token_accuracy": 0.18086985647678375,
      "num_tokens": 16173954.0,
      "step": 7245
    },
    {
      "entropy": 5.586259508132935,
      "epoch": 0.81428651653844,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004944279987072238,
      "loss": 5.5687,
      "mean_token_accuracy": 0.17602573484182357,
      "num_tokens": 16185491.0,
      "step": 7250
    },
    {
      "entropy": 5.497488117218017,
      "epoch": 0.8148480934463975,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004944191155713531,
      "loss": 5.4231,
      "mean_token_accuracy": 0.19166387915611266,
      "num_tokens": 16195301.0,
      "step": 7255
    },
    {
      "entropy": 5.659170913696289,
      "epoch": 0.8154096703543551,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004944102254490928,
      "loss": 5.5571,
      "mean_token_accuracy": 0.17331268936395644,
      "num_tokens": 16206620.0,
      "step": 7260
    },
    {
      "entropy": 5.648644590377808,
      "epoch": 0.8159712472623126,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004944013283407257,
      "loss": 5.6375,
      "mean_token_accuracy": 0.17144708782434465,
      "num_tokens": 16218693.0,
      "step": 7265
    },
    {
      "entropy": 5.7061583518981935,
      "epoch": 0.8165328241702701,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004943924242465352,
      "loss": 5.5237,
      "mean_token_accuracy": 0.18040265440940856,
      "num_tokens": 16229124.0,
      "step": 7270
    },
    {
      "entropy": 5.559056901931763,
      "epoch": 0.8170944010782276,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004943835131668049,
      "loss": 5.5089,
      "mean_token_accuracy": 0.17560482025146484,
      "num_tokens": 16239999.0,
      "step": 7275
    },
    {
      "entropy": 5.628651857376099,
      "epoch": 0.8176559779861852,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004943745951018184,
      "loss": 5.571,
      "mean_token_accuracy": 0.18164185136556626,
      "num_tokens": 16251027.0,
      "step": 7280
    },
    {
      "entropy": 5.667185211181641,
      "epoch": 0.8182175548941427,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004943656700518597,
      "loss": 5.5314,
      "mean_token_accuracy": 0.1802248865365982,
      "num_tokens": 16262924.0,
      "step": 7285
    },
    {
      "entropy": 5.697255945205688,
      "epoch": 0.8187791318021003,
      "grad_norm": 1.125,
      "learning_rate": 0.0004943567380172129,
      "loss": 5.6383,
      "mean_token_accuracy": 0.17880614697933198,
      "num_tokens": 16274481.0,
      "step": 7290
    },
    {
      "entropy": 5.637316036224365,
      "epoch": 0.8193407087100578,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004943477989981627,
      "loss": 5.5605,
      "mean_token_accuracy": 0.1813304215669632,
      "num_tokens": 16285214.0,
      "step": 7295
    },
    {
      "entropy": 5.672893047332764,
      "epoch": 0.8199022856180154,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004943388529949933,
      "loss": 5.6307,
      "mean_token_accuracy": 0.1746322527527809,
      "num_tokens": 16296077.0,
      "step": 7300
    },
    {
      "entropy": 5.700228357315064,
      "epoch": 0.8204638625259729,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004943299000079899,
      "loss": 5.5368,
      "mean_token_accuracy": 0.17813003361225127,
      "num_tokens": 16307514.0,
      "step": 7305
    },
    {
      "entropy": 5.680727100372314,
      "epoch": 0.8210254394339305,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004943209400374373,
      "loss": 5.5332,
      "mean_token_accuracy": 0.17984469681978227,
      "num_tokens": 16318734.0,
      "step": 7310
    },
    {
      "entropy": 5.579897356033325,
      "epoch": 0.821587016341888,
      "grad_norm": 1.046875,
      "learning_rate": 0.000494311973083621,
      "loss": 5.5628,
      "mean_token_accuracy": 0.17839521467685698,
      "num_tokens": 16329057.0,
      "step": 7315
    },
    {
      "entropy": 5.579474544525146,
      "epoch": 0.8221485932498456,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004943029991468264,
      "loss": 5.5024,
      "mean_token_accuracy": 0.18085712641477586,
      "num_tokens": 16339646.0,
      "step": 7320
    },
    {
      "entropy": 5.652965307235718,
      "epoch": 0.8227101701578031,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942940182273393,
      "loss": 5.5888,
      "mean_token_accuracy": 0.17340163737535477,
      "num_tokens": 16349783.0,
      "step": 7325
    },
    {
      "entropy": 5.668771743774414,
      "epoch": 0.8232717470657607,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004942850303254455,
      "loss": 5.5021,
      "mean_token_accuracy": 0.18269888162612916,
      "num_tokens": 16361078.0,
      "step": 7330
    },
    {
      "entropy": 5.528364753723144,
      "epoch": 0.8238333239737182,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004942760354414313,
      "loss": 5.4976,
      "mean_token_accuracy": 0.18162541687488556,
      "num_tokens": 16371308.0,
      "step": 7335
    },
    {
      "entropy": 5.628272008895874,
      "epoch": 0.8243949008816758,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004942670335755831,
      "loss": 5.5518,
      "mean_token_accuracy": 0.18339990228414535,
      "num_tokens": 16382682.0,
      "step": 7340
    },
    {
      "entropy": 5.7226831912994385,
      "epoch": 0.8249564777896333,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004942580247281876,
      "loss": 5.5871,
      "mean_token_accuracy": 0.17570515275001525,
      "num_tokens": 16393715.0,
      "step": 7345
    },
    {
      "entropy": 5.574453592300415,
      "epoch": 0.8255180546975909,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004942490088995315,
      "loss": 5.5441,
      "mean_token_accuracy": 0.1787221685051918,
      "num_tokens": 16406332.0,
      "step": 7350
    },
    {
      "entropy": 5.5816456317901615,
      "epoch": 0.8260796316055484,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004942399860899018,
      "loss": 5.4876,
      "mean_token_accuracy": 0.18046286553144456,
      "num_tokens": 16417702.0,
      "step": 7355
    },
    {
      "entropy": 5.6356000900268555,
      "epoch": 0.826641208513506,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004942309562995862,
      "loss": 5.5239,
      "mean_token_accuracy": 0.17496747821569442,
      "num_tokens": 16428497.0,
      "step": 7360
    },
    {
      "entropy": 5.63239164352417,
      "epoch": 0.8272027854214634,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004942219195288716,
      "loss": 5.5331,
      "mean_token_accuracy": 0.17938542664051055,
      "num_tokens": 16440637.0,
      "step": 7365
    },
    {
      "entropy": 5.597147846221924,
      "epoch": 0.827764362329421,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004942128757780463,
      "loss": 5.547,
      "mean_token_accuracy": 0.1762922652065754,
      "num_tokens": 16453213.0,
      "step": 7370
    },
    {
      "entropy": 5.601246643066406,
      "epoch": 0.8283259392373785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004942038250473979,
      "loss": 5.5616,
      "mean_token_accuracy": 0.1792689397931099,
      "num_tokens": 16464345.0,
      "step": 7375
    },
    {
      "entropy": 5.570774888992309,
      "epoch": 0.8288875161453361,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004941947673372147,
      "loss": 5.4458,
      "mean_token_accuracy": 0.1834470361471176,
      "num_tokens": 16475399.0,
      "step": 7380
    },
    {
      "entropy": 5.552033090591431,
      "epoch": 0.8294490930532936,
      "grad_norm": 1.0,
      "learning_rate": 0.0004941857026477851,
      "loss": 5.4295,
      "mean_token_accuracy": 0.18250147253274918,
      "num_tokens": 16485439.0,
      "step": 7385
    },
    {
      "entropy": 5.557843732833862,
      "epoch": 0.8300106699612512,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004941766309793978,
      "loss": 5.4211,
      "mean_token_accuracy": 0.18478182405233384,
      "num_tokens": 16496765.0,
      "step": 7390
    },
    {
      "entropy": 5.50337781906128,
      "epoch": 0.8305722468692087,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004941675523323414,
      "loss": 5.4742,
      "mean_token_accuracy": 0.18376317471265793,
      "num_tokens": 16509199.0,
      "step": 7395
    },
    {
      "entropy": 5.668723487854004,
      "epoch": 0.8311338237771663,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004941584667069052,
      "loss": 5.6168,
      "mean_token_accuracy": 0.17944542020559312,
      "num_tokens": 16520073.0,
      "step": 7400
    },
    {
      "entropy": 5.6101233005523685,
      "epoch": 0.8316954006851238,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004941493741033784,
      "loss": 5.4928,
      "mean_token_accuracy": 0.17857892513275148,
      "num_tokens": 16530919.0,
      "step": 7405
    },
    {
      "entropy": 5.585693073272705,
      "epoch": 0.8322569775930814,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004941402745220507,
      "loss": 5.5152,
      "mean_token_accuracy": 0.18006205558776855,
      "num_tokens": 16542591.0,
      "step": 7410
    },
    {
      "entropy": 5.639138460159302,
      "epoch": 0.8328185545010389,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004941311679632115,
      "loss": 5.5166,
      "mean_token_accuracy": 0.17835039049386978,
      "num_tokens": 16554984.0,
      "step": 7415
    },
    {
      "entropy": 5.569074535369873,
      "epoch": 0.8333801314089965,
      "grad_norm": 1.0,
      "learning_rate": 0.000494122054427151,
      "loss": 5.4353,
      "mean_token_accuracy": 0.18038525581359863,
      "num_tokens": 16566598.0,
      "step": 7420
    },
    {
      "entropy": 5.645244073867798,
      "epoch": 0.833941708316954,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004941129339141592,
      "loss": 5.5727,
      "mean_token_accuracy": 0.17438869774341584,
      "num_tokens": 16577868.0,
      "step": 7425
    },
    {
      "entropy": 5.605511569976807,
      "epoch": 0.8345032852249116,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004941038064245267,
      "loss": 5.5199,
      "mean_token_accuracy": 0.17894256263971328,
      "num_tokens": 16587891.0,
      "step": 7430
    },
    {
      "entropy": 5.581606817245484,
      "epoch": 0.8350648621328691,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000494094671958544,
      "loss": 5.4789,
      "mean_token_accuracy": 0.185255566239357,
      "num_tokens": 16599924.0,
      "step": 7435
    },
    {
      "entropy": 5.564290761947632,
      "epoch": 0.8356264390408267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004940855305165019,
      "loss": 5.4221,
      "mean_token_accuracy": 0.1857829600572586,
      "num_tokens": 16609958.0,
      "step": 7440
    },
    {
      "entropy": 5.592352771759034,
      "epoch": 0.8361880159487842,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004940763820986916,
      "loss": 5.5451,
      "mean_token_accuracy": 0.17597225308418274,
      "num_tokens": 16621159.0,
      "step": 7445
    },
    {
      "entropy": 5.665910339355468,
      "epoch": 0.8367495928567418,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004940672267054044,
      "loss": 5.6495,
      "mean_token_accuracy": 0.17905145585536958,
      "num_tokens": 16632263.0,
      "step": 7450
    },
    {
      "entropy": 5.610781002044678,
      "epoch": 0.8373111697646993,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004940580643369316,
      "loss": 5.5127,
      "mean_token_accuracy": 0.1742522358894348,
      "num_tokens": 16644117.0,
      "step": 7455
    },
    {
      "entropy": 5.504357767105103,
      "epoch": 0.8378727466726568,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004940488949935651,
      "loss": 5.4364,
      "mean_token_accuracy": 0.1833467125892639,
      "num_tokens": 16654529.0,
      "step": 7460
    },
    {
      "entropy": 5.552529144287109,
      "epoch": 0.8384343235806143,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004940397186755968,
      "loss": 5.4581,
      "mean_token_accuracy": 0.18409384936094284,
      "num_tokens": 16666191.0,
      "step": 7465
    },
    {
      "entropy": 5.602557229995727,
      "epoch": 0.8389959004885719,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004940305353833191,
      "loss": 5.4532,
      "mean_token_accuracy": 0.18632374703884125,
      "num_tokens": 16677252.0,
      "step": 7470
    },
    {
      "entropy": 5.514086151123047,
      "epoch": 0.8395574773965294,
      "grad_norm": 1.0625,
      "learning_rate": 0.000494021345117024,
      "loss": 5.5045,
      "mean_token_accuracy": 0.1806218221783638,
      "num_tokens": 16689487.0,
      "step": 7475
    },
    {
      "entropy": 5.5145933628082275,
      "epoch": 0.840119054304487,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004940121478770044,
      "loss": 5.3797,
      "mean_token_accuracy": 0.18738940060138704,
      "num_tokens": 16699955.0,
      "step": 7480
    },
    {
      "entropy": 5.490093421936035,
      "epoch": 0.8406806312124445,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004940029436635531,
      "loss": 5.4548,
      "mean_token_accuracy": 0.18099683672189712,
      "num_tokens": 16711071.0,
      "step": 7485
    },
    {
      "entropy": 5.5467767238616945,
      "epoch": 0.8412422081204021,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939937324769631,
      "loss": 5.4595,
      "mean_token_accuracy": 0.18532306998968123,
      "num_tokens": 16722115.0,
      "step": 7490
    },
    {
      "entropy": 5.573823499679565,
      "epoch": 0.8418037850283596,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004939845143175276,
      "loss": 5.4847,
      "mean_token_accuracy": 0.1772324413061142,
      "num_tokens": 16733146.0,
      "step": 7495
    },
    {
      "entropy": 5.528927516937256,
      "epoch": 0.8423653619363172,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004939752891855404,
      "loss": 5.4651,
      "mean_token_accuracy": 0.17675635814666749,
      "num_tokens": 16745543.0,
      "step": 7500
    },
    {
      "entropy": 5.652834272384643,
      "epoch": 0.8429269388442747,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004939660570812951,
      "loss": 5.5381,
      "mean_token_accuracy": 0.18086207807064056,
      "num_tokens": 16755782.0,
      "step": 7505
    },
    {
      "entropy": 5.671106624603271,
      "epoch": 0.8434885157522323,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939568180050855,
      "loss": 5.5523,
      "mean_token_accuracy": 0.17602942287921905,
      "num_tokens": 16767509.0,
      "step": 7510
    },
    {
      "entropy": 5.536266279220581,
      "epoch": 0.8440500926601898,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004939475719572059,
      "loss": 5.4468,
      "mean_token_accuracy": 0.18344242572784425,
      "num_tokens": 16779257.0,
      "step": 7515
    },
    {
      "entropy": 5.494624519348145,
      "epoch": 0.8446116695681474,
      "grad_norm": 1.0,
      "learning_rate": 0.0004939383189379507,
      "loss": 5.4771,
      "mean_token_accuracy": 0.1824428215622902,
      "num_tokens": 16790769.0,
      "step": 7520
    },
    {
      "entropy": 5.61299409866333,
      "epoch": 0.8451732464761049,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004939290589476145,
      "loss": 5.4786,
      "mean_token_accuracy": 0.1779421776533127,
      "num_tokens": 16800054.0,
      "step": 7525
    },
    {
      "entropy": 5.661533546447754,
      "epoch": 0.8457348233840625,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939197919864922,
      "loss": 5.5634,
      "mean_token_accuracy": 0.17657241970300674,
      "num_tokens": 16811485.0,
      "step": 7530
    },
    {
      "entropy": 5.6560791492462155,
      "epoch": 0.84629640029202,
      "grad_norm": 1.125,
      "learning_rate": 0.0004939105180548788,
      "loss": 5.6189,
      "mean_token_accuracy": 0.18123217225074767,
      "num_tokens": 16822683.0,
      "step": 7535
    },
    {
      "entropy": 5.659206247329712,
      "epoch": 0.8468579771999776,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004939012371530696,
      "loss": 5.4949,
      "mean_token_accuracy": 0.18079070150852203,
      "num_tokens": 16834200.0,
      "step": 7540
    },
    {
      "entropy": 5.665460824966431,
      "epoch": 0.8474195541079351,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004938919492813601,
      "loss": 5.552,
      "mean_token_accuracy": 0.1727047935128212,
      "num_tokens": 16846399.0,
      "step": 7545
    },
    {
      "entropy": 5.670234203338623,
      "epoch": 0.8479811310158927,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004938826544400461,
      "loss": 5.6857,
      "mean_token_accuracy": 0.17400158941745758,
      "num_tokens": 16858322.0,
      "step": 7550
    },
    {
      "entropy": 5.603891086578369,
      "epoch": 0.8485427079238502,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004938733526294234,
      "loss": 5.5161,
      "mean_token_accuracy": 0.18002558946609498,
      "num_tokens": 16869537.0,
      "step": 7555
    },
    {
      "entropy": 5.7228978157043455,
      "epoch": 0.8491042848318077,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004938640438497883,
      "loss": 5.5528,
      "mean_token_accuracy": 0.1807059243321419,
      "num_tokens": 16881566.0,
      "step": 7560
    },
    {
      "entropy": 5.505270767211914,
      "epoch": 0.8496658617397652,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004938547281014372,
      "loss": 5.4836,
      "mean_token_accuracy": 0.18655762821435928,
      "num_tokens": 16892959.0,
      "step": 7565
    },
    {
      "entropy": 5.627756643295288,
      "epoch": 0.8502274386477228,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004938454053846666,
      "loss": 5.5327,
      "mean_token_accuracy": 0.17650023996829986,
      "num_tokens": 16904809.0,
      "step": 7570
    },
    {
      "entropy": 5.6635175228118895,
      "epoch": 0.8507890155556803,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004938360756997735,
      "loss": 5.5555,
      "mean_token_accuracy": 0.1722218871116638,
      "num_tokens": 16916969.0,
      "step": 7575
    },
    {
      "entropy": 5.547809886932373,
      "epoch": 0.8513505924636379,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004938267390470547,
      "loss": 5.4493,
      "mean_token_accuracy": 0.18617565035820008,
      "num_tokens": 16928253.0,
      "step": 7580
    },
    {
      "entropy": 5.5388415336608885,
      "epoch": 0.8519121693715954,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004938173954268079,
      "loss": 5.4925,
      "mean_token_accuracy": 0.1745935708284378,
      "num_tokens": 16939875.0,
      "step": 7585
    },
    {
      "entropy": 5.716016340255737,
      "epoch": 0.852473746279553,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004938080448393302,
      "loss": 5.6158,
      "mean_token_accuracy": 0.17623613476753236,
      "num_tokens": 16950173.0,
      "step": 7590
    },
    {
      "entropy": 5.595033264160156,
      "epoch": 0.8530353231875105,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004937986872849195,
      "loss": 5.5257,
      "mean_token_accuracy": 0.1806575506925583,
      "num_tokens": 16960933.0,
      "step": 7595
    },
    {
      "entropy": 5.53103814125061,
      "epoch": 0.8535969000954681,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004937893227638738,
      "loss": 5.4185,
      "mean_token_accuracy": 0.18351842612028121,
      "num_tokens": 16972284.0,
      "step": 7600
    },
    {
      "entropy": 5.60010724067688,
      "epoch": 0.8541584770034256,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004937799512764912,
      "loss": 5.4401,
      "mean_token_accuracy": 0.18844986706972122,
      "num_tokens": 16983541.0,
      "step": 7605
    },
    {
      "entropy": 5.6820151805877686,
      "epoch": 0.8547200539113832,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00049377057282307,
      "loss": 5.5941,
      "mean_token_accuracy": 0.17229961454868317,
      "num_tokens": 16993773.0,
      "step": 7610
    },
    {
      "entropy": 5.676660823822021,
      "epoch": 0.8552816308193407,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493761187403909,
      "loss": 5.5623,
      "mean_token_accuracy": 0.17587521821260452,
      "num_tokens": 17005287.0,
      "step": 7615
    },
    {
      "entropy": 5.610708808898925,
      "epoch": 0.8558432077272983,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000493751795019307,
      "loss": 5.4996,
      "mean_token_accuracy": 0.18012612909078599,
      "num_tokens": 17015664.0,
      "step": 7620
    },
    {
      "entropy": 5.535806179046631,
      "epoch": 0.8564047846352558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004937423956695629,
      "loss": 5.4242,
      "mean_token_accuracy": 0.18311096727848053,
      "num_tokens": 17026876.0,
      "step": 7625
    },
    {
      "entropy": 5.506599283218383,
      "epoch": 0.8569663615432134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004937329893549762,
      "loss": 5.4716,
      "mean_token_accuracy": 0.19143964052200318,
      "num_tokens": 17037321.0,
      "step": 7630
    },
    {
      "entropy": 5.6227470397949215,
      "epoch": 0.8575279384511709,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004937235760758462,
      "loss": 5.5503,
      "mean_token_accuracy": 0.1803719401359558,
      "num_tokens": 17047506.0,
      "step": 7635
    },
    {
      "entropy": 5.7091453075408936,
      "epoch": 0.8580895153591285,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004937141558324726,
      "loss": 5.5766,
      "mean_token_accuracy": 0.1779593601822853,
      "num_tokens": 17059710.0,
      "step": 7640
    },
    {
      "entropy": 5.681270933151245,
      "epoch": 0.858651092267086,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004937047286251556,
      "loss": 5.6052,
      "mean_token_accuracy": 0.1721954107284546,
      "num_tokens": 17070956.0,
      "step": 7645
    },
    {
      "entropy": 5.67075514793396,
      "epoch": 0.8592126691750436,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004936952944541952,
      "loss": 5.569,
      "mean_token_accuracy": 0.17900737822055818,
      "num_tokens": 17081404.0,
      "step": 7650
    },
    {
      "entropy": 5.679018259048462,
      "epoch": 0.859774246083001,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004936858533198917,
      "loss": 5.5798,
      "mean_token_accuracy": 0.17749113738536834,
      "num_tokens": 17092692.0,
      "step": 7655
    },
    {
      "entropy": 5.587037515640259,
      "epoch": 0.8603358229909586,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000493676405222546,
      "loss": 5.4279,
      "mean_token_accuracy": 0.18479957133531572,
      "num_tokens": 17103730.0,
      "step": 7660
    },
    {
      "entropy": 5.5954996109008786,
      "epoch": 0.8608973998989161,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004936669501624586,
      "loss": 5.4967,
      "mean_token_accuracy": 0.17741685062646867,
      "num_tokens": 17115086.0,
      "step": 7665
    },
    {
      "entropy": 5.562705373764038,
      "epoch": 0.8614589768068737,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004936574881399308,
      "loss": 5.5177,
      "mean_token_accuracy": 0.17818861007690429,
      "num_tokens": 17125928.0,
      "step": 7670
    },
    {
      "entropy": 5.633713483810425,
      "epoch": 0.8620205537148312,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004936480191552637,
      "loss": 5.5344,
      "mean_token_accuracy": 0.17698321789503096,
      "num_tokens": 17137328.0,
      "step": 7675
    },
    {
      "entropy": 5.56837158203125,
      "epoch": 0.8625821306227888,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004936385432087589,
      "loss": 5.5116,
      "mean_token_accuracy": 0.1792793244123459,
      "num_tokens": 17148274.0,
      "step": 7680
    },
    {
      "entropy": 5.567792510986328,
      "epoch": 0.8631437075307463,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004936290603007181,
      "loss": 5.5524,
      "mean_token_accuracy": 0.1794579342007637,
      "num_tokens": 17158834.0,
      "step": 7685
    },
    {
      "entropy": 5.5691451072692875,
      "epoch": 0.8637052844387039,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004936195704314432,
      "loss": 5.3966,
      "mean_token_accuracy": 0.18155379891395568,
      "num_tokens": 17169276.0,
      "step": 7690
    },
    {
      "entropy": 5.575228214263916,
      "epoch": 0.8642668613466614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004936100736012363,
      "loss": 5.4476,
      "mean_token_accuracy": 0.18865410089492798,
      "num_tokens": 17180642.0,
      "step": 7695
    },
    {
      "entropy": 5.5229919910430905,
      "epoch": 0.864828438254619,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004936005698104,
      "loss": 5.5157,
      "mean_token_accuracy": 0.1827980622649193,
      "num_tokens": 17192477.0,
      "step": 7700
    },
    {
      "entropy": 5.683761119842529,
      "epoch": 0.8653900151625765,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004935910590592367,
      "loss": 5.648,
      "mean_token_accuracy": 0.17949193716049194,
      "num_tokens": 17202653.0,
      "step": 7705
    },
    {
      "entropy": 5.522699785232544,
      "epoch": 0.8659515920705341,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004935815413480494,
      "loss": 5.3877,
      "mean_token_accuracy": 0.18448692113161086,
      "num_tokens": 17212607.0,
      "step": 7710
    },
    {
      "entropy": 5.577323579788208,
      "epoch": 0.8665131689784916,
      "grad_norm": 0.984375,
      "learning_rate": 0.000493572016677141,
      "loss": 5.5312,
      "mean_token_accuracy": 0.18407864719629288,
      "num_tokens": 17223701.0,
      "step": 7715
    },
    {
      "entropy": 5.618215799331665,
      "epoch": 0.8670747458864492,
      "grad_norm": 1.0,
      "learning_rate": 0.0004935624850468148,
      "loss": 5.4317,
      "mean_token_accuracy": 0.1864711031317711,
      "num_tokens": 17234878.0,
      "step": 7720
    },
    {
      "entropy": 5.507683944702149,
      "epoch": 0.8676363227944067,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004935529464573742,
      "loss": 5.5017,
      "mean_token_accuracy": 0.1830043986439705,
      "num_tokens": 17245813.0,
      "step": 7725
    },
    {
      "entropy": 5.595293378829956,
      "epoch": 0.8681978997023643,
      "grad_norm": 1.0,
      "learning_rate": 0.0004935434009091232,
      "loss": 5.5815,
      "mean_token_accuracy": 0.18132366389036178,
      "num_tokens": 17257378.0,
      "step": 7730
    },
    {
      "entropy": 5.655421590805053,
      "epoch": 0.8687594766103218,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004935338484023655,
      "loss": 5.5168,
      "mean_token_accuracy": 0.1777476266026497,
      "num_tokens": 17269108.0,
      "step": 7735
    },
    {
      "entropy": 5.585360717773438,
      "epoch": 0.8693210535182794,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004935242889374052,
      "loss": 5.4943,
      "mean_token_accuracy": 0.17545924484729766,
      "num_tokens": 17280643.0,
      "step": 7740
    },
    {
      "entropy": 5.690800285339355,
      "epoch": 0.8698826304262369,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004935147225145469,
      "loss": 5.5603,
      "mean_token_accuracy": 0.17998234629631044,
      "num_tokens": 17291236.0,
      "step": 7745
    },
    {
      "entropy": 5.7303039073944095,
      "epoch": 0.8704442073341944,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004935051491340951,
      "loss": 5.6305,
      "mean_token_accuracy": 0.1695631831884384,
      "num_tokens": 17303547.0,
      "step": 7750
    },
    {
      "entropy": 5.726223516464233,
      "epoch": 0.8710057842421519,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004934955687963546,
      "loss": 5.6027,
      "mean_token_accuracy": 0.17243569046258928,
      "num_tokens": 17314803.0,
      "step": 7755
    },
    {
      "entropy": 5.568598127365112,
      "epoch": 0.8715673611501095,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004934859815016304,
      "loss": 5.4792,
      "mean_token_accuracy": 0.1801397055387497,
      "num_tokens": 17326009.0,
      "step": 7760
    },
    {
      "entropy": 5.635626983642578,
      "epoch": 0.872128938058067,
      "grad_norm": 1.0,
      "learning_rate": 0.0004934763872502278,
      "loss": 5.4904,
      "mean_token_accuracy": 0.18378062695264816,
      "num_tokens": 17337078.0,
      "step": 7765
    },
    {
      "entropy": 5.6486553192138675,
      "epoch": 0.8726905149660246,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004934667860424523,
      "loss": 5.6029,
      "mean_token_accuracy": 0.17538124173879624,
      "num_tokens": 17349266.0,
      "step": 7770
    },
    {
      "entropy": 5.464595460891724,
      "epoch": 0.8732520918739821,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004934571778786096,
      "loss": 5.4034,
      "mean_token_accuracy": 0.18144616186618806,
      "num_tokens": 17360327.0,
      "step": 7775
    },
    {
      "entropy": 5.679508781433105,
      "epoch": 0.8738136687819397,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004934475627590057,
      "loss": 5.5593,
      "mean_token_accuracy": 0.17058178782463074,
      "num_tokens": 17372727.0,
      "step": 7780
    },
    {
      "entropy": 5.712011480331421,
      "epoch": 0.8743752456898972,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004934379406839466,
      "loss": 5.6117,
      "mean_token_accuracy": 0.17371562570333482,
      "num_tokens": 17384160.0,
      "step": 7785
    },
    {
      "entropy": 5.605370426177979,
      "epoch": 0.8749368225978548,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934283116537387,
      "loss": 5.5269,
      "mean_token_accuracy": 0.17663144171237946,
      "num_tokens": 17396009.0,
      "step": 7790
    },
    {
      "entropy": 5.570238256454468,
      "epoch": 0.8754983995058123,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004934186756686888,
      "loss": 5.5287,
      "mean_token_accuracy": 0.18028004467487335,
      "num_tokens": 17406903.0,
      "step": 7795
    },
    {
      "entropy": 5.586734771728516,
      "epoch": 0.8760599764137699,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004934090327291034,
      "loss": 5.3941,
      "mean_token_accuracy": 0.18741592317819594,
      "num_tokens": 17417694.0,
      "step": 7800
    },
    {
      "entropy": 5.5621052265167235,
      "epoch": 0.8766215533217274,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004933993828352898,
      "loss": 5.6368,
      "mean_token_accuracy": 0.1808529257774353,
      "num_tokens": 17428731.0,
      "step": 7805
    },
    {
      "entropy": 5.559157037734986,
      "epoch": 0.877183130229685,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004933897259875551,
      "loss": 5.3837,
      "mean_token_accuracy": 0.18207520693540574,
      "num_tokens": 17439242.0,
      "step": 7810
    },
    {
      "entropy": 5.6173442840576175,
      "epoch": 0.8777447071376425,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004933800621862069,
      "loss": 5.4863,
      "mean_token_accuracy": 0.17750116288661957,
      "num_tokens": 17449980.0,
      "step": 7815
    },
    {
      "entropy": 5.623306751251221,
      "epoch": 0.8783062840456001,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004933703914315527,
      "loss": 5.592,
      "mean_token_accuracy": 0.16987940967082976,
      "num_tokens": 17464140.0,
      "step": 7820
    },
    {
      "entropy": 5.725029325485229,
      "epoch": 0.8788678609535576,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004933607137239006,
      "loss": 5.5997,
      "mean_token_accuracy": 0.17623118460178375,
      "num_tokens": 17476246.0,
      "step": 7825
    },
    {
      "entropy": 5.660158109664917,
      "epoch": 0.8794294378615152,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004933510290635587,
      "loss": 5.6103,
      "mean_token_accuracy": 0.16733368188142778,
      "num_tokens": 17488169.0,
      "step": 7830
    },
    {
      "entropy": 5.5224262237548825,
      "epoch": 0.8799910147694727,
      "grad_norm": 1.0,
      "learning_rate": 0.0004933413374508353,
      "loss": 5.4298,
      "mean_token_accuracy": 0.18274897038936616,
      "num_tokens": 17498465.0,
      "step": 7835
    },
    {
      "entropy": 5.593181943893432,
      "epoch": 0.8805525916774303,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004933316388860391,
      "loss": 5.4271,
      "mean_token_accuracy": 0.18860915303230286,
      "num_tokens": 17509217.0,
      "step": 7840
    },
    {
      "entropy": 5.572647142410278,
      "epoch": 0.8811141685853877,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004933219333694787,
      "loss": 5.4612,
      "mean_token_accuracy": 0.18687058687210084,
      "num_tokens": 17519851.0,
      "step": 7845
    },
    {
      "entropy": 5.569455528259278,
      "epoch": 0.8816757454933453,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004933122209014634,
      "loss": 5.5286,
      "mean_token_accuracy": 0.1822614923119545,
      "num_tokens": 17532144.0,
      "step": 7850
    },
    {
      "entropy": 5.6123206615448,
      "epoch": 0.8822373224013028,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004933025014823022,
      "loss": 5.4477,
      "mean_token_accuracy": 0.18794535398483275,
      "num_tokens": 17542929.0,
      "step": 7855
    },
    {
      "entropy": 5.6424397945404055,
      "epoch": 0.8827988993092604,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004932927751123048,
      "loss": 5.6132,
      "mean_token_accuracy": 0.1730242982506752,
      "num_tokens": 17554931.0,
      "step": 7860
    },
    {
      "entropy": 5.524155998229981,
      "epoch": 0.8833604762172179,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004932830417917808,
      "loss": 5.4173,
      "mean_token_accuracy": 0.18392786830663682,
      "num_tokens": 17566026.0,
      "step": 7865
    },
    {
      "entropy": 5.602767896652222,
      "epoch": 0.8839220531251755,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00049327330152104,
      "loss": 5.525,
      "mean_token_accuracy": 0.18074765652418137,
      "num_tokens": 17578396.0,
      "step": 7870
    },
    {
      "entropy": 5.66438627243042,
      "epoch": 0.884483630033133,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004932635543003927,
      "loss": 5.5199,
      "mean_token_accuracy": 0.182549124956131,
      "num_tokens": 17590559.0,
      "step": 7875
    },
    {
      "entropy": 5.553819227218628,
      "epoch": 0.8850452069410906,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004932538001301491,
      "loss": 5.4882,
      "mean_token_accuracy": 0.17640931904315948,
      "num_tokens": 17600833.0,
      "step": 7880
    },
    {
      "entropy": 5.5474964618682865,
      "epoch": 0.8856067838490481,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004932440390106199,
      "loss": 5.4701,
      "mean_token_accuracy": 0.18419776856899261,
      "num_tokens": 17612166.0,
      "step": 7885
    },
    {
      "entropy": 5.653920030593872,
      "epoch": 0.8861683607570057,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004932342709421159,
      "loss": 5.5832,
      "mean_token_accuracy": 0.17716454714536667,
      "num_tokens": 17623070.0,
      "step": 7890
    },
    {
      "entropy": 5.673707628250122,
      "epoch": 0.8867299376649632,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000493224495924948,
      "loss": 5.6412,
      "mean_token_accuracy": 0.17596878111362457,
      "num_tokens": 17633319.0,
      "step": 7895
    },
    {
      "entropy": 5.601524114608765,
      "epoch": 0.8872915145729208,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004932147139594276,
      "loss": 5.4527,
      "mean_token_accuracy": 0.18103813529014587,
      "num_tokens": 17643818.0,
      "step": 7900
    },
    {
      "entropy": 5.580865526199341,
      "epoch": 0.8878530914808783,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004932049250458659,
      "loss": 5.4478,
      "mean_token_accuracy": 0.18300647139549256,
      "num_tokens": 17655665.0,
      "step": 7905
    },
    {
      "entropy": 5.56022891998291,
      "epoch": 0.8884146683888359,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004931951291845749,
      "loss": 5.3737,
      "mean_token_accuracy": 0.1908242642879486,
      "num_tokens": 17666200.0,
      "step": 7910
    },
    {
      "entropy": 5.550500535964966,
      "epoch": 0.8889762452967934,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004931853263758663,
      "loss": 5.5856,
      "mean_token_accuracy": 0.17779973596334459,
      "num_tokens": 17677299.0,
      "step": 7915
    },
    {
      "entropy": 5.598557758331299,
      "epoch": 0.889537822204751,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004931755166200523,
      "loss": 5.4302,
      "mean_token_accuracy": 0.1908480182290077,
      "num_tokens": 17687870.0,
      "step": 7920
    },
    {
      "entropy": 5.579096794128418,
      "epoch": 0.8900993991127085,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004931656999174452,
      "loss": 5.4567,
      "mean_token_accuracy": 0.18092368841171264,
      "num_tokens": 17698853.0,
      "step": 7925
    },
    {
      "entropy": 5.567067193984985,
      "epoch": 0.8906609760206661,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004931558762683577,
      "loss": 5.5095,
      "mean_token_accuracy": 0.18371810168027877,
      "num_tokens": 17709897.0,
      "step": 7930
    },
    {
      "entropy": 5.595690393447876,
      "epoch": 0.8912225529286236,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004931460456731024,
      "loss": 5.5107,
      "mean_token_accuracy": 0.1784860759973526,
      "num_tokens": 17720974.0,
      "step": 7935
    },
    {
      "entropy": 5.736320924758911,
      "epoch": 0.8917841298365812,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004931362081319925,
      "loss": 5.6477,
      "mean_token_accuracy": 0.17302757054567336,
      "num_tokens": 17734449.0,
      "step": 7940
    },
    {
      "entropy": 5.485985565185547,
      "epoch": 0.8923457067445386,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004931263636453409,
      "loss": 5.3903,
      "mean_token_accuracy": 0.18473621755838393,
      "num_tokens": 17744735.0,
      "step": 7945
    },
    {
      "entropy": 5.533071231842041,
      "epoch": 0.8929072836524962,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004931165122134616,
      "loss": 5.4178,
      "mean_token_accuracy": 0.18630819767713547,
      "num_tokens": 17755641.0,
      "step": 7950
    },
    {
      "entropy": 5.641396427154541,
      "epoch": 0.8934688605604537,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004931066538366677,
      "loss": 5.5324,
      "mean_token_accuracy": 0.17234542667865754,
      "num_tokens": 17767621.0,
      "step": 7955
    },
    {
      "entropy": 5.680049991607666,
      "epoch": 0.8940304374684113,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004930967885152734,
      "loss": 5.6277,
      "mean_token_accuracy": 0.1737061321735382,
      "num_tokens": 17779507.0,
      "step": 7960
    },
    {
      "entropy": 5.543487548828125,
      "epoch": 0.8945920143763688,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004930869162495929,
      "loss": 5.4693,
      "mean_token_accuracy": 0.17781761288642883,
      "num_tokens": 17790371.0,
      "step": 7965
    },
    {
      "entropy": 5.603549337387085,
      "epoch": 0.8951535912843264,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004930770370399403,
      "loss": 5.4923,
      "mean_token_accuracy": 0.1885839194059372,
      "num_tokens": 17800903.0,
      "step": 7970
    },
    {
      "entropy": 5.712726497650147,
      "epoch": 0.8957151681922839,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004930671508866303,
      "loss": 5.6715,
      "mean_token_accuracy": 0.16963825598359109,
      "num_tokens": 17812053.0,
      "step": 7975
    },
    {
      "entropy": 5.577488851547241,
      "epoch": 0.8962767451002415,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004930572577899776,
      "loss": 5.4038,
      "mean_token_accuracy": 0.18471201211214067,
      "num_tokens": 17822113.0,
      "step": 7980
    },
    {
      "entropy": 5.536539506912232,
      "epoch": 0.896838322008199,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004930473577502972,
      "loss": 5.4442,
      "mean_token_accuracy": 0.18067950308322905,
      "num_tokens": 17832741.0,
      "step": 7985
    },
    {
      "entropy": 5.445788621902466,
      "epoch": 0.8973998989161566,
      "grad_norm": 1.0,
      "learning_rate": 0.0004930374507679044,
      "loss": 5.3738,
      "mean_token_accuracy": 0.17739315032958985,
      "num_tokens": 17842968.0,
      "step": 7990
    },
    {
      "entropy": 5.68609652519226,
      "epoch": 0.8979614758241141,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004930275368431146,
      "loss": 5.6441,
      "mean_token_accuracy": 0.167500863969326,
      "num_tokens": 17854295.0,
      "step": 7995
    },
    {
      "entropy": 5.724550437927246,
      "epoch": 0.8985230527320717,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004930176159762435,
      "loss": 5.565,
      "mean_token_accuracy": 0.18175942301750184,
      "num_tokens": 17865597.0,
      "step": 8000
    },
    {
      "entropy": 5.452319383621216,
      "epoch": 0.8990846296400292,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004930076881676069,
      "loss": 5.342,
      "mean_token_accuracy": 0.18693463057279586,
      "num_tokens": 17877405.0,
      "step": 8005
    },
    {
      "entropy": 5.69283709526062,
      "epoch": 0.8996462065479868,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004929977534175211,
      "loss": 5.5456,
      "mean_token_accuracy": 0.17947353273630143,
      "num_tokens": 17889273.0,
      "step": 8010
    },
    {
      "entropy": 5.5469183921813965,
      "epoch": 0.9002077834559443,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004929878117263021,
      "loss": 5.4435,
      "mean_token_accuracy": 0.1848416194319725,
      "num_tokens": 17899761.0,
      "step": 8015
    },
    {
      "entropy": 5.553060245513916,
      "epoch": 0.9007693603639019,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004929778630942668,
      "loss": 5.4433,
      "mean_token_accuracy": 0.18374273478984832,
      "num_tokens": 17910607.0,
      "step": 8020
    },
    {
      "entropy": 5.6265302181243895,
      "epoch": 0.9013309372718594,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004929679075217317,
      "loss": 5.5055,
      "mean_token_accuracy": 0.17801229804754257,
      "num_tokens": 17922538.0,
      "step": 8025
    },
    {
      "entropy": 5.638562345504761,
      "epoch": 0.901892514179817,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004929579450090139,
      "loss": 5.5479,
      "mean_token_accuracy": 0.1817954033613205,
      "num_tokens": 17933313.0,
      "step": 8030
    },
    {
      "entropy": 5.673544836044312,
      "epoch": 0.9024540910877745,
      "grad_norm": 1.0,
      "learning_rate": 0.0004929479755564305,
      "loss": 5.632,
      "mean_token_accuracy": 0.1749018445611,
      "num_tokens": 17944893.0,
      "step": 8035
    },
    {
      "entropy": 5.606023740768433,
      "epoch": 0.903015667995732,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004929379991642991,
      "loss": 5.5149,
      "mean_token_accuracy": 0.1786926954984665,
      "num_tokens": 17955019.0,
      "step": 8040
    },
    {
      "entropy": 5.6634604930877686,
      "epoch": 0.9035772449036895,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004929280158329373,
      "loss": 5.5422,
      "mean_token_accuracy": 0.18044948875904082,
      "num_tokens": 17965793.0,
      "step": 8045
    },
    {
      "entropy": 5.551074838638305,
      "epoch": 0.9041388218116471,
      "grad_norm": 1.125,
      "learning_rate": 0.000492918025562663,
      "loss": 5.3656,
      "mean_token_accuracy": 0.18530188649892806,
      "num_tokens": 17975709.0,
      "step": 8050
    },
    {
      "entropy": 5.63584451675415,
      "epoch": 0.9047003987196046,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004929080283537942,
      "loss": 5.5954,
      "mean_token_accuracy": 0.17713356465101243,
      "num_tokens": 17985405.0,
      "step": 8055
    },
    {
      "entropy": 5.69279899597168,
      "epoch": 0.9052619756275622,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004928980242066492,
      "loss": 5.6199,
      "mean_token_accuracy": 0.17835464477539062,
      "num_tokens": 17997105.0,
      "step": 8060
    },
    {
      "entropy": 5.630202341079712,
      "epoch": 0.9058235525355197,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004928880131215467,
      "loss": 5.5167,
      "mean_token_accuracy": 0.18688323348760605,
      "num_tokens": 18007914.0,
      "step": 8065
    },
    {
      "entropy": 5.534109878540039,
      "epoch": 0.9063851294434773,
      "grad_norm": 1.125,
      "learning_rate": 0.0004928779950988053,
      "loss": 5.5166,
      "mean_token_accuracy": 0.18196988403797149,
      "num_tokens": 18018328.0,
      "step": 8070
    },
    {
      "entropy": 5.592997598648071,
      "epoch": 0.9069467063514348,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000492867970138744,
      "loss": 5.4732,
      "mean_token_accuracy": 0.18148116916418075,
      "num_tokens": 18028568.0,
      "step": 8075
    },
    {
      "entropy": 5.627342462539673,
      "epoch": 0.9075082832593924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004928579382416821,
      "loss": 5.4917,
      "mean_token_accuracy": 0.18100592792034148,
      "num_tokens": 18038655.0,
      "step": 8080
    },
    {
      "entropy": 5.5573585510253904,
      "epoch": 0.9080698601673499,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492847899407939,
      "loss": 5.4214,
      "mean_token_accuracy": 0.18531959056854247,
      "num_tokens": 18049716.0,
      "step": 8085
    },
    {
      "entropy": 5.418399143218994,
      "epoch": 0.9086314370753075,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004928378536378341,
      "loss": 5.4153,
      "mean_token_accuracy": 0.18372095227241517,
      "num_tokens": 18062530.0,
      "step": 8090
    },
    {
      "entropy": 5.559502267837525,
      "epoch": 0.909193013983265,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004928278009316877,
      "loss": 5.508,
      "mean_token_accuracy": 0.17610028833150865,
      "num_tokens": 18073076.0,
      "step": 8095
    },
    {
      "entropy": 5.695038843154907,
      "epoch": 0.9097545908912226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004928177412898195,
      "loss": 5.4742,
      "mean_token_accuracy": 0.17959958761930467,
      "num_tokens": 18083759.0,
      "step": 8100
    },
    {
      "entropy": 5.470184278488159,
      "epoch": 0.9103161677991801,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004928076747125499,
      "loss": 5.4145,
      "mean_token_accuracy": 0.1759122133255005,
      "num_tokens": 18095980.0,
      "step": 8105
    },
    {
      "entropy": 5.582666730880737,
      "epoch": 0.9108777447071377,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004927976012001995,
      "loss": 5.5138,
      "mean_token_accuracy": 0.18386195302009584,
      "num_tokens": 18106672.0,
      "step": 8110
    },
    {
      "entropy": 5.564055871963501,
      "epoch": 0.9114393216150952,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004927875207530889,
      "loss": 5.5107,
      "mean_token_accuracy": 0.1788438469171524,
      "num_tokens": 18117481.0,
      "step": 8115
    },
    {
      "entropy": 5.546481370925903,
      "epoch": 0.9120008985230528,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004927774333715394,
      "loss": 5.4453,
      "mean_token_accuracy": 0.18264017403125762,
      "num_tokens": 18127602.0,
      "step": 8120
    },
    {
      "entropy": 5.665448760986328,
      "epoch": 0.9125624754310103,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004927673390558718,
      "loss": 5.5225,
      "mean_token_accuracy": 0.17809127867221833,
      "num_tokens": 18140109.0,
      "step": 8125
    },
    {
      "entropy": 5.551574611663819,
      "epoch": 0.9131240523389679,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004927572378064077,
      "loss": 5.4111,
      "mean_token_accuracy": 0.18579337745904922,
      "num_tokens": 18150271.0,
      "step": 8130
    },
    {
      "entropy": 5.551731109619141,
      "epoch": 0.9136856292469253,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004927471296234687,
      "loss": 5.5347,
      "mean_token_accuracy": 0.18156878054141998,
      "num_tokens": 18160746.0,
      "step": 8135
    },
    {
      "entropy": 5.681353235244751,
      "epoch": 0.9142472061548829,
      "grad_norm": 1.0,
      "learning_rate": 0.0004927370145073767,
      "loss": 5.5083,
      "mean_token_accuracy": 0.1790511950850487,
      "num_tokens": 18172271.0,
      "step": 8140
    },
    {
      "entropy": 5.579694557189941,
      "epoch": 0.9148087830628404,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004927268924584536,
      "loss": 5.4419,
      "mean_token_accuracy": 0.18102690130472182,
      "num_tokens": 18183116.0,
      "step": 8145
    },
    {
      "entropy": 5.51436710357666,
      "epoch": 0.915370359970798,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004927167634770219,
      "loss": 5.4092,
      "mean_token_accuracy": 0.19384038895368577,
      "num_tokens": 18195147.0,
      "step": 8150
    },
    {
      "entropy": 5.570596933364868,
      "epoch": 0.9159319368787555,
      "grad_norm": 1.21875,
      "learning_rate": 0.000492706627563404,
      "loss": 5.4841,
      "mean_token_accuracy": 0.1778377816081047,
      "num_tokens": 18206217.0,
      "step": 8155
    },
    {
      "entropy": 5.550006151199341,
      "epoch": 0.9164935137867131,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004926964847179225,
      "loss": 5.4855,
      "mean_token_accuracy": 0.1773753508925438,
      "num_tokens": 18217434.0,
      "step": 8160
    },
    {
      "entropy": 5.608560085296631,
      "epoch": 0.9170550906946706,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004926863349409006,
      "loss": 5.4922,
      "mean_token_accuracy": 0.17910895496606827,
      "num_tokens": 18229277.0,
      "step": 8165
    },
    {
      "entropy": 5.526487493515015,
      "epoch": 0.9176166676026282,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004926761782326613,
      "loss": 5.3497,
      "mean_token_accuracy": 0.18623987585306168,
      "num_tokens": 18240320.0,
      "step": 8170
    },
    {
      "entropy": 5.463681316375732,
      "epoch": 0.9181782445105857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004926660145935282,
      "loss": 5.3871,
      "mean_token_accuracy": 0.18620867878198624,
      "num_tokens": 18250380.0,
      "step": 8175
    },
    {
      "entropy": 5.583861589431763,
      "epoch": 0.9187398214185433,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004926558440238247,
      "loss": 5.5694,
      "mean_token_accuracy": 0.18073032200336456,
      "num_tokens": 18261703.0,
      "step": 8180
    },
    {
      "entropy": 5.621598863601685,
      "epoch": 0.9193013983265008,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004926456665238746,
      "loss": 5.496,
      "mean_token_accuracy": 0.1756924122571945,
      "num_tokens": 18272472.0,
      "step": 8185
    },
    {
      "entropy": 5.674906826019287,
      "epoch": 0.9198629752344584,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004926354820940023,
      "loss": 5.5723,
      "mean_token_accuracy": 0.1792101725935936,
      "num_tokens": 18282168.0,
      "step": 8190
    },
    {
      "entropy": 5.687933826446534,
      "epoch": 0.9204245521424159,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004926252907345317,
      "loss": 5.5731,
      "mean_token_accuracy": 0.17163886725902558,
      "num_tokens": 18293750.0,
      "step": 8195
    },
    {
      "entropy": 5.5886390686035154,
      "epoch": 0.9209861290503735,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004926150924457874,
      "loss": 5.5,
      "mean_token_accuracy": 0.18049391061067582,
      "num_tokens": 18304568.0,
      "step": 8200
    },
    {
      "entropy": 5.562101364135742,
      "epoch": 0.921547705958331,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004926048872280943,
      "loss": 5.4241,
      "mean_token_accuracy": 0.1870491161942482,
      "num_tokens": 18315129.0,
      "step": 8205
    },
    {
      "entropy": 5.434761047363281,
      "epoch": 0.9221092828662886,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492594675081777,
      "loss": 5.3787,
      "mean_token_accuracy": 0.18584675639867781,
      "num_tokens": 18324628.0,
      "step": 8210
    },
    {
      "entropy": 5.466102457046508,
      "epoch": 0.9226708597742461,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000492584456007161,
      "loss": 5.4009,
      "mean_token_accuracy": 0.18903011679649354,
      "num_tokens": 18334899.0,
      "step": 8215
    },
    {
      "entropy": 5.617522048950195,
      "epoch": 0.9232324366822037,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004925742300045716,
      "loss": 5.4078,
      "mean_token_accuracy": 0.1870238482952118,
      "num_tokens": 18345719.0,
      "step": 8220
    },
    {
      "entropy": 5.561037397384643,
      "epoch": 0.9237940135901612,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004925639970743341,
      "loss": 5.4906,
      "mean_token_accuracy": 0.18776837289333342,
      "num_tokens": 18355814.0,
      "step": 8225
    },
    {
      "entropy": 5.588109254837036,
      "epoch": 0.9243555904981187,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004925537572167746,
      "loss": 5.5545,
      "mean_token_accuracy": 0.17607142478227616,
      "num_tokens": 18366746.0,
      "step": 8230
    },
    {
      "entropy": 5.51508412361145,
      "epoch": 0.9249171674060762,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004925435104322192,
      "loss": 5.3827,
      "mean_token_accuracy": 0.1891380324959755,
      "num_tokens": 18375846.0,
      "step": 8235
    },
    {
      "entropy": 5.514292907714844,
      "epoch": 0.9254787443140338,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492533256720994,
      "loss": 5.5051,
      "mean_token_accuracy": 0.1810300201177597,
      "num_tokens": 18386850.0,
      "step": 8240
    },
    {
      "entropy": 5.538437843322754,
      "epoch": 0.9260403212219913,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004925229960834256,
      "loss": 5.3958,
      "mean_token_accuracy": 0.18939542025327682,
      "num_tokens": 18396347.0,
      "step": 8245
    },
    {
      "entropy": 5.567156553268433,
      "epoch": 0.9266018981299489,
      "grad_norm": 1.125,
      "learning_rate": 0.0004925127285198407,
      "loss": 5.4436,
      "mean_token_accuracy": 0.18140185326337815,
      "num_tokens": 18406583.0,
      "step": 8250
    },
    {
      "entropy": 5.679202651977539,
      "epoch": 0.9271634750379064,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000492502454030566,
      "loss": 5.5593,
      "mean_token_accuracy": 0.17679764479398727,
      "num_tokens": 18418586.0,
      "step": 8255
    },
    {
      "entropy": 5.5931939601898195,
      "epoch": 0.927725051945864,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004924921726159288,
      "loss": 5.4015,
      "mean_token_accuracy": 0.18384822010993956,
      "num_tokens": 18429752.0,
      "step": 8260
    },
    {
      "entropy": 5.575476551055909,
      "epoch": 0.9282866288538215,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004924818842762565,
      "loss": 5.4969,
      "mean_token_accuracy": 0.1765847384929657,
      "num_tokens": 18440663.0,
      "step": 8265
    },
    {
      "entropy": 5.456879949569702,
      "epoch": 0.9288482057617791,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004924715890118768,
      "loss": 5.4371,
      "mean_token_accuracy": 0.18375977724790574,
      "num_tokens": 18451497.0,
      "step": 8270
    },
    {
      "entropy": 5.543888425827026,
      "epoch": 0.9294097826697366,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004924612868231173,
      "loss": 5.431,
      "mean_token_accuracy": 0.17825741618871688,
      "num_tokens": 18462794.0,
      "step": 8275
    },
    {
      "entropy": 5.6037522315979,
      "epoch": 0.9299713595776942,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004924509777103059,
      "loss": 5.4277,
      "mean_token_accuracy": 0.18658117055892945,
      "num_tokens": 18472973.0,
      "step": 8280
    },
    {
      "entropy": 5.635124921798706,
      "epoch": 0.9305329364856517,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004924406616737712,
      "loss": 5.5903,
      "mean_token_accuracy": 0.1796143800020218,
      "num_tokens": 18484451.0,
      "step": 8285
    },
    {
      "entropy": 5.611140489578247,
      "epoch": 0.9310945133936093,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004924303387138415,
      "loss": 5.4427,
      "mean_token_accuracy": 0.1901862293481827,
      "num_tokens": 18496412.0,
      "step": 8290
    },
    {
      "entropy": 5.556115007400512,
      "epoch": 0.9316560903015668,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004924200088308455,
      "loss": 5.4706,
      "mean_token_accuracy": 0.1861564189195633,
      "num_tokens": 18508547.0,
      "step": 8295
    },
    {
      "entropy": 5.514981889724732,
      "epoch": 0.9322176672095244,
      "grad_norm": 1.09375,
      "learning_rate": 0.000492409672025112,
      "loss": 5.513,
      "mean_token_accuracy": 0.175574454665184,
      "num_tokens": 18518954.0,
      "step": 8300
    },
    {
      "entropy": 5.528355646133423,
      "epoch": 0.9327792441174819,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923993282969703,
      "loss": 5.3676,
      "mean_token_accuracy": 0.18600136190652847,
      "num_tokens": 18530444.0,
      "step": 8305
    },
    {
      "entropy": 5.545026731491089,
      "epoch": 0.9333408210254395,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923889776467495,
      "loss": 5.4264,
      "mean_token_accuracy": 0.18126846700906754,
      "num_tokens": 18541612.0,
      "step": 8310
    },
    {
      "entropy": 5.616086959838867,
      "epoch": 0.933902397933397,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004923786200747795,
      "loss": 5.5977,
      "mean_token_accuracy": 0.17694150507450104,
      "num_tokens": 18553316.0,
      "step": 8315
    },
    {
      "entropy": 5.60566201210022,
      "epoch": 0.9344639748413546,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049236825558139,
      "loss": 5.4627,
      "mean_token_accuracy": 0.18421150743961334,
      "num_tokens": 18564468.0,
      "step": 8320
    },
    {
      "entropy": 5.501161336898804,
      "epoch": 0.9350255517493121,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004923578841669108,
      "loss": 5.3976,
      "mean_token_accuracy": 0.18672065287828446,
      "num_tokens": 18575536.0,
      "step": 8325
    },
    {
      "entropy": 5.5153326988220215,
      "epoch": 0.9355871286572696,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004923475058316723,
      "loss": 5.5212,
      "mean_token_accuracy": 0.1734934478998184,
      "num_tokens": 18586906.0,
      "step": 8330
    },
    {
      "entropy": 5.623011255264283,
      "epoch": 0.9361487055652271,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923371205760049,
      "loss": 5.503,
      "mean_token_accuracy": 0.18519700616598128,
      "num_tokens": 18599064.0,
      "step": 8335
    },
    {
      "entropy": 5.60233063697815,
      "epoch": 0.9367102824731847,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004923267284002394,
      "loss": 5.5367,
      "mean_token_accuracy": 0.18010803312063217,
      "num_tokens": 18610057.0,
      "step": 8340
    },
    {
      "entropy": 5.434579849243164,
      "epoch": 0.9372718593811422,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004923163293047065,
      "loss": 5.3641,
      "mean_token_accuracy": 0.18760113418102264,
      "num_tokens": 18620008.0,
      "step": 8345
    },
    {
      "entropy": 5.479099082946777,
      "epoch": 0.9378334362890998,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004923059232897374,
      "loss": 5.3609,
      "mean_token_accuracy": 0.18557711094617843,
      "num_tokens": 18630599.0,
      "step": 8350
    },
    {
      "entropy": 5.573970603942871,
      "epoch": 0.9383950131970573,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004922955103556633,
      "loss": 5.5379,
      "mean_token_accuracy": 0.1812148243188858,
      "num_tokens": 18643898.0,
      "step": 8355
    },
    {
      "entropy": 5.660813236236573,
      "epoch": 0.9389565901050149,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000492285090502816,
      "loss": 5.5152,
      "mean_token_accuracy": 0.17586288154125213,
      "num_tokens": 18654760.0,
      "step": 8360
    },
    {
      "entropy": 5.6697930812835695,
      "epoch": 0.9395181670129724,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004922746637315272,
      "loss": 5.49,
      "mean_token_accuracy": 0.18287123441696168,
      "num_tokens": 18666036.0,
      "step": 8365
    },
    {
      "entropy": 5.622818899154663,
      "epoch": 0.94007974392093,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004922642300421288,
      "loss": 5.5341,
      "mean_token_accuracy": 0.18295766413211823,
      "num_tokens": 18677806.0,
      "step": 8370
    },
    {
      "entropy": 5.612994861602783,
      "epoch": 0.9406413208288875,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000492253789434953,
      "loss": 5.6516,
      "mean_token_accuracy": 0.1783110097050667,
      "num_tokens": 18688971.0,
      "step": 8375
    },
    {
      "entropy": 5.589858484268189,
      "epoch": 0.9412028977368451,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004922433419103323,
      "loss": 5.5247,
      "mean_token_accuracy": 0.17627423405647277,
      "num_tokens": 18700153.0,
      "step": 8380
    },
    {
      "entropy": 5.715108442306518,
      "epoch": 0.9417644746448026,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004922328874685993,
      "loss": 5.5649,
      "mean_token_accuracy": 0.179406775534153,
      "num_tokens": 18711935.0,
      "step": 8385
    },
    {
      "entropy": 5.487036943435669,
      "epoch": 0.9423260515527602,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004922224261100869,
      "loss": 5.3781,
      "mean_token_accuracy": 0.18586247116327287,
      "num_tokens": 18722694.0,
      "step": 8390
    },
    {
      "entropy": 5.586714458465576,
      "epoch": 0.9428876284607177,
      "grad_norm": 1.0,
      "learning_rate": 0.0004922119578351283,
      "loss": 5.5388,
      "mean_token_accuracy": 0.18508632779121398,
      "num_tokens": 18733443.0,
      "step": 8395
    },
    {
      "entropy": 5.6142387866973875,
      "epoch": 0.9434492053686753,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004922014826440566,
      "loss": 5.4596,
      "mean_token_accuracy": 0.18226936012506484,
      "num_tokens": 18743965.0,
      "step": 8400
    },
    {
      "entropy": 5.560277986526489,
      "epoch": 0.9440107822766328,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004921910005372056,
      "loss": 5.5514,
      "mean_token_accuracy": 0.17946840226650237,
      "num_tokens": 18755712.0,
      "step": 8405
    },
    {
      "entropy": 5.579993200302124,
      "epoch": 0.9445723591845904,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004921805115149088,
      "loss": 5.4461,
      "mean_token_accuracy": 0.18157842755317688,
      "num_tokens": 18766756.0,
      "step": 8410
    },
    {
      "entropy": 5.581419563293457,
      "epoch": 0.9451339360925479,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004921700155775002,
      "loss": 5.5481,
      "mean_token_accuracy": 0.17493135780096053,
      "num_tokens": 18776932.0,
      "step": 8415
    },
    {
      "entropy": 5.607511425018311,
      "epoch": 0.9456955130005055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004921595127253142,
      "loss": 5.4681,
      "mean_token_accuracy": 0.18051346987485886,
      "num_tokens": 18787197.0,
      "step": 8420
    },
    {
      "entropy": 5.512584781646728,
      "epoch": 0.9462570899084629,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004921490029586852,
      "loss": 5.4303,
      "mean_token_accuracy": 0.18762568533420562,
      "num_tokens": 18797861.0,
      "step": 8425
    },
    {
      "entropy": 5.535125875473023,
      "epoch": 0.9468186668164205,
      "grad_norm": 1.0,
      "learning_rate": 0.0004921384862779475,
      "loss": 5.4368,
      "mean_token_accuracy": 0.18415761291980742,
      "num_tokens": 18809111.0,
      "step": 8430
    },
    {
      "entropy": 5.6051353931427,
      "epoch": 0.947380243724378,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004921279626834363,
      "loss": 5.4542,
      "mean_token_accuracy": 0.1931390032172203,
      "num_tokens": 18820236.0,
      "step": 8435
    },
    {
      "entropy": 5.523641204833984,
      "epoch": 0.9479418206323356,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004921174321754865,
      "loss": 5.4247,
      "mean_token_accuracy": 0.18735523670911788,
      "num_tokens": 18831942.0,
      "step": 8440
    },
    {
      "entropy": 5.559072494506836,
      "epoch": 0.9485033975402931,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004921068947544335,
      "loss": 5.4836,
      "mean_token_accuracy": 0.18535945266485215,
      "num_tokens": 18843271.0,
      "step": 8445
    },
    {
      "entropy": 5.581586647033691,
      "epoch": 0.9490649744482507,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004920963504206128,
      "loss": 5.5181,
      "mean_token_accuracy": 0.18354747593402862,
      "num_tokens": 18856285.0,
      "step": 8450
    },
    {
      "entropy": 5.667445230484009,
      "epoch": 0.9496265513562082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00049208579917436,
      "loss": 5.5354,
      "mean_token_accuracy": 0.17863369584083558,
      "num_tokens": 18867203.0,
      "step": 8455
    },
    {
      "entropy": 5.6186872005462645,
      "epoch": 0.9501881282641658,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004920752410160113,
      "loss": 5.5895,
      "mean_token_accuracy": 0.17248342782258988,
      "num_tokens": 18878598.0,
      "step": 8460
    },
    {
      "entropy": 5.656963253021241,
      "epoch": 0.9507497051721233,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004920646759459026,
      "loss": 5.486,
      "mean_token_accuracy": 0.18291398584842683,
      "num_tokens": 18890233.0,
      "step": 8465
    },
    {
      "entropy": 5.614354228973388,
      "epoch": 0.9513112820800809,
      "grad_norm": 1.125,
      "learning_rate": 0.0004920541039643705,
      "loss": 5.5555,
      "mean_token_accuracy": 0.17716809660196303,
      "num_tokens": 18900362.0,
      "step": 8470
    },
    {
      "entropy": 5.601590394973755,
      "epoch": 0.9518728589880384,
      "grad_norm": 1.125,
      "learning_rate": 0.0004920435250717515,
      "loss": 5.5011,
      "mean_token_accuracy": 0.18710409849882126,
      "num_tokens": 18911924.0,
      "step": 8475
    },
    {
      "entropy": 5.6120593547821045,
      "epoch": 0.952434435895996,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004920329392683826,
      "loss": 5.5114,
      "mean_token_accuracy": 0.1798287019133568,
      "num_tokens": 18923931.0,
      "step": 8480
    },
    {
      "entropy": 5.569701528549194,
      "epoch": 0.9529960128039535,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004920223465546006,
      "loss": 5.4259,
      "mean_token_accuracy": 0.18338057398796082,
      "num_tokens": 18935859.0,
      "step": 8485
    },
    {
      "entropy": 5.604145479202271,
      "epoch": 0.9535575897119111,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004920117469307429,
      "loss": 5.5531,
      "mean_token_accuracy": 0.17262073755264282,
      "num_tokens": 18946939.0,
      "step": 8490
    },
    {
      "entropy": 5.617562627792358,
      "epoch": 0.9541191666198686,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004920011403971471,
      "loss": 5.5297,
      "mean_token_accuracy": 0.17581794559955596,
      "num_tokens": 18958067.0,
      "step": 8495
    },
    {
      "entropy": 5.567305421829223,
      "epoch": 0.9546807435278262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004919905269541508,
      "loss": 5.4349,
      "mean_token_accuracy": 0.18979771435260773,
      "num_tokens": 18967920.0,
      "step": 8500
    },
    {
      "entropy": 5.5682543277740475,
      "epoch": 0.9552423204357837,
      "grad_norm": 1.078125,
      "learning_rate": 0.000491979906602092,
      "loss": 5.4654,
      "mean_token_accuracy": 0.18660126626491547,
      "num_tokens": 18978244.0,
      "step": 8505
    },
    {
      "entropy": 5.584858179092407,
      "epoch": 0.9558038973437413,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004919692793413089,
      "loss": 5.4249,
      "mean_token_accuracy": 0.18310647010803222,
      "num_tokens": 18988464.0,
      "step": 8510
    },
    {
      "entropy": 5.487370204925537,
      "epoch": 0.9563654742516988,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004919586451721396,
      "loss": 5.3872,
      "mean_token_accuracy": 0.1830912634730339,
      "num_tokens": 18998318.0,
      "step": 8515
    },
    {
      "entropy": 5.42850890159607,
      "epoch": 0.9569270511596563,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004919480040949229,
      "loss": 5.3693,
      "mean_token_accuracy": 0.187215156853199,
      "num_tokens": 19009802.0,
      "step": 8520
    },
    {
      "entropy": 5.691130685806274,
      "epoch": 0.9574886280676138,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004919373561099977,
      "loss": 5.5971,
      "mean_token_accuracy": 0.1797230139374733,
      "num_tokens": 19020174.0,
      "step": 8525
    },
    {
      "entropy": 5.621046876907348,
      "epoch": 0.9580502049755714,
      "grad_norm": 1.0,
      "learning_rate": 0.000491926701217703,
      "loss": 5.5008,
      "mean_token_accuracy": 0.17844899594783784,
      "num_tokens": 19031505.0,
      "step": 8530
    },
    {
      "entropy": 5.537479114532471,
      "epoch": 0.9586117818835289,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004919160394183779,
      "loss": 5.4162,
      "mean_token_accuracy": 0.18335415422916412,
      "num_tokens": 19042052.0,
      "step": 8535
    },
    {
      "entropy": 5.73462405204773,
      "epoch": 0.9591733587914865,
      "grad_norm": 1.046875,
      "learning_rate": 0.000491905370712362,
      "loss": 5.6042,
      "mean_token_accuracy": 0.17487475723028184,
      "num_tokens": 19052050.0,
      "step": 8540
    },
    {
      "entropy": 5.642431116104126,
      "epoch": 0.959734935699444,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004918946950999949,
      "loss": 5.5188,
      "mean_token_accuracy": 0.17947326004505157,
      "num_tokens": 19062053.0,
      "step": 8545
    },
    {
      "entropy": 5.530859279632568,
      "epoch": 0.9602965126074016,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004918840125816167,
      "loss": 5.4788,
      "mean_token_accuracy": 0.18358104079961776,
      "num_tokens": 19073095.0,
      "step": 8550
    },
    {
      "entropy": 5.615685939788818,
      "epoch": 0.9608580895153591,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004918733231575673,
      "loss": 5.5361,
      "mean_token_accuracy": 0.18429580926895142,
      "num_tokens": 19084521.0,
      "step": 8555
    },
    {
      "entropy": 5.64449553489685,
      "epoch": 0.9614196664233167,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004918626268281874,
      "loss": 5.5147,
      "mean_token_accuracy": 0.18146532624959946,
      "num_tokens": 19096372.0,
      "step": 8560
    },
    {
      "entropy": 5.6696178913116455,
      "epoch": 0.9619812433312742,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004918519235938171,
      "loss": 5.6342,
      "mean_token_accuracy": 0.1733379617333412,
      "num_tokens": 19108535.0,
      "step": 8565
    },
    {
      "entropy": 5.70990104675293,
      "epoch": 0.9625428202392318,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004918412134547976,
      "loss": 5.4846,
      "mean_token_accuracy": 0.18403034955263137,
      "num_tokens": 19119480.0,
      "step": 8570
    },
    {
      "entropy": 5.588913059234619,
      "epoch": 0.9631043971471893,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004918304964114696,
      "loss": 5.49,
      "mean_token_accuracy": 0.18181291073560715,
      "num_tokens": 19131168.0,
      "step": 8575
    },
    {
      "entropy": 5.604631042480468,
      "epoch": 0.9636659740551469,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004918197724641747,
      "loss": 5.621,
      "mean_token_accuracy": 0.17726342529058456,
      "num_tokens": 19142454.0,
      "step": 8580
    },
    {
      "entropy": 5.597691631317138,
      "epoch": 0.9642275509631044,
      "grad_norm": 0.921875,
      "learning_rate": 0.000491809041613254,
      "loss": 5.4787,
      "mean_token_accuracy": 0.18039288967847825,
      "num_tokens": 19154682.0,
      "step": 8585
    },
    {
      "entropy": 5.572999572753906,
      "epoch": 0.964789127871062,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004917983038590495,
      "loss": 5.4083,
      "mean_token_accuracy": 0.17833466827869415,
      "num_tokens": 19165190.0,
      "step": 8590
    },
    {
      "entropy": 5.609723138809204,
      "epoch": 0.9653507047790195,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004917875592019028,
      "loss": 5.5429,
      "mean_token_accuracy": 0.18797778934240342,
      "num_tokens": 19176075.0,
      "step": 8595
    },
    {
      "entropy": 5.6036262035369875,
      "epoch": 0.9659122816869771,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004917768076421561,
      "loss": 5.5542,
      "mean_token_accuracy": 0.17775871157646178,
      "num_tokens": 19188450.0,
      "step": 8600
    },
    {
      "entropy": 5.532081127166748,
      "epoch": 0.9664738585949346,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004917660491801518,
      "loss": 5.4163,
      "mean_token_accuracy": 0.19002363234758377,
      "num_tokens": 19201254.0,
      "step": 8605
    },
    {
      "entropy": 5.505360412597656,
      "epoch": 0.9670354355028922,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004917552838162324,
      "loss": 5.4888,
      "mean_token_accuracy": 0.18102688789367677,
      "num_tokens": 19214062.0,
      "step": 8610
    },
    {
      "entropy": 5.557801866531372,
      "epoch": 0.9675970124108496,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004917445115507408,
      "loss": 5.4048,
      "mean_token_accuracy": 0.18907466828823088,
      "num_tokens": 19224844.0,
      "step": 8615
    },
    {
      "entropy": 5.545368671417236,
      "epoch": 0.9681585893188072,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004917337323840198,
      "loss": 5.5717,
      "mean_token_accuracy": 0.17443536072969437,
      "num_tokens": 19237347.0,
      "step": 8620
    },
    {
      "entropy": 5.599031209945679,
      "epoch": 0.9687201662267647,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004917229463164128,
      "loss": 5.4525,
      "mean_token_accuracy": 0.18267473876476287,
      "num_tokens": 19248743.0,
      "step": 8625
    },
    {
      "entropy": 5.597088861465454,
      "epoch": 0.9692817431347223,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004917121533482631,
      "loss": 5.4472,
      "mean_token_accuracy": 0.18550329953432082,
      "num_tokens": 19259904.0,
      "step": 8630
    },
    {
      "entropy": 5.488122892379761,
      "epoch": 0.9698433200426798,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004917013534799144,
      "loss": 5.3845,
      "mean_token_accuracy": 0.1810299724340439,
      "num_tokens": 19271388.0,
      "step": 8635
    },
    {
      "entropy": 5.54411039352417,
      "epoch": 0.9704048969506374,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004916905467117104,
      "loss": 5.5203,
      "mean_token_accuracy": 0.18260214924812318,
      "num_tokens": 19282790.0,
      "step": 8640
    },
    {
      "entropy": 5.609619045257569,
      "epoch": 0.9709664738585949,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004916797330439955,
      "loss": 5.4125,
      "mean_token_accuracy": 0.1887241557240486,
      "num_tokens": 19293389.0,
      "step": 8645
    },
    {
      "entropy": 5.559426498413086,
      "epoch": 0.9715280507665525,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004916689124771138,
      "loss": 5.4924,
      "mean_token_accuracy": 0.18119360953569413,
      "num_tokens": 19304593.0,
      "step": 8650
    },
    {
      "entropy": 5.577017593383789,
      "epoch": 0.97208962767451,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004916580850114099,
      "loss": 5.4134,
      "mean_token_accuracy": 0.18593296855688096,
      "num_tokens": 19317388.0,
      "step": 8655
    },
    {
      "entropy": 5.610171699523926,
      "epoch": 0.9726512045824676,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004916472506472286,
      "loss": 5.4759,
      "mean_token_accuracy": 0.184097957611084,
      "num_tokens": 19329228.0,
      "step": 8660
    },
    {
      "entropy": 5.570147609710693,
      "epoch": 0.9732127814904251,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004916364093849147,
      "loss": 5.439,
      "mean_token_accuracy": 0.18606530874967575,
      "num_tokens": 19339421.0,
      "step": 8665
    },
    {
      "entropy": 5.468735218048096,
      "epoch": 0.9737743583983827,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004916255612248136,
      "loss": 5.3461,
      "mean_token_accuracy": 0.18966307491064072,
      "num_tokens": 19349618.0,
      "step": 8670
    },
    {
      "entropy": 5.42205400466919,
      "epoch": 0.9743359353063402,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004916147061672706,
      "loss": 5.446,
      "mean_token_accuracy": 0.18687217086553573,
      "num_tokens": 19361324.0,
      "step": 8675
    },
    {
      "entropy": 5.5919225215911865,
      "epoch": 0.9748975122142978,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004916038442126314,
      "loss": 5.4594,
      "mean_token_accuracy": 0.18415236473083496,
      "num_tokens": 19373631.0,
      "step": 8680
    },
    {
      "entropy": 5.509151649475098,
      "epoch": 0.9754590891222553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004915929753612416,
      "loss": 5.3967,
      "mean_token_accuracy": 0.1740434393286705,
      "num_tokens": 19384682.0,
      "step": 8685
    },
    {
      "entropy": 5.621597242355347,
      "epoch": 0.9760206660302129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004915820996134477,
      "loss": 5.5324,
      "mean_token_accuracy": 0.17876951843500138,
      "num_tokens": 19396344.0,
      "step": 8690
    },
    {
      "entropy": 5.5938441276550295,
      "epoch": 0.9765822429381704,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004915712169695956,
      "loss": 5.3791,
      "mean_token_accuracy": 0.18256505727767944,
      "num_tokens": 19406921.0,
      "step": 8695
    },
    {
      "entropy": 5.543767309188842,
      "epoch": 0.977143819846128,
      "grad_norm": 1.015625,
      "learning_rate": 0.000491560327430032,
      "loss": 5.5064,
      "mean_token_accuracy": 0.1808873161673546,
      "num_tokens": 19416851.0,
      "step": 8700
    },
    {
      "entropy": 5.577412939071655,
      "epoch": 0.9777053967540855,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004915494309951035,
      "loss": 5.4647,
      "mean_token_accuracy": 0.1804933190345764,
      "num_tokens": 19428180.0,
      "step": 8705
    },
    {
      "entropy": 5.691901683807373,
      "epoch": 0.9782669736620431,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004915385276651573,
      "loss": 5.5244,
      "mean_token_accuracy": 0.1787811204791069,
      "num_tokens": 19439575.0,
      "step": 8710
    },
    {
      "entropy": 5.61657485961914,
      "epoch": 0.9788285505700005,
      "grad_norm": 1.0,
      "learning_rate": 0.0004915276174405403,
      "loss": 5.4937,
      "mean_token_accuracy": 0.18056508004665375,
      "num_tokens": 19452074.0,
      "step": 8715
    },
    {
      "entropy": 5.4843040943145756,
      "epoch": 0.9793901274779581,
      "grad_norm": 1.0,
      "learning_rate": 0.0004915167003215999,
      "loss": 5.423,
      "mean_token_accuracy": 0.18719545304775237,
      "num_tokens": 19463085.0,
      "step": 8720
    },
    {
      "entropy": 5.699585342407227,
      "epoch": 0.9799517043859156,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004915057763086838,
      "loss": 5.5769,
      "mean_token_accuracy": 0.17127641886472703,
      "num_tokens": 19472726.0,
      "step": 8725
    },
    {
      "entropy": 5.559404087066651,
      "epoch": 0.9805132812938732,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049149484540214,
      "loss": 5.4047,
      "mean_token_accuracy": 0.18202415704727173,
      "num_tokens": 19482275.0,
      "step": 8730
    },
    {
      "entropy": 5.5444048881530765,
      "epoch": 0.9810748582018307,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004914839076023161,
      "loss": 5.5032,
      "mean_token_accuracy": 0.18604240268468858,
      "num_tokens": 19492782.0,
      "step": 8735
    },
    {
      "entropy": 5.496974086761474,
      "epoch": 0.9816364351097883,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004914729629095609,
      "loss": 5.3922,
      "mean_token_accuracy": 0.18256796002388,
      "num_tokens": 19505709.0,
      "step": 8740
    },
    {
      "entropy": 5.586014890670777,
      "epoch": 0.9821980120177458,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004914620113242224,
      "loss": 5.4607,
      "mean_token_accuracy": 0.1845390111207962,
      "num_tokens": 19516269.0,
      "step": 8745
    },
    {
      "entropy": 5.560197639465332,
      "epoch": 0.9827595889257034,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004914510528466495,
      "loss": 5.4564,
      "mean_token_accuracy": 0.18129220306873323,
      "num_tokens": 19526786.0,
      "step": 8750
    },
    {
      "entropy": 5.616420602798462,
      "epoch": 0.9833211658336609,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004914400874771913,
      "loss": 5.4868,
      "mean_token_accuracy": 0.17830210477113723,
      "num_tokens": 19537865.0,
      "step": 8755
    },
    {
      "entropy": 5.50406551361084,
      "epoch": 0.9838827427416185,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004914291152161967,
      "loss": 5.4231,
      "mean_token_accuracy": 0.19029248505830765,
      "num_tokens": 19548320.0,
      "step": 8760
    },
    {
      "entropy": 5.565492105484009,
      "epoch": 0.984444319649576,
      "grad_norm": 1.078125,
      "learning_rate": 0.000491418136064015,
      "loss": 5.4142,
      "mean_token_accuracy": 0.18464195132255554,
      "num_tokens": 19558856.0,
      "step": 8765
    },
    {
      "entropy": 5.56738600730896,
      "epoch": 0.9850058965575336,
      "grad_norm": 1.078125,
      "learning_rate": 0.000491407150020996,
      "loss": 5.5146,
      "mean_token_accuracy": 0.1884256422519684,
      "num_tokens": 19570792.0,
      "step": 8770
    },
    {
      "entropy": 5.599941253662109,
      "epoch": 0.9855674734654911,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004913961570874895,
      "loss": 5.4351,
      "mean_token_accuracy": 0.18806307315826415,
      "num_tokens": 19581982.0,
      "step": 8775
    },
    {
      "entropy": 5.512841844558716,
      "epoch": 0.9861290503734487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004913851572638453,
      "loss": 5.3796,
      "mean_token_accuracy": 0.1856641262769699,
      "num_tokens": 19592882.0,
      "step": 8780
    },
    {
      "entropy": 5.651014518737793,
      "epoch": 0.9866906272814062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004913741505504138,
      "loss": 5.5993,
      "mean_token_accuracy": 0.17879325300455093,
      "num_tokens": 19605967.0,
      "step": 8785
    },
    {
      "entropy": 5.645204353332519,
      "epoch": 0.9872522041893638,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004913631369475454,
      "loss": 5.4488,
      "mean_token_accuracy": 0.18693710565567018,
      "num_tokens": 19616494.0,
      "step": 8790
    },
    {
      "entropy": 5.582544946670533,
      "epoch": 0.9878137810973213,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004913521164555909,
      "loss": 5.5433,
      "mean_token_accuracy": 0.1812344253063202,
      "num_tokens": 19627013.0,
      "step": 8795
    },
    {
      "entropy": 5.5240236759185795,
      "epoch": 0.9883753580052789,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000491341089074901,
      "loss": 5.4555,
      "mean_token_accuracy": 0.18922722935676575,
      "num_tokens": 19639099.0,
      "step": 8800
    },
    {
      "entropy": 5.535636138916016,
      "epoch": 0.9889369349132364,
      "grad_norm": 1.046875,
      "learning_rate": 0.000491330054805827,
      "loss": 5.4266,
      "mean_token_accuracy": 0.18756327778100967,
      "num_tokens": 19650650.0,
      "step": 8805
    },
    {
      "entropy": 5.602877569198609,
      "epoch": 0.9894985118211939,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004913190136487201,
      "loss": 5.4585,
      "mean_token_accuracy": 0.18385438323020936,
      "num_tokens": 19662869.0,
      "step": 8810
    },
    {
      "entropy": 5.521768283843994,
      "epoch": 0.9900600887291514,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004913079656039318,
      "loss": 5.4634,
      "mean_token_accuracy": 0.18303659409284592,
      "num_tokens": 19673232.0,
      "step": 8815
    },
    {
      "entropy": 5.587317323684692,
      "epoch": 0.990621665637109,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004912969106718142,
      "loss": 5.4813,
      "mean_token_accuracy": 0.1822674974799156,
      "num_tokens": 19685069.0,
      "step": 8820
    },
    {
      "entropy": 5.6474627494812015,
      "epoch": 0.9911832425450665,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000491285848852719,
      "loss": 5.5704,
      "mean_token_accuracy": 0.17589719742536544,
      "num_tokens": 19697018.0,
      "step": 8825
    },
    {
      "entropy": 5.591678476333618,
      "epoch": 0.9917448194530241,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004912747801469984,
      "loss": 5.4193,
      "mean_token_accuracy": 0.1865503191947937,
      "num_tokens": 19707922.0,
      "step": 8830
    },
    {
      "entropy": 5.410496044158935,
      "epoch": 0.9923063963609816,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000491263704555005,
      "loss": 5.37,
      "mean_token_accuracy": 0.18731970489025115,
      "num_tokens": 19718693.0,
      "step": 8835
    },
    {
      "entropy": 5.526531171798706,
      "epoch": 0.9928679732689392,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004912526220770913,
      "loss": 5.4524,
      "mean_token_accuracy": 0.17944142073392869,
      "num_tokens": 19730846.0,
      "step": 8840
    },
    {
      "entropy": 5.548200798034668,
      "epoch": 0.9934295501768967,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004912415327136103,
      "loss": 5.4637,
      "mean_token_accuracy": 0.18399159759283065,
      "num_tokens": 19742099.0,
      "step": 8845
    },
    {
      "entropy": 5.576262474060059,
      "epoch": 0.9939911270848543,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000491230436464915,
      "loss": 5.4773,
      "mean_token_accuracy": 0.18230542540550232,
      "num_tokens": 19753216.0,
      "step": 8850
    },
    {
      "entropy": 5.6016761302948,
      "epoch": 0.9945527039928118,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004912193333313589,
      "loss": 5.4088,
      "mean_token_accuracy": 0.18425570875406266,
      "num_tokens": 19765316.0,
      "step": 8855
    },
    {
      "entropy": 5.562940406799316,
      "epoch": 0.9951142809007694,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004912082233132951,
      "loss": 5.4337,
      "mean_token_accuracy": 0.18214668035507203,
      "num_tokens": 19776208.0,
      "step": 8860
    },
    {
      "entropy": 5.593229818344116,
      "epoch": 0.9956758578087269,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004911971064110778,
      "loss": 5.4778,
      "mean_token_accuracy": 0.1849060982465744,
      "num_tokens": 19788065.0,
      "step": 8865
    },
    {
      "entropy": 5.582309246063232,
      "epoch": 0.9962374347166845,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004911859826250607,
      "loss": 5.3996,
      "mean_token_accuracy": 0.18344223350286484,
      "num_tokens": 19799388.0,
      "step": 8870
    },
    {
      "entropy": 5.531808280944825,
      "epoch": 0.996799011624642,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004911748519555981,
      "loss": 5.491,
      "mean_token_accuracy": 0.17774271667003633,
      "num_tokens": 19809933.0,
      "step": 8875
    },
    {
      "entropy": 5.685315370559692,
      "epoch": 0.9973605885325996,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004911637144030444,
      "loss": 5.5734,
      "mean_token_accuracy": 0.1801734372973442,
      "num_tokens": 19821566.0,
      "step": 8880
    },
    {
      "entropy": 5.6796534061431885,
      "epoch": 0.9979221654405571,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004911525699677541,
      "loss": 5.525,
      "mean_token_accuracy": 0.18233054876327515,
      "num_tokens": 19833727.0,
      "step": 8885
    },
    {
      "entropy": 5.610952854156494,
      "epoch": 0.9984837423485147,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004911414186500822,
      "loss": 5.5418,
      "mean_token_accuracy": 0.1787643238902092,
      "num_tokens": 19845176.0,
      "step": 8890
    },
    {
      "entropy": 5.476277828216553,
      "epoch": 0.9990453192564722,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004911302604503837,
      "loss": 5.4456,
      "mean_token_accuracy": 0.17728385776281358,
      "num_tokens": 19857528.0,
      "step": 8895
    },
    {
      "entropy": 5.590266132354737,
      "epoch": 0.9996068961644298,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004911190953690139,
      "loss": 5.373,
      "mean_token_accuracy": 0.18531470745801926,
      "num_tokens": 19868232.0,
      "step": 8900
    },
    {
      "entropy": 5.526914808485243,
      "epoch": 1.0001123153815914,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004911079234063284,
      "loss": 5.3123,
      "mean_token_accuracy": 0.18969544933901894,
      "num_tokens": 19877881.0,
      "step": 8905
    },
    {
      "entropy": 5.416088581085205,
      "epoch": 1.000673892289549,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004910967445626826,
      "loss": 5.4079,
      "mean_token_accuracy": 0.19263606518507004,
      "num_tokens": 19888398.0,
      "step": 8910
    },
    {
      "entropy": 5.603803396224976,
      "epoch": 1.0012354691975065,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004910855588384327,
      "loss": 5.3823,
      "mean_token_accuracy": 0.18999042063951493,
      "num_tokens": 19899661.0,
      "step": 8915
    },
    {
      "entropy": 5.565470886230469,
      "epoch": 1.0017970461054642,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004910743662339348,
      "loss": 5.4358,
      "mean_token_accuracy": 0.19043354392051698,
      "num_tokens": 19911073.0,
      "step": 8920
    },
    {
      "entropy": 5.462346839904785,
      "epoch": 1.0023586230134216,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004910631667495452,
      "loss": 5.3594,
      "mean_token_accuracy": 0.18880361914634705,
      "num_tokens": 19922567.0,
      "step": 8925
    },
    {
      "entropy": 5.537394285202026,
      "epoch": 1.0029201999213793,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004910519603856207,
      "loss": 5.4645,
      "mean_token_accuracy": 0.18664260804653168,
      "num_tokens": 19934547.0,
      "step": 8930
    },
    {
      "entropy": 5.555563068389892,
      "epoch": 1.0034817768293367,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004910407471425179,
      "loss": 5.4688,
      "mean_token_accuracy": 0.18146257624030113,
      "num_tokens": 19947659.0,
      "step": 8935
    },
    {
      "entropy": 5.588209009170532,
      "epoch": 1.0040433537372944,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000491029527020594,
      "loss": 5.3665,
      "mean_token_accuracy": 0.18237140774726868,
      "num_tokens": 19958150.0,
      "step": 8940
    },
    {
      "entropy": 5.49904727935791,
      "epoch": 1.0046049306452518,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004910183000202061,
      "loss": 5.361,
      "mean_token_accuracy": 0.18561754673719405,
      "num_tokens": 19968173.0,
      "step": 8945
    },
    {
      "entropy": 5.523112201690674,
      "epoch": 1.0051665075532095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004910070661417119,
      "loss": 5.4053,
      "mean_token_accuracy": 0.1836772307753563,
      "num_tokens": 19978543.0,
      "step": 8950
    },
    {
      "entropy": 5.5357386589050295,
      "epoch": 1.005728084461167,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004909958253854687,
      "loss": 5.358,
      "mean_token_accuracy": 0.18340061455965043,
      "num_tokens": 19989586.0,
      "step": 8955
    },
    {
      "entropy": 5.521802759170532,
      "epoch": 1.0062896613691246,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004909845777518347,
      "loss": 5.3736,
      "mean_token_accuracy": 0.1864837720990181,
      "num_tokens": 20000524.0,
      "step": 8960
    },
    {
      "entropy": 5.5667784214019775,
      "epoch": 1.006851238277082,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004909733232411681,
      "loss": 5.494,
      "mean_token_accuracy": 0.18308472037315368,
      "num_tokens": 20012935.0,
      "step": 8965
    },
    {
      "entropy": 5.56749906539917,
      "epoch": 1.0074128151850397,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004909620618538271,
      "loss": 5.46,
      "mean_token_accuracy": 0.18224538564682008,
      "num_tokens": 20024435.0,
      "step": 8970
    },
    {
      "entropy": 5.587008428573609,
      "epoch": 1.007974392092997,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004909507935901702,
      "loss": 5.4596,
      "mean_token_accuracy": 0.17887090891599655,
      "num_tokens": 20036017.0,
      "step": 8975
    },
    {
      "entropy": 5.5421875,
      "epoch": 1.0085359690009548,
      "grad_norm": 1.125,
      "learning_rate": 0.0004909395184505564,
      "loss": 5.441,
      "mean_token_accuracy": 0.1802113637328148,
      "num_tokens": 20047216.0,
      "step": 8980
    },
    {
      "entropy": 5.53623366355896,
      "epoch": 1.0090975459089122,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004909282364353445,
      "loss": 5.3309,
      "mean_token_accuracy": 0.185406456887722,
      "num_tokens": 20057439.0,
      "step": 8985
    },
    {
      "entropy": 5.550770998001099,
      "epoch": 1.0096591228168699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004909169475448938,
      "loss": 5.4458,
      "mean_token_accuracy": 0.18372596949338912,
      "num_tokens": 20069602.0,
      "step": 8990
    },
    {
      "entropy": 5.505699682235718,
      "epoch": 1.0102206997248273,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004909056517795638,
      "loss": 5.396,
      "mean_token_accuracy": 0.18840807676315308,
      "num_tokens": 20081366.0,
      "step": 8995
    },
    {
      "entropy": 5.554901027679444,
      "epoch": 1.010782276632785,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004908943491397141,
      "loss": 5.3988,
      "mean_token_accuracy": 0.18910853266716005,
      "num_tokens": 20091384.0,
      "step": 9000
    },
    {
      "epoch": 1.010782276632785,
      "eval_entropy": 5.457474174898003,
      "eval_loss": 5.499125957489014,
      "eval_mean_token_accuracy": 0.18832171713534107,
      "eval_num_tokens": 20091384.0,
      "eval_runtime": 58.5199,
      "eval_samples_per_second": 529.956,
      "eval_steps_per_second": 66.251,
      "step": 9000
    },
    {
      "entropy": 5.616485166549682,
      "epoch": 1.0113438535407424,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004908830396257045,
      "loss": 5.4308,
      "mean_token_accuracy": 0.18055644035339355,
      "num_tokens": 20102953.0,
      "step": 9005
    },
    {
      "entropy": 5.5084339618682865,
      "epoch": 1.0119054304486998,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004908717232378953,
      "loss": 5.3845,
      "mean_token_accuracy": 0.1884632647037506,
      "num_tokens": 20113672.0,
      "step": 9010
    },
    {
      "entropy": 5.578459787368774,
      "epoch": 1.0124670073566575,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004908603999766467,
      "loss": 5.4048,
      "mean_token_accuracy": 0.186214779317379,
      "num_tokens": 20125143.0,
      "step": 9015
    },
    {
      "entropy": 5.5441014766693115,
      "epoch": 1.013028584264615,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004908490698423191,
      "loss": 5.4129,
      "mean_token_accuracy": 0.18566339910030366,
      "num_tokens": 20135489.0,
      "step": 9020
    },
    {
      "entropy": 5.513772201538086,
      "epoch": 1.0135901611725726,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004908377328352736,
      "loss": 5.4008,
      "mean_token_accuracy": 0.18757753372192382,
      "num_tokens": 20146768.0,
      "step": 9025
    },
    {
      "entropy": 5.329649972915649,
      "epoch": 1.01415173808053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004908263889558709,
      "loss": 5.2137,
      "mean_token_accuracy": 0.18804068714380265,
      "num_tokens": 20156894.0,
      "step": 9030
    },
    {
      "entropy": 5.542341899871826,
      "epoch": 1.0147133149884877,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004908150382044722,
      "loss": 5.4357,
      "mean_token_accuracy": 0.1877053663134575,
      "num_tokens": 20168174.0,
      "step": 9035
    },
    {
      "entropy": 5.622216033935547,
      "epoch": 1.0152748918964452,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000490803680581439,
      "loss": 5.5145,
      "mean_token_accuracy": 0.17867926359176636,
      "num_tokens": 20179962.0,
      "step": 9040
    },
    {
      "entropy": 5.598105525970459,
      "epoch": 1.0158364688044028,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000490792316087133,
      "loss": 5.4298,
      "mean_token_accuracy": 0.1849098801612854,
      "num_tokens": 20189842.0,
      "step": 9045
    },
    {
      "entropy": 5.5179485321044925,
      "epoch": 1.0163980457123603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004907809447219158,
      "loss": 5.3081,
      "mean_token_accuracy": 0.1819055676460266,
      "num_tokens": 20200471.0,
      "step": 9050
    },
    {
      "entropy": 5.528376197814941,
      "epoch": 1.016959622620318,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004907695664861498,
      "loss": 5.384,
      "mean_token_accuracy": 0.1852393552660942,
      "num_tokens": 20212145.0,
      "step": 9055
    },
    {
      "entropy": 5.502276039123535,
      "epoch": 1.0175211995282754,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004907581813801971,
      "loss": 5.3749,
      "mean_token_accuracy": 0.18804652988910675,
      "num_tokens": 20222454.0,
      "step": 9060
    },
    {
      "entropy": 5.597353744506836,
      "epoch": 1.018082776436233,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004907467894044202,
      "loss": 5.5113,
      "mean_token_accuracy": 0.1800994336605072,
      "num_tokens": 20233402.0,
      "step": 9065
    },
    {
      "entropy": 5.487265300750733,
      "epoch": 1.0186443533441905,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004907353905591817,
      "loss": 5.3143,
      "mean_token_accuracy": 0.19445917904376983,
      "num_tokens": 20244102.0,
      "step": 9070
    },
    {
      "entropy": 5.594339370727539,
      "epoch": 1.0192059302521481,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000490723984844845,
      "loss": 5.5201,
      "mean_token_accuracy": 0.18200081139802932,
      "num_tokens": 20255003.0,
      "step": 9075
    },
    {
      "entropy": 5.551009750366211,
      "epoch": 1.0197675071601056,
      "grad_norm": 1.125,
      "learning_rate": 0.0004907125722617727,
      "loss": 5.3524,
      "mean_token_accuracy": 0.18559956699609756,
      "num_tokens": 20265211.0,
      "step": 9080
    },
    {
      "entropy": 5.475327205657959,
      "epoch": 1.0203290840680632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004907011528103285,
      "loss": 5.3241,
      "mean_token_accuracy": 0.18909415304660798,
      "num_tokens": 20276552.0,
      "step": 9085
    },
    {
      "entropy": 5.559994316101074,
      "epoch": 1.0208906609760207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004906897264908759,
      "loss": 5.4538,
      "mean_token_accuracy": 0.18224422484636307,
      "num_tokens": 20286805.0,
      "step": 9090
    },
    {
      "entropy": 5.557529258728027,
      "epoch": 1.0214522378839783,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004906782933037789,
      "loss": 5.3922,
      "mean_token_accuracy": 0.18283914625644684,
      "num_tokens": 20298536.0,
      "step": 9095
    },
    {
      "entropy": 5.530365753173828,
      "epoch": 1.0220138147919358,
      "grad_norm": 1.125,
      "learning_rate": 0.0004906668532494012,
      "loss": 5.4019,
      "mean_token_accuracy": 0.18648336976766586,
      "num_tokens": 20309842.0,
      "step": 9100
    },
    {
      "entropy": 5.517857694625855,
      "epoch": 1.0225753916998932,
      "grad_norm": 1.125,
      "learning_rate": 0.0004906554063281074,
      "loss": 5.3217,
      "mean_token_accuracy": 0.1901137039065361,
      "num_tokens": 20320413.0,
      "step": 9105
    },
    {
      "entropy": 5.549293136596679,
      "epoch": 1.0231369686078509,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004906439525402618,
      "loss": 5.3748,
      "mean_token_accuracy": 0.1853329733014107,
      "num_tokens": 20331812.0,
      "step": 9110
    },
    {
      "entropy": 5.475205707550049,
      "epoch": 1.0236985455158083,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004906324918862291,
      "loss": 5.3204,
      "mean_token_accuracy": 0.1900770142674446,
      "num_tokens": 20342148.0,
      "step": 9115
    },
    {
      "entropy": 5.516740608215332,
      "epoch": 1.024260122423766,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004906210243663742,
      "loss": 5.4226,
      "mean_token_accuracy": 0.18869406282901763,
      "num_tokens": 20353843.0,
      "step": 9120
    },
    {
      "entropy": 5.548943662643433,
      "epoch": 1.0248216993317234,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004906095499810622,
      "loss": 5.4208,
      "mean_token_accuracy": 0.18616133481264113,
      "num_tokens": 20364846.0,
      "step": 9125
    },
    {
      "entropy": 5.565707874298096,
      "epoch": 1.025383276239681,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004905980687306586,
      "loss": 5.3691,
      "mean_token_accuracy": 0.1888698384165764,
      "num_tokens": 20376874.0,
      "step": 9130
    },
    {
      "entropy": 5.49728045463562,
      "epoch": 1.0259448531476385,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004905865806155288,
      "loss": 5.3856,
      "mean_token_accuracy": 0.18706257939338683,
      "num_tokens": 20388189.0,
      "step": 9135
    },
    {
      "entropy": 5.536900901794434,
      "epoch": 1.0265064300555962,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004905750856360387,
      "loss": 5.4226,
      "mean_token_accuracy": 0.18550562411546706,
      "num_tokens": 20399155.0,
      "step": 9140
    },
    {
      "entropy": 5.513206052780151,
      "epoch": 1.0270680069635536,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004905635837925543,
      "loss": 5.2928,
      "mean_token_accuracy": 0.1886839136481285,
      "num_tokens": 20410569.0,
      "step": 9145
    },
    {
      "entropy": 5.426887464523316,
      "epoch": 1.0276295838715113,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004905520750854418,
      "loss": 5.3469,
      "mean_token_accuracy": 0.18439428955316545,
      "num_tokens": 20420438.0,
      "step": 9150
    },
    {
      "entropy": 5.56753511428833,
      "epoch": 1.0281911607794687,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004905405595150674,
      "loss": 5.3972,
      "mean_token_accuracy": 0.1855535164475441,
      "num_tokens": 20431874.0,
      "step": 9155
    },
    {
      "entropy": 5.522236251831055,
      "epoch": 1.0287527376874264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004905290370817982,
      "loss": 5.3824,
      "mean_token_accuracy": 0.18540708273649215,
      "num_tokens": 20442792.0,
      "step": 9160
    },
    {
      "entropy": 5.567186307907105,
      "epoch": 1.0293143145953838,
      "grad_norm": 1.0625,
      "learning_rate": 0.000490517507786001,
      "loss": 5.3997,
      "mean_token_accuracy": 0.18406621366739273,
      "num_tokens": 20454410.0,
      "step": 9165
    },
    {
      "entropy": 5.655465030670166,
      "epoch": 1.0298758915033415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004905059716280425,
      "loss": 5.5354,
      "mean_token_accuracy": 0.1752452775835991,
      "num_tokens": 20467072.0,
      "step": 9170
    },
    {
      "entropy": 5.57279372215271,
      "epoch": 1.030437468411299,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004904944286082904,
      "loss": 5.4558,
      "mean_token_accuracy": 0.18608310520648957,
      "num_tokens": 20477697.0,
      "step": 9175
    },
    {
      "entropy": 5.52084174156189,
      "epoch": 1.0309990453192566,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004904828787271121,
      "loss": 5.2931,
      "mean_token_accuracy": 0.1903368964791298,
      "num_tokens": 20489024.0,
      "step": 9180
    },
    {
      "entropy": 5.500179243087769,
      "epoch": 1.031560622227214,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004904713219848753,
      "loss": 5.3832,
      "mean_token_accuracy": 0.18683810234069825,
      "num_tokens": 20499928.0,
      "step": 9185
    },
    {
      "entropy": 5.44192795753479,
      "epoch": 1.0321221991351717,
      "grad_norm": 1.125,
      "learning_rate": 0.0004904597583819481,
      "loss": 5.4284,
      "mean_token_accuracy": 0.1826045975089073,
      "num_tokens": 20511085.0,
      "step": 9190
    },
    {
      "entropy": 5.55776572227478,
      "epoch": 1.032683776043129,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004904481879186987,
      "loss": 5.2971,
      "mean_token_accuracy": 0.18520693182945253,
      "num_tokens": 20522164.0,
      "step": 9195
    },
    {
      "entropy": 5.586649608612061,
      "epoch": 1.0332453529510865,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004904366105954954,
      "loss": 5.4627,
      "mean_token_accuracy": 0.186662457883358,
      "num_tokens": 20534187.0,
      "step": 9200
    },
    {
      "entropy": 5.59186577796936,
      "epoch": 1.0338069298590442,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004904250264127067,
      "loss": 5.4379,
      "mean_token_accuracy": 0.1879410594701767,
      "num_tokens": 20545819.0,
      "step": 9205
    },
    {
      "entropy": 5.610886096954346,
      "epoch": 1.0343685067670016,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004904134353707018,
      "loss": 5.449,
      "mean_token_accuracy": 0.17868003100156785,
      "num_tokens": 20556803.0,
      "step": 9210
    },
    {
      "entropy": 5.562782859802246,
      "epoch": 1.0349300836749593,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004904018374698496,
      "loss": 5.4392,
      "mean_token_accuracy": 0.17891971468925477,
      "num_tokens": 20567366.0,
      "step": 9215
    },
    {
      "entropy": 5.5469132423400875,
      "epoch": 1.0354916605829168,
      "grad_norm": 1.0,
      "learning_rate": 0.0004903902327105191,
      "loss": 5.4597,
      "mean_token_accuracy": 0.1822926864027977,
      "num_tokens": 20579182.0,
      "step": 9220
    },
    {
      "entropy": 5.515037584304809,
      "epoch": 1.0360532374908744,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004903786210930803,
      "loss": 5.3174,
      "mean_token_accuracy": 0.18531483113765718,
      "num_tokens": 20590346.0,
      "step": 9225
    },
    {
      "entropy": 5.52108678817749,
      "epoch": 1.0366148143988319,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004903670026179025,
      "loss": 5.4186,
      "mean_token_accuracy": 0.18376436084508896,
      "num_tokens": 20601279.0,
      "step": 9230
    },
    {
      "entropy": 5.487068128585816,
      "epoch": 1.0371763913067895,
      "grad_norm": 1.21875,
      "learning_rate": 0.000490355377285356,
      "loss": 5.3454,
      "mean_token_accuracy": 0.18688741326332092,
      "num_tokens": 20612163.0,
      "step": 9235
    },
    {
      "entropy": 5.550949001312256,
      "epoch": 1.037737968214747,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004903437450958106,
      "loss": 5.3069,
      "mean_token_accuracy": 0.1903759852051735,
      "num_tokens": 20622587.0,
      "step": 9240
    },
    {
      "entropy": 5.528546476364136,
      "epoch": 1.0382995451227046,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004903321060496369,
      "loss": 5.4755,
      "mean_token_accuracy": 0.17715801149606705,
      "num_tokens": 20633341.0,
      "step": 9245
    },
    {
      "entropy": 5.494615077972412,
      "epoch": 1.038861122030662,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004903204601472054,
      "loss": 5.3492,
      "mean_token_accuracy": 0.18869257122278213,
      "num_tokens": 20644582.0,
      "step": 9250
    },
    {
      "entropy": 5.527494478225708,
      "epoch": 1.0394226989386197,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004903088073888869,
      "loss": 5.3163,
      "mean_token_accuracy": 0.1975536048412323,
      "num_tokens": 20654936.0,
      "step": 9255
    },
    {
      "entropy": 5.434938859939575,
      "epoch": 1.0399842758465772,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004902971477750526,
      "loss": 5.3381,
      "mean_token_accuracy": 0.1842801809310913,
      "num_tokens": 20666348.0,
      "step": 9260
    },
    {
      "entropy": 5.5995471477508545,
      "epoch": 1.0405458527545348,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004902854813060736,
      "loss": 5.5141,
      "mean_token_accuracy": 0.18195110112428664,
      "num_tokens": 20677482.0,
      "step": 9265
    },
    {
      "entropy": 5.579679870605469,
      "epoch": 1.0411074296624923,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004902738079823213,
      "loss": 5.425,
      "mean_token_accuracy": 0.19345418363809586,
      "num_tokens": 20688369.0,
      "step": 9270
    },
    {
      "entropy": 5.517041492462158,
      "epoch": 1.04166900657045,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902621278041674,
      "loss": 5.437,
      "mean_token_accuracy": 0.18158860951662065,
      "num_tokens": 20698933.0,
      "step": 9275
    },
    {
      "entropy": 5.515263271331787,
      "epoch": 1.0422305834784074,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004902504407719839,
      "loss": 5.3057,
      "mean_token_accuracy": 0.1855906665325165,
      "num_tokens": 20710192.0,
      "step": 9280
    },
    {
      "entropy": 5.498098850250244,
      "epoch": 1.042792160386365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004902387468861429,
      "loss": 5.362,
      "mean_token_accuracy": 0.1852239578962326,
      "num_tokens": 20721408.0,
      "step": 9285
    },
    {
      "entropy": 5.419748306274414,
      "epoch": 1.0433537372943225,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004902270461470169,
      "loss": 5.272,
      "mean_token_accuracy": 0.19246668368577957,
      "num_tokens": 20732155.0,
      "step": 9290
    },
    {
      "entropy": 5.4803431034088135,
      "epoch": 1.04391531420228,
      "grad_norm": 1.09375,
      "learning_rate": 0.000490215338554978,
      "loss": 5.2751,
      "mean_token_accuracy": 0.19139215499162673,
      "num_tokens": 20741819.0,
      "step": 9295
    },
    {
      "entropy": 5.537647771835327,
      "epoch": 1.0444768911102376,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004902036241103994,
      "loss": 5.4062,
      "mean_token_accuracy": 0.18556405305862428,
      "num_tokens": 20754931.0,
      "step": 9300
    },
    {
      "entropy": 5.5981498718261715,
      "epoch": 1.045038468018195,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004901919028136539,
      "loss": 5.4801,
      "mean_token_accuracy": 0.1823945239186287,
      "num_tokens": 20766533.0,
      "step": 9305
    },
    {
      "entropy": 5.502779340744018,
      "epoch": 1.0456000449261527,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004901801746651147,
      "loss": 5.346,
      "mean_token_accuracy": 0.19028619229793547,
      "num_tokens": 20777055.0,
      "step": 9310
    },
    {
      "entropy": 5.418141555786133,
      "epoch": 1.04616162183411,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004901684396651554,
      "loss": 5.2939,
      "mean_token_accuracy": 0.18705033957958223,
      "num_tokens": 20786871.0,
      "step": 9315
    },
    {
      "entropy": 5.565149259567261,
      "epoch": 1.0467231987420678,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004901566978141494,
      "loss": 5.4456,
      "mean_token_accuracy": 0.182110396027565,
      "num_tokens": 20798546.0,
      "step": 9320
    },
    {
      "entropy": 5.530739164352417,
      "epoch": 1.0472847756500252,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004901449491124708,
      "loss": 5.38,
      "mean_token_accuracy": 0.18562474697828293,
      "num_tokens": 20809657.0,
      "step": 9325
    },
    {
      "entropy": 5.46672682762146,
      "epoch": 1.0478463525579829,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004901331935604936,
      "loss": 5.3978,
      "mean_token_accuracy": 0.18574664890766143,
      "num_tokens": 20819662.0,
      "step": 9330
    },
    {
      "entropy": 5.450799417495728,
      "epoch": 1.0484079294659403,
      "grad_norm": 1.046875,
      "learning_rate": 0.000490121431158592,
      "loss": 5.2663,
      "mean_token_accuracy": 0.19446823745965958,
      "num_tokens": 20830554.0,
      "step": 9335
    },
    {
      "entropy": 5.571157646179199,
      "epoch": 1.048969506373898,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004901096619071407,
      "loss": 5.4337,
      "mean_token_accuracy": 0.1810958907008171,
      "num_tokens": 20842003.0,
      "step": 9340
    },
    {
      "entropy": 5.559390020370484,
      "epoch": 1.0495310832818554,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004900978858065142,
      "loss": 5.4178,
      "mean_token_accuracy": 0.1890198290348053,
      "num_tokens": 20853453.0,
      "step": 9345
    },
    {
      "entropy": 5.6493274688720705,
      "epoch": 1.050092660189813,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004900861028570877,
      "loss": 5.4767,
      "mean_token_accuracy": 0.1816459745168686,
      "num_tokens": 20867296.0,
      "step": 9350
    },
    {
      "entropy": 5.473407840728759,
      "epoch": 1.0506542370977705,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004900743130592362,
      "loss": 5.3723,
      "mean_token_accuracy": 0.18959630727767945,
      "num_tokens": 20878951.0,
      "step": 9355
    },
    {
      "entropy": 5.556913042068482,
      "epoch": 1.0512158140057282,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004900625164133351,
      "loss": 5.4364,
      "mean_token_accuracy": 0.18974503874778748,
      "num_tokens": 20891103.0,
      "step": 9360
    },
    {
      "entropy": 5.632567405700684,
      "epoch": 1.0517773909136856,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900507129197601,
      "loss": 5.5273,
      "mean_token_accuracy": 0.1795767903327942,
      "num_tokens": 20903570.0,
      "step": 9365
    },
    {
      "entropy": 5.604644870758056,
      "epoch": 1.0523389678216433,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900389025788871,
      "loss": 5.4774,
      "mean_token_accuracy": 0.182725690305233,
      "num_tokens": 20914156.0,
      "step": 9370
    },
    {
      "entropy": 5.50830864906311,
      "epoch": 1.0529005447296007,
      "grad_norm": 0.9375,
      "learning_rate": 0.000490027085391092,
      "loss": 5.3634,
      "mean_token_accuracy": 0.18633141964673997,
      "num_tokens": 20925192.0,
      "step": 9375
    },
    {
      "entropy": 5.559418725967407,
      "epoch": 1.0534621216375584,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000490015261356751,
      "loss": 5.2579,
      "mean_token_accuracy": 0.20030733942985535,
      "num_tokens": 20935281.0,
      "step": 9380
    },
    {
      "entropy": 5.603970336914062,
      "epoch": 1.0540236985455158,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004900034304762408,
      "loss": 5.4817,
      "mean_token_accuracy": 0.18674329668283463,
      "num_tokens": 20947397.0,
      "step": 9385
    },
    {
      "entropy": 5.462509298324585,
      "epoch": 1.0545852754534732,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489991592749938,
      "loss": 5.3527,
      "mean_token_accuracy": 0.18962117433547973,
      "num_tokens": 20958473.0,
      "step": 9390
    },
    {
      "entropy": 5.624472379684448,
      "epoch": 1.055146852361431,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004899797481782193,
      "loss": 5.4511,
      "mean_token_accuracy": 0.17956508100032806,
      "num_tokens": 20970112.0,
      "step": 9395
    },
    {
      "entropy": 5.5020238876342775,
      "epoch": 1.0557084292693883,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004899678967614623,
      "loss": 5.3606,
      "mean_token_accuracy": 0.18722193837165832,
      "num_tokens": 20980762.0,
      "step": 9400
    },
    {
      "entropy": 5.46553521156311,
      "epoch": 1.056270006177346,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004899560385000439,
      "loss": 5.2688,
      "mean_token_accuracy": 0.19204796999692916,
      "num_tokens": 20991663.0,
      "step": 9405
    },
    {
      "entropy": 5.472022581100464,
      "epoch": 1.0568315830853034,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000489944173394342,
      "loss": 5.2958,
      "mean_token_accuracy": 0.19311801642179488,
      "num_tokens": 21002121.0,
      "step": 9410
    },
    {
      "entropy": 5.4545652866363525,
      "epoch": 1.0573931599932611,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004899323014447342,
      "loss": 5.3564,
      "mean_token_accuracy": 0.1815158247947693,
      "num_tokens": 21013791.0,
      "step": 9415
    },
    {
      "entropy": 5.448367071151734,
      "epoch": 1.0579547369012186,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004899204226515985,
      "loss": 5.3091,
      "mean_token_accuracy": 0.18644743114709855,
      "num_tokens": 21022977.0,
      "step": 9420
    },
    {
      "entropy": 5.521354913711548,
      "epoch": 1.0585163138091762,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004899085370153133,
      "loss": 5.4551,
      "mean_token_accuracy": 0.18517175167798997,
      "num_tokens": 21036372.0,
      "step": 9425
    },
    {
      "entropy": 5.680944919586182,
      "epoch": 1.0590778907171337,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004898966445362569,
      "loss": 5.5038,
      "mean_token_accuracy": 0.17759744226932525,
      "num_tokens": 21047984.0,
      "step": 9430
    },
    {
      "entropy": 5.606042003631591,
      "epoch": 1.0596394676250913,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004898847452148079,
      "loss": 5.3977,
      "mean_token_accuracy": 0.18267227858304977,
      "num_tokens": 21060127.0,
      "step": 9435
    },
    {
      "entropy": 5.482920551300049,
      "epoch": 1.0602010445330488,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004898728390513454,
      "loss": 5.3766,
      "mean_token_accuracy": 0.18611850142478942,
      "num_tokens": 21070659.0,
      "step": 9440
    },
    {
      "entropy": 5.472131824493408,
      "epoch": 1.0607626214410064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004898609260462483,
      "loss": 5.3223,
      "mean_token_accuracy": 0.19197576344013215,
      "num_tokens": 21081667.0,
      "step": 9445
    },
    {
      "entropy": 5.533467817306518,
      "epoch": 1.0613241983489639,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004898490061998961,
      "loss": 5.3432,
      "mean_token_accuracy": 0.18749047070741653,
      "num_tokens": 21094275.0,
      "step": 9450
    },
    {
      "entropy": 5.549678325653076,
      "epoch": 1.0618857752569215,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004898370795126681,
      "loss": 5.4108,
      "mean_token_accuracy": 0.18219694793224334,
      "num_tokens": 21106360.0,
      "step": 9455
    },
    {
      "entropy": 5.505332660675049,
      "epoch": 1.062447352164879,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004898251459849441,
      "loss": 5.3758,
      "mean_token_accuracy": 0.1860309511423111,
      "num_tokens": 21117647.0,
      "step": 9460
    },
    {
      "entropy": 5.3988227367401125,
      "epoch": 1.0630089290728366,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004898132056171041,
      "loss": 5.2327,
      "mean_token_accuracy": 0.19441557824611663,
      "num_tokens": 21129599.0,
      "step": 9465
    },
    {
      "entropy": 5.485315370559692,
      "epoch": 1.063570505980794,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004898012584095285,
      "loss": 5.3569,
      "mean_token_accuracy": 0.1858839511871338,
      "num_tokens": 21140573.0,
      "step": 9470
    },
    {
      "entropy": 5.580947637557983,
      "epoch": 1.0641320828887517,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004897893043625975,
      "loss": 5.4311,
      "mean_token_accuracy": 0.1874650239944458,
      "num_tokens": 21151899.0,
      "step": 9475
    },
    {
      "entropy": 5.602825069427491,
      "epoch": 1.0646936597967092,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004897773434766917,
      "loss": 5.4881,
      "mean_token_accuracy": 0.17654771059751512,
      "num_tokens": 21162748.0,
      "step": 9480
    },
    {
      "entropy": 5.550855731964111,
      "epoch": 1.0652552367046666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000489765375752192,
      "loss": 5.4042,
      "mean_token_accuracy": 0.183906626701355,
      "num_tokens": 21172962.0,
      "step": 9485
    },
    {
      "entropy": 5.54920072555542,
      "epoch": 1.0658168136126243,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004897534011894793,
      "loss": 5.3891,
      "mean_token_accuracy": 0.18949666619300842,
      "num_tokens": 21183805.0,
      "step": 9490
    },
    {
      "entropy": 5.438908433914184,
      "epoch": 1.0663783905205817,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004897414197889352,
      "loss": 5.2922,
      "mean_token_accuracy": 0.19069612473249437,
      "num_tokens": 21195129.0,
      "step": 9495
    },
    {
      "entropy": 5.4912261962890625,
      "epoch": 1.0669399674285394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000489729431550941,
      "loss": 5.4097,
      "mean_token_accuracy": 0.1807676836848259,
      "num_tokens": 21206300.0,
      "step": 9500
    },
    {
      "entropy": 5.565548753738403,
      "epoch": 1.0675015443364968,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004897174364758783,
      "loss": 5.3464,
      "mean_token_accuracy": 0.18868244737386702,
      "num_tokens": 21217137.0,
      "step": 9505
    },
    {
      "entropy": 5.481102037429809,
      "epoch": 1.0680631212444545,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004897054345641291,
      "loss": 5.3322,
      "mean_token_accuracy": 0.19163161218166352,
      "num_tokens": 21228399.0,
      "step": 9510
    },
    {
      "entropy": 5.528740978240966,
      "epoch": 1.068624698152412,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004896934258160757,
      "loss": 5.4411,
      "mean_token_accuracy": 0.18235985487699508,
      "num_tokens": 21239396.0,
      "step": 9515
    },
    {
      "entropy": 5.504907941818237,
      "epoch": 1.0691862750603696,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004896814102321004,
      "loss": 5.3433,
      "mean_token_accuracy": 0.1862892672419548,
      "num_tokens": 21249966.0,
      "step": 9520
    },
    {
      "entropy": 5.5240458965301515,
      "epoch": 1.069747851968327,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004896693878125857,
      "loss": 5.3671,
      "mean_token_accuracy": 0.18622422069311143,
      "num_tokens": 21260157.0,
      "step": 9525
    },
    {
      "entropy": 5.523571395874024,
      "epoch": 1.0703094288762847,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004896573585579144,
      "loss": 5.4258,
      "mean_token_accuracy": 0.18874689936637878,
      "num_tokens": 21270738.0,
      "step": 9530
    },
    {
      "entropy": 5.4886860847473145,
      "epoch": 1.070871005784242,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004896453224684695,
      "loss": 5.3561,
      "mean_token_accuracy": 0.18944532722234725,
      "num_tokens": 21282469.0,
      "step": 9535
    },
    {
      "entropy": 5.496334505081177,
      "epoch": 1.0714325826921998,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004896332795446345,
      "loss": 5.3063,
      "mean_token_accuracy": 0.1876911997795105,
      "num_tokens": 21293474.0,
      "step": 9540
    },
    {
      "entropy": 5.541016721725464,
      "epoch": 1.0719941596001572,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004896212297867924,
      "loss": 5.4524,
      "mean_token_accuracy": 0.1798918917775154,
      "num_tokens": 21304904.0,
      "step": 9545
    },
    {
      "entropy": 5.574141883850098,
      "epoch": 1.0725557365081149,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004896091731953271,
      "loss": 5.3766,
      "mean_token_accuracy": 0.18713437765836716,
      "num_tokens": 21315407.0,
      "step": 9550
    },
    {
      "entropy": 5.469040870666504,
      "epoch": 1.0731173134160723,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004895971097706227,
      "loss": 5.3148,
      "mean_token_accuracy": 0.19208430349826813,
      "num_tokens": 21326496.0,
      "step": 9555
    },
    {
      "entropy": 5.515277433395386,
      "epoch": 1.07367889032403,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004895850395130629,
      "loss": 5.4026,
      "mean_token_accuracy": 0.1864038661122322,
      "num_tokens": 21339449.0,
      "step": 9560
    },
    {
      "entropy": 5.527117300033569,
      "epoch": 1.0742404672319874,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004895729624230323,
      "loss": 5.4633,
      "mean_token_accuracy": 0.18114811927080154,
      "num_tokens": 21350431.0,
      "step": 9565
    },
    {
      "entropy": 5.497921133041382,
      "epoch": 1.074802044139945,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004895608785009153,
      "loss": 5.3559,
      "mean_token_accuracy": 0.18641359955072404,
      "num_tokens": 21361168.0,
      "step": 9570
    },
    {
      "entropy": 5.535551834106445,
      "epoch": 1.0753636210479025,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004895487877470969,
      "loss": 5.3458,
      "mean_token_accuracy": 0.1902617409825325,
      "num_tokens": 21372635.0,
      "step": 9575
    },
    {
      "entropy": 5.446909427642822,
      "epoch": 1.07592519795586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004895366901619617,
      "loss": 5.3517,
      "mean_token_accuracy": 0.18696731477975845,
      "num_tokens": 21384051.0,
      "step": 9580
    },
    {
      "entropy": 5.511615562438965,
      "epoch": 1.0764867748638176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004895245857458951,
      "loss": 5.3767,
      "mean_token_accuracy": 0.17953940927982331,
      "num_tokens": 21397496.0,
      "step": 9585
    },
    {
      "entropy": 5.510258388519287,
      "epoch": 1.077048351771775,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004895124744992825,
      "loss": 5.4354,
      "mean_token_accuracy": 0.1822570264339447,
      "num_tokens": 21409349.0,
      "step": 9590
    },
    {
      "entropy": 5.527426385879517,
      "epoch": 1.0776099286797327,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004895003564225095,
      "loss": 5.3954,
      "mean_token_accuracy": 0.1880100414156914,
      "num_tokens": 21421843.0,
      "step": 9595
    },
    {
      "entropy": 5.507824325561524,
      "epoch": 1.0781715055876901,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004894882315159621,
      "loss": 5.275,
      "mean_token_accuracy": 0.18905281722545625,
      "num_tokens": 21433230.0,
      "step": 9600
    },
    {
      "entropy": 5.541763591766357,
      "epoch": 1.0787330824956478,
      "grad_norm": 1.109375,
      "learning_rate": 0.000489476099780026,
      "loss": 5.412,
      "mean_token_accuracy": 0.1857568547129631,
      "num_tokens": 21444295.0,
      "step": 9605
    },
    {
      "entropy": 5.568008661270142,
      "epoch": 1.0792946594036052,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004894639612150879,
      "loss": 5.3545,
      "mean_token_accuracy": 0.1823578417301178,
      "num_tokens": 21455721.0,
      "step": 9610
    },
    {
      "entropy": 5.496934843063355,
      "epoch": 1.079856236311563,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004894518158215339,
      "loss": 5.3864,
      "mean_token_accuracy": 0.18858225643634796,
      "num_tokens": 21466514.0,
      "step": 9615
    },
    {
      "entropy": 5.519130754470825,
      "epoch": 1.0804178132195204,
      "grad_norm": 1.21875,
      "learning_rate": 0.000489439663599751,
      "loss": 5.3797,
      "mean_token_accuracy": 0.18560597747564317,
      "num_tokens": 21478782.0,
      "step": 9620
    },
    {
      "entropy": 5.6073750972747805,
      "epoch": 1.080979390127478,
      "grad_norm": 1.09375,
      "learning_rate": 0.000489427504550126,
      "loss": 5.3838,
      "mean_token_accuracy": 0.19413492530584336,
      "num_tokens": 21490428.0,
      "step": 9625
    },
    {
      "entropy": 5.39088807106018,
      "epoch": 1.0815409670354355,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004894153386730462,
      "loss": 5.2444,
      "mean_token_accuracy": 0.18894489258527755,
      "num_tokens": 21501530.0,
      "step": 9630
    },
    {
      "entropy": 5.502291059494018,
      "epoch": 1.0821025439433931,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004894031659688988,
      "loss": 5.3327,
      "mean_token_accuracy": 0.19191098362207412,
      "num_tokens": 21512122.0,
      "step": 9635
    },
    {
      "entropy": 5.477610349655151,
      "epoch": 1.0826641208513506,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004893909864380714,
      "loss": 5.2542,
      "mean_token_accuracy": 0.18706072121858597,
      "num_tokens": 21521896.0,
      "step": 9640
    },
    {
      "entropy": 5.502262163162231,
      "epoch": 1.0832256977593082,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000489378800080952,
      "loss": 5.4232,
      "mean_token_accuracy": 0.18383677750825883,
      "num_tokens": 21533584.0,
      "step": 9645
    },
    {
      "entropy": 5.418279790878296,
      "epoch": 1.0837872746672657,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004893666068979283,
      "loss": 5.3868,
      "mean_token_accuracy": 0.19056437462568282,
      "num_tokens": 21545432.0,
      "step": 9650
    },
    {
      "entropy": 5.534392881393432,
      "epoch": 1.0843488515752233,
      "grad_norm": 1.125,
      "learning_rate": 0.0004893544068893887,
      "loss": 5.3864,
      "mean_token_accuracy": 0.1842215746641159,
      "num_tokens": 21556547.0,
      "step": 9655
    },
    {
      "entropy": 5.5567206859588625,
      "epoch": 1.0849104284831808,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004893422000557217,
      "loss": 5.4479,
      "mean_token_accuracy": 0.1804027333855629,
      "num_tokens": 21567354.0,
      "step": 9660
    },
    {
      "entropy": 5.567187309265137,
      "epoch": 1.0854720053911384,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004893299863973159,
      "loss": 5.3219,
      "mean_token_accuracy": 0.18749481737613677,
      "num_tokens": 21578694.0,
      "step": 9665
    },
    {
      "entropy": 5.474405145645141,
      "epoch": 1.0860335822990959,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004893177659145601,
      "loss": 5.274,
      "mean_token_accuracy": 0.19735342413187026,
      "num_tokens": 21589365.0,
      "step": 9670
    },
    {
      "entropy": 5.520022010803222,
      "epoch": 1.0865951592070533,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004893055386078438,
      "loss": 5.3314,
      "mean_token_accuracy": 0.18308279663324356,
      "num_tokens": 21600766.0,
      "step": 9675
    },
    {
      "entropy": 5.533956050872803,
      "epoch": 1.087156736115011,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004892933044775558,
      "loss": 5.534,
      "mean_token_accuracy": 0.178544320166111,
      "num_tokens": 21613969.0,
      "step": 9680
    },
    {
      "entropy": 5.494122648239136,
      "epoch": 1.0877183130229684,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004892810635240859,
      "loss": 5.3484,
      "mean_token_accuracy": 0.18665636330842972,
      "num_tokens": 21626302.0,
      "step": 9685
    },
    {
      "entropy": 5.582063674926758,
      "epoch": 1.088279889930926,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004892688157478238,
      "loss": 5.4242,
      "mean_token_accuracy": 0.18517893850803374,
      "num_tokens": 21637058.0,
      "step": 9690
    },
    {
      "entropy": 5.57568507194519,
      "epoch": 1.0888414668388835,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004892565611491597,
      "loss": 5.4384,
      "mean_token_accuracy": 0.17996353805065154,
      "num_tokens": 21648729.0,
      "step": 9695
    },
    {
      "entropy": 5.421488428115845,
      "epoch": 1.0894030437468412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004892442997284834,
      "loss": 5.2302,
      "mean_token_accuracy": 0.19667865931987763,
      "num_tokens": 21659961.0,
      "step": 9700
    },
    {
      "entropy": 5.4760987758636475,
      "epoch": 1.0899646206547986,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004892320314861855,
      "loss": 5.3471,
      "mean_token_accuracy": 0.1910601496696472,
      "num_tokens": 21670577.0,
      "step": 9705
    },
    {
      "entropy": 5.474126720428467,
      "epoch": 1.0905261975627563,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004892197564226568,
      "loss": 5.3096,
      "mean_token_accuracy": 0.195725616812706,
      "num_tokens": 21679882.0,
      "step": 9710
    },
    {
      "entropy": 5.569431114196777,
      "epoch": 1.0910877744707137,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004892074745382878,
      "loss": 5.4347,
      "mean_token_accuracy": 0.19174170345067978,
      "num_tokens": 21690595.0,
      "step": 9715
    },
    {
      "entropy": 5.459299182891845,
      "epoch": 1.0916493513786714,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004891951858334698,
      "loss": 5.2853,
      "mean_token_accuracy": 0.19190464466810225,
      "num_tokens": 21701251.0,
      "step": 9720
    },
    {
      "entropy": 5.563612699508667,
      "epoch": 1.0922109282866288,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004891828903085939,
      "loss": 5.4836,
      "mean_token_accuracy": 0.18425981253385543,
      "num_tokens": 21713286.0,
      "step": 9725
    },
    {
      "entropy": 5.58175835609436,
      "epoch": 1.0927725051945865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004891705879640518,
      "loss": 5.5244,
      "mean_token_accuracy": 0.1812610000371933,
      "num_tokens": 21725508.0,
      "step": 9730
    },
    {
      "entropy": 5.508434200286866,
      "epoch": 1.093334082102544,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004891582788002351,
      "loss": 5.3307,
      "mean_token_accuracy": 0.18853978514671327,
      "num_tokens": 21736467.0,
      "step": 9735
    },
    {
      "entropy": 5.599121046066284,
      "epoch": 1.0938956590105016,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004891459628175358,
      "loss": 5.4172,
      "mean_token_accuracy": 0.1899057596921921,
      "num_tokens": 21746114.0,
      "step": 9740
    },
    {
      "entropy": 5.482849264144898,
      "epoch": 1.094457235918459,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000489133640016346,
      "loss": 5.4166,
      "mean_token_accuracy": 0.1853234961628914,
      "num_tokens": 21757145.0,
      "step": 9745
    },
    {
      "entropy": 5.4897778034210205,
      "epoch": 1.0950188128264167,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004891213103970579,
      "loss": 5.3906,
      "mean_token_accuracy": 0.18188587725162506,
      "num_tokens": 21768294.0,
      "step": 9750
    },
    {
      "entropy": 5.565903949737549,
      "epoch": 1.095580389734374,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004891089739600643,
      "loss": 5.4016,
      "mean_token_accuracy": 0.18574694842100142,
      "num_tokens": 21779667.0,
      "step": 9755
    },
    {
      "entropy": 5.454914569854736,
      "epoch": 1.0961419666423318,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004890966307057581,
      "loss": 5.2588,
      "mean_token_accuracy": 0.19705150574445723,
      "num_tokens": 21789635.0,
      "step": 9760
    },
    {
      "entropy": 5.410786819458008,
      "epoch": 1.0967035435502892,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000489084280634532,
      "loss": 5.2576,
      "mean_token_accuracy": 0.19602492451667786,
      "num_tokens": 21799462.0,
      "step": 9765
    },
    {
      "entropy": 5.416881847381592,
      "epoch": 1.0972651204582466,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004890719237467794,
      "loss": 5.3138,
      "mean_token_accuracy": 0.1895948529243469,
      "num_tokens": 21811520.0,
      "step": 9770
    },
    {
      "entropy": 5.622190618515015,
      "epoch": 1.0978266973662043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004890595600428937,
      "loss": 5.3953,
      "mean_token_accuracy": 0.18550430834293366,
      "num_tokens": 21821997.0,
      "step": 9775
    },
    {
      "entropy": 5.57149920463562,
      "epoch": 1.0983882742741617,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004890471895232684,
      "loss": 5.4493,
      "mean_token_accuracy": 0.18469329476356505,
      "num_tokens": 21833353.0,
      "step": 9780
    },
    {
      "entropy": 5.491009521484375,
      "epoch": 1.0989498511821194,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004890348121882978,
      "loss": 5.3606,
      "mean_token_accuracy": 0.1915326952934265,
      "num_tokens": 21843691.0,
      "step": 9785
    },
    {
      "entropy": 5.468305063247681,
      "epoch": 1.0995114280900768,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004890224280383758,
      "loss": 5.3459,
      "mean_token_accuracy": 0.19110767245292665,
      "num_tokens": 21853771.0,
      "step": 9790
    },
    {
      "entropy": 5.413413190841675,
      "epoch": 1.1000730049980345,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004890100370738966,
      "loss": 5.2153,
      "mean_token_accuracy": 0.19955267161130905,
      "num_tokens": 21863424.0,
      "step": 9795
    },
    {
      "entropy": 5.507965183258056,
      "epoch": 1.100634581905992,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004889976392952548,
      "loss": 5.3446,
      "mean_token_accuracy": 0.18554313480854034,
      "num_tokens": 21875407.0,
      "step": 9800
    },
    {
      "entropy": 5.513894748687744,
      "epoch": 1.1011961588139496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004889852347028452,
      "loss": 5.4146,
      "mean_token_accuracy": 0.18251552283763886,
      "num_tokens": 21887081.0,
      "step": 9805
    },
    {
      "entropy": 5.485643911361694,
      "epoch": 1.101757735721907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004889728232970626,
      "loss": 5.3215,
      "mean_token_accuracy": 0.18611925095319748,
      "num_tokens": 21897858.0,
      "step": 9810
    },
    {
      "entropy": 5.50360255241394,
      "epoch": 1.1023193126298647,
      "grad_norm": 1.125,
      "learning_rate": 0.0004889604050783025,
      "loss": 5.367,
      "mean_token_accuracy": 0.1846587523818016,
      "num_tokens": 21907985.0,
      "step": 9815
    },
    {
      "entropy": 5.49332365989685,
      "epoch": 1.1028808895378222,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048894798004696,
      "loss": 5.4128,
      "mean_token_accuracy": 0.18538483530282973,
      "num_tokens": 21918726.0,
      "step": 9820
    },
    {
      "entropy": 5.463434791564941,
      "epoch": 1.1034424664457798,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004889355482034309,
      "loss": 5.3261,
      "mean_token_accuracy": 0.18947667479515076,
      "num_tokens": 21929027.0,
      "step": 9825
    },
    {
      "entropy": 5.456523656845093,
      "epoch": 1.1040040433537373,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004889231095481111,
      "loss": 5.257,
      "mean_token_accuracy": 0.19990183860063554,
      "num_tokens": 21938356.0,
      "step": 9830
    },
    {
      "entropy": 5.46082034111023,
      "epoch": 1.104565620261695,
      "grad_norm": 1.125,
      "learning_rate": 0.0004889106640813964,
      "loss": 5.2924,
      "mean_token_accuracy": 0.19054461866617203,
      "num_tokens": 21947889.0,
      "step": 9835
    },
    {
      "entropy": 5.50875072479248,
      "epoch": 1.1051271971696524,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004888982118036831,
      "loss": 5.3827,
      "mean_token_accuracy": 0.18880287557840347,
      "num_tokens": 21958409.0,
      "step": 9840
    },
    {
      "entropy": 5.438988304138183,
      "epoch": 1.10568877407761,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004888857527153679,
      "loss": 5.3542,
      "mean_token_accuracy": 0.18572020828723906,
      "num_tokens": 21970120.0,
      "step": 9845
    },
    {
      "entropy": 5.491817474365234,
      "epoch": 1.1062503509855675,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004888732868168473,
      "loss": 5.36,
      "mean_token_accuracy": 0.1902606576681137,
      "num_tokens": 21981499.0,
      "step": 9850
    },
    {
      "entropy": 5.42215747833252,
      "epoch": 1.1068119278935251,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004888608141085183,
      "loss": 5.2482,
      "mean_token_accuracy": 0.20036101937294007,
      "num_tokens": 21991263.0,
      "step": 9855
    },
    {
      "entropy": 5.460143899917602,
      "epoch": 1.1073735048014826,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004888483345907781,
      "loss": 5.2468,
      "mean_token_accuracy": 0.1842061311006546,
      "num_tokens": 22001582.0,
      "step": 9860
    },
    {
      "entropy": 5.446298122406006,
      "epoch": 1.10793508170944,
      "grad_norm": 1.140625,
      "learning_rate": 0.000488835848264024,
      "loss": 5.3266,
      "mean_token_accuracy": 0.19687162339687347,
      "num_tokens": 22012447.0,
      "step": 9865
    },
    {
      "entropy": 5.496721887588501,
      "epoch": 1.1084966586173977,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004888233551286534,
      "loss": 5.4143,
      "mean_token_accuracy": 0.18907029330730438,
      "num_tokens": 22024865.0,
      "step": 9870
    },
    {
      "entropy": 5.469445371627808,
      "epoch": 1.109058235525355,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004888108551850644,
      "loss": 5.3311,
      "mean_token_accuracy": 0.190682515501976,
      "num_tokens": 22036298.0,
      "step": 9875
    },
    {
      "entropy": 5.547363233566284,
      "epoch": 1.1096198124333128,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004887983484336548,
      "loss": 5.3401,
      "mean_token_accuracy": 0.1882574364542961,
      "num_tokens": 22047499.0,
      "step": 9880
    },
    {
      "entropy": 5.517907047271729,
      "epoch": 1.1101813893412702,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004887858348748229,
      "loss": 5.3848,
      "mean_token_accuracy": 0.18664682954549788,
      "num_tokens": 22057612.0,
      "step": 9885
    },
    {
      "entropy": 5.4974652290344235,
      "epoch": 1.1107429662492279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000488773314508967,
      "loss": 5.4156,
      "mean_token_accuracy": 0.18530358076095582,
      "num_tokens": 22068566.0,
      "step": 9890
    },
    {
      "entropy": 5.473715686798096,
      "epoch": 1.1113045431571853,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004887607873364859,
      "loss": 5.2856,
      "mean_token_accuracy": 0.1880566105246544,
      "num_tokens": 22079821.0,
      "step": 9895
    },
    {
      "entropy": 5.489977693557739,
      "epoch": 1.111866120065143,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004887482533577786,
      "loss": 5.335,
      "mean_token_accuracy": 0.1866757184267044,
      "num_tokens": 22090143.0,
      "step": 9900
    },
    {
      "entropy": 5.445848846435547,
      "epoch": 1.1124276969731004,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004887357125732438,
      "loss": 5.395,
      "mean_token_accuracy": 0.18721785098314286,
      "num_tokens": 22101601.0,
      "step": 9905
    },
    {
      "entropy": 5.50714430809021,
      "epoch": 1.112989273881058,
      "grad_norm": 1.109375,
      "learning_rate": 0.000488723164983281,
      "loss": 5.3121,
      "mean_token_accuracy": 0.1923036679625511,
      "num_tokens": 22113102.0,
      "step": 9910
    },
    {
      "entropy": 5.5688591480255125,
      "epoch": 1.1135508507890155,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004887106105882898,
      "loss": 5.4613,
      "mean_token_accuracy": 0.18365613073110582,
      "num_tokens": 22124650.0,
      "step": 9915
    },
    {
      "entropy": 5.545090818405152,
      "epoch": 1.1141124276969732,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004886980493886699,
      "loss": 5.388,
      "mean_token_accuracy": 0.1909029722213745,
      "num_tokens": 22136637.0,
      "step": 9920
    },
    {
      "entropy": 5.465398502349854,
      "epoch": 1.1146740046049306,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004886854813848212,
      "loss": 5.3099,
      "mean_token_accuracy": 0.19008917212486268,
      "num_tokens": 22149318.0,
      "step": 9925
    },
    {
      "entropy": 5.384241342544556,
      "epoch": 1.1152355815128883,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004886729065771439,
      "loss": 5.1965,
      "mean_token_accuracy": 0.19351300746202468,
      "num_tokens": 22159410.0,
      "step": 9930
    },
    {
      "entropy": 5.425002002716065,
      "epoch": 1.1157971584208457,
      "grad_norm": 1.125,
      "learning_rate": 0.0004886603249660384,
      "loss": 5.2794,
      "mean_token_accuracy": 0.18795328289270402,
      "num_tokens": 22170017.0,
      "step": 9935
    },
    {
      "entropy": 5.449740934371948,
      "epoch": 1.1163587353288034,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004886477365519052,
      "loss": 5.2828,
      "mean_token_accuracy": 0.1909773141145706,
      "num_tokens": 22179961.0,
      "step": 9940
    },
    {
      "entropy": 5.555003643035889,
      "epoch": 1.1169203122367608,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004886351413351453,
      "loss": 5.4265,
      "mean_token_accuracy": 0.1794225588440895,
      "num_tokens": 22191415.0,
      "step": 9945
    },
    {
      "entropy": 5.509974718093872,
      "epoch": 1.1174818891447185,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004886225393161596,
      "loss": 5.3652,
      "mean_token_accuracy": 0.19871425330638887,
      "num_tokens": 22205354.0,
      "step": 9950
    },
    {
      "entropy": 5.595667791366577,
      "epoch": 1.118043466052676,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004886099304953495,
      "loss": 5.4172,
      "mean_token_accuracy": 0.1836776241660118,
      "num_tokens": 22216989.0,
      "step": 9955
    },
    {
      "entropy": 5.58510389328003,
      "epoch": 1.1186050429606333,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004885973148731162,
      "loss": 5.4163,
      "mean_token_accuracy": 0.18037967681884765,
      "num_tokens": 22228421.0,
      "step": 9960
    },
    {
      "entropy": 5.463361406326294,
      "epoch": 1.119166619868591,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004885846924498617,
      "loss": 5.2873,
      "mean_token_accuracy": 0.18683721423149108,
      "num_tokens": 22240621.0,
      "step": 9965
    },
    {
      "entropy": 5.410719871520996,
      "epoch": 1.1197281967765484,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004885720632259876,
      "loss": 5.2622,
      "mean_token_accuracy": 0.19615741223096847,
      "num_tokens": 22250936.0,
      "step": 9970
    },
    {
      "entropy": 5.475217628479004,
      "epoch": 1.120289773684506,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004885594272018963,
      "loss": 5.3882,
      "mean_token_accuracy": 0.18656658232212067,
      "num_tokens": 22261766.0,
      "step": 9975
    },
    {
      "entropy": 5.475364446640015,
      "epoch": 1.1208513505924635,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048854678437799,
      "loss": 5.319,
      "mean_token_accuracy": 0.1898321345448494,
      "num_tokens": 22271831.0,
      "step": 9980
    },
    {
      "entropy": 5.427309131622314,
      "epoch": 1.1214129275004212,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004885341347546713,
      "loss": 5.3458,
      "mean_token_accuracy": 0.18728182464838028,
      "num_tokens": 22283756.0,
      "step": 9985
    },
    {
      "entropy": 5.4209051609039305,
      "epoch": 1.1219745044083786,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004885214783323428,
      "loss": 5.2965,
      "mean_token_accuracy": 0.18717077374458313,
      "num_tokens": 22294576.0,
      "step": 9990
    },
    {
      "entropy": 5.495046854019165,
      "epoch": 1.1225360813163363,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004885088151114076,
      "loss": 5.3587,
      "mean_token_accuracy": 0.18578314632177353,
      "num_tokens": 22305003.0,
      "step": 9995
    },
    {
      "entropy": 5.46302809715271,
      "epoch": 1.1230976582242937,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004884961450922691,
      "loss": 5.3157,
      "mean_token_accuracy": 0.19266755282878875,
      "num_tokens": 22316505.0,
      "step": 10000
    },
    {
      "entropy": 5.515137434005737,
      "epoch": 1.1236592351322514,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884834682753305,
      "loss": 5.3772,
      "mean_token_accuracy": 0.19046587198972703,
      "num_tokens": 22327853.0,
      "step": 10005
    },
    {
      "entropy": 5.558890104293823,
      "epoch": 1.1242208120402089,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004884707846609953,
      "loss": 5.2952,
      "mean_token_accuracy": 0.188982817530632,
      "num_tokens": 22337386.0,
      "step": 10010
    },
    {
      "entropy": 5.539405822753906,
      "epoch": 1.1247823889481665,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004884580942496677,
      "loss": 5.4759,
      "mean_token_accuracy": 0.18247807174921035,
      "num_tokens": 22348742.0,
      "step": 10015
    },
    {
      "entropy": 5.583350753784179,
      "epoch": 1.125343965856124,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004884453970417515,
      "loss": 5.4014,
      "mean_token_accuracy": 0.1852103129029274,
      "num_tokens": 22360083.0,
      "step": 10020
    },
    {
      "entropy": 5.500770807266235,
      "epoch": 1.1259055427640816,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000488432693037651,
      "loss": 5.3609,
      "mean_token_accuracy": 0.1815495163202286,
      "num_tokens": 22370982.0,
      "step": 10025
    },
    {
      "entropy": 5.451694679260254,
      "epoch": 1.126467119672039,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000488419982237771,
      "loss": 5.3063,
      "mean_token_accuracy": 0.20067557394504548,
      "num_tokens": 22382383.0,
      "step": 10030
    },
    {
      "entropy": 5.519569683074951,
      "epoch": 1.1270286965799967,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004884072646425157,
      "loss": 5.3186,
      "mean_token_accuracy": 0.18760548233985902,
      "num_tokens": 22393994.0,
      "step": 10035
    },
    {
      "entropy": 5.419602727890014,
      "epoch": 1.1275902734879542,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004883945402522907,
      "loss": 5.2394,
      "mean_token_accuracy": 0.19836754649877547,
      "num_tokens": 22405354.0,
      "step": 10040
    },
    {
      "entropy": 5.387629604339599,
      "epoch": 1.1281518503959118,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004883818090675005,
      "loss": 5.2984,
      "mean_token_accuracy": 0.18958158791065216,
      "num_tokens": 22416708.0,
      "step": 10045
    },
    {
      "entropy": 5.441644954681396,
      "epoch": 1.1287134273038693,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004883690710885507,
      "loss": 5.2888,
      "mean_token_accuracy": 0.19262771159410477,
      "num_tokens": 22427724.0,
      "step": 10050
    },
    {
      "entropy": 5.5489989757537845,
      "epoch": 1.1292750042118267,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004883563263158471,
      "loss": 5.3878,
      "mean_token_accuracy": 0.18374199718236922,
      "num_tokens": 22437823.0,
      "step": 10055
    },
    {
      "entropy": 5.523835563659668,
      "epoch": 1.1298365811197844,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004883435747497952,
      "loss": 5.4159,
      "mean_token_accuracy": 0.1875823125243187,
      "num_tokens": 22450155.0,
      "step": 10060
    },
    {
      "entropy": 5.53595986366272,
      "epoch": 1.130398158027742,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004883308163908012,
      "loss": 5.3377,
      "mean_token_accuracy": 0.19261740893125534,
      "num_tokens": 22460484.0,
      "step": 10065
    },
    {
      "entropy": 5.434231376647949,
      "epoch": 1.1309597349356995,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004883180512392714,
      "loss": 5.3338,
      "mean_token_accuracy": 0.18646218478679658,
      "num_tokens": 22472063.0,
      "step": 10070
    },
    {
      "entropy": 5.421738958358764,
      "epoch": 1.131521311843657,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004883052792956119,
      "loss": 5.3373,
      "mean_token_accuracy": 0.19664047062397003,
      "num_tokens": 22483669.0,
      "step": 10075
    },
    {
      "entropy": 5.559923267364502,
      "epoch": 1.1320828887516146,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004882925005602296,
      "loss": 5.4136,
      "mean_token_accuracy": 0.18587087243795394,
      "num_tokens": 22495505.0,
      "step": 10080
    },
    {
      "entropy": 5.499471282958984,
      "epoch": 1.132644465659572,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004882797150335314,
      "loss": 5.3628,
      "mean_token_accuracy": 0.18663066774606704,
      "num_tokens": 22505260.0,
      "step": 10085
    },
    {
      "entropy": 5.515253400802612,
      "epoch": 1.1332060425675297,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048826692271592434,
      "loss": 5.4206,
      "mean_token_accuracy": 0.1849902629852295,
      "num_tokens": 22516362.0,
      "step": 10090
    },
    {
      "entropy": 5.570123624801636,
      "epoch": 1.133767619475487,
      "grad_norm": 1.125,
      "learning_rate": 0.0004882541236078158,
      "loss": 5.3758,
      "mean_token_accuracy": 0.18518655598163605,
      "num_tokens": 22527760.0,
      "step": 10095
    },
    {
      "entropy": 5.689021730422974,
      "epoch": 1.1343291963834448,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048824131770961316,
      "loss": 5.4669,
      "mean_token_accuracy": 0.17877719551324844,
      "num_tokens": 22540601.0,
      "step": 10100
    },
    {
      "entropy": 5.5218689918518065,
      "epoch": 1.1348907732914022,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048822850502172437,
      "loss": 5.3798,
      "mean_token_accuracy": 0.18207437992095948,
      "num_tokens": 22552469.0,
      "step": 10105
    },
    {
      "entropy": 5.379859828948975,
      "epoch": 1.1354523501993599,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004882156855445571,
      "loss": 5.3306,
      "mean_token_accuracy": 0.1915702611207962,
      "num_tokens": 22563552.0,
      "step": 10110
    },
    {
      "entropy": 5.4801393985748295,
      "epoch": 1.1360139271073173,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004882028592785199,
      "loss": 5.306,
      "mean_token_accuracy": 0.19141027182340623,
      "num_tokens": 22575619.0,
      "step": 10115
    },
    {
      "entropy": 5.467701959609985,
      "epoch": 1.136575504015275,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048819002622402095,
      "loss": 5.2863,
      "mean_token_accuracy": 0.1910892367362976,
      "num_tokens": 22585854.0,
      "step": 10120
    },
    {
      "entropy": 5.524414014816284,
      "epoch": 1.1371370809232324,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004881771863814688,
      "loss": 5.3095,
      "mean_token_accuracy": 0.19435176998376846,
      "num_tokens": 22595899.0,
      "step": 10125
    },
    {
      "entropy": 5.407589101791382,
      "epoch": 1.13769865783119,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048816433975127244,
      "loss": 5.2913,
      "mean_token_accuracy": 0.19016751348972322,
      "num_tokens": 22607517.0,
      "step": 10130
    },
    {
      "entropy": 5.501319694519043,
      "epoch": 1.1382602347391475,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004881514863338407,
      "loss": 5.4031,
      "mean_token_accuracy": 0.18679691404104232,
      "num_tokens": 22618531.0,
      "step": 10135
    },
    {
      "entropy": 5.388666439056396,
      "epoch": 1.1388218116471052,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048813862612958307,
      "loss": 5.2495,
      "mean_token_accuracy": 0.1937074512243271,
      "num_tokens": 22629193.0,
      "step": 10140
    },
    {
      "entropy": 5.523422384262085,
      "epoch": 1.1393833885550626,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004881257591389089,
      "loss": 5.3675,
      "mean_token_accuracy": 0.19131201058626174,
      "num_tokens": 22639794.0,
      "step": 10145
    },
    {
      "entropy": 5.546735048294067,
      "epoch": 1.13994496546302,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048811288536222794,
      "loss": 5.3187,
      "mean_token_accuracy": 0.1938394457101822,
      "num_tokens": 22650061.0,
      "step": 10150
    },
    {
      "entropy": 5.480262565612793,
      "epoch": 1.1405065423709777,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048810000479995,
      "loss": 5.4005,
      "mean_token_accuracy": 0.18121499419212342,
      "num_tokens": 22661399.0,
      "step": 10155
    },
    {
      "entropy": 5.635375690460205,
      "epoch": 1.1410681192789354,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048808711745248524,
      "loss": 5.4886,
      "mean_token_accuracy": 0.18087699860334397,
      "num_tokens": 22672445.0,
      "step": 10160
    },
    {
      "entropy": 5.555114841461181,
      "epoch": 1.1416296961868928,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048807422332024404,
      "loss": 5.2962,
      "mean_token_accuracy": 0.19056684970855714,
      "num_tokens": 22682545.0,
      "step": 10165
    },
    {
      "entropy": 5.468265962600708,
      "epoch": 1.1421912730948502,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004880613224036369,
      "loss": 5.3508,
      "mean_token_accuracy": 0.1824083372950554,
      "num_tokens": 22693885.0,
      "step": 10170
    },
    {
      "entropy": 5.4522919178009035,
      "epoch": 1.142752850002808,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048804841470307466,
      "loss": 5.2971,
      "mean_token_accuracy": 0.18848683089017867,
      "num_tokens": 22703841.0,
      "step": 10175
    },
    {
      "entropy": 5.590392589569092,
      "epoch": 1.1433144269107653,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048803550021896824,
      "loss": 5.4413,
      "mean_token_accuracy": 0.1882365733385086,
      "num_tokens": 22714142.0,
      "step": 10180
    },
    {
      "entropy": 5.53915376663208,
      "epoch": 1.143876003818723,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004880225789517289,
      "loss": 5.3155,
      "mean_token_accuracy": 0.18552613705396653,
      "num_tokens": 22725056.0,
      "step": 10185
    },
    {
      "entropy": 5.4065001010894775,
      "epoch": 1.1444375807266804,
      "grad_norm": 1.3125,
      "learning_rate": 0.000488009650901768,
      "loss": 5.3452,
      "mean_token_accuracy": 0.1924957364797592,
      "num_tokens": 22735664.0,
      "step": 10190
    },
    {
      "entropy": 5.450916433334351,
      "epoch": 1.144999157634638,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048799671606949724,
      "loss": 5.3151,
      "mean_token_accuracy": 0.19113609790802003,
      "num_tokens": 22748071.0,
      "step": 10195
    },
    {
      "entropy": 5.533325099945069,
      "epoch": 1.1455607345425955,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048798377445532837,
      "loss": 5.3782,
      "mean_token_accuracy": 0.1861809179186821,
      "num_tokens": 22759477.0,
      "step": 10200
    },
    {
      "entropy": 5.493514156341552,
      "epoch": 1.1461223114505532,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004879708260596736,
      "loss": 5.2539,
      "mean_token_accuracy": 0.1958856463432312,
      "num_tokens": 22769955.0,
      "step": 10205
    },
    {
      "entropy": 5.419086170196533,
      "epoch": 1.1466838883585107,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004879578708829451,
      "loss": 5.3503,
      "mean_token_accuracy": 0.18614689409732818,
      "num_tokens": 22781216.0,
      "step": 10210
    },
    {
      "entropy": 5.436713027954101,
      "epoch": 1.1472454652664683,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004879449089255553,
      "loss": 5.2921,
      "mean_token_accuracy": 0.18933203518390657,
      "num_tokens": 22792581.0,
      "step": 10215
    },
    {
      "entropy": 5.422752380371094,
      "epoch": 1.1478070421744258,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004879319401879171,
      "loss": 5.2549,
      "mean_token_accuracy": 0.19116601198911667,
      "num_tokens": 22803305.0,
      "step": 10220
    },
    {
      "entropy": 5.560664892196655,
      "epoch": 1.1483686190823834,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048791896467044335,
      "loss": 5.3888,
      "mean_token_accuracy": 0.18957023620605468,
      "num_tokens": 22815558.0,
      "step": 10225
    },
    {
      "entropy": 5.477578258514404,
      "epoch": 1.1489301959903409,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048790598237354716,
      "loss": 5.2999,
      "mean_token_accuracy": 0.19268295168876648,
      "num_tokens": 22825428.0,
      "step": 10230
    },
    {
      "entropy": 5.495377683639527,
      "epoch": 1.1494917728982985,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004878929932976419,
      "loss": 5.3487,
      "mean_token_accuracy": 0.19178205877542495,
      "num_tokens": 22836294.0,
      "step": 10235
    },
    {
      "entropy": 5.4914785861969,
      "epoch": 1.150053349806256,
      "grad_norm": 1.25,
      "learning_rate": 0.0004878799974431412,
      "loss": 5.4317,
      "mean_token_accuracy": 0.19050806164741516,
      "num_tokens": 22847835.0,
      "step": 10240
    },
    {
      "entropy": 5.513508224487305,
      "epoch": 1.1506149267142134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004878669948104588,
      "loss": 5.3659,
      "mean_token_accuracy": 0.18653237372636794,
      "num_tokens": 22859726.0,
      "step": 10245
    },
    {
      "entropy": 5.433402252197266,
      "epoch": 1.151176503622171,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048785398540000867,
      "loss": 5.3195,
      "mean_token_accuracy": 0.18754699677228928,
      "num_tokens": 22871250.0,
      "step": 10250
    },
    {
      "entropy": 5.541570472717285,
      "epoch": 1.1517380805301287,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048784096921220514,
      "loss": 5.3975,
      "mean_token_accuracy": 0.1847960025072098,
      "num_tokens": 22882034.0,
      "step": 10255
    },
    {
      "entropy": 5.631517171859741,
      "epoch": 1.1522996574380862,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004878279462474627,
      "loss": 5.4759,
      "mean_token_accuracy": 0.18365521728992462,
      "num_tokens": 22894606.0,
      "step": 10260
    },
    {
      "entropy": 5.440877389907837,
      "epoch": 1.1528612343460436,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048781491650619574,
      "loss": 5.3115,
      "mean_token_accuracy": 0.19253068268299103,
      "num_tokens": 22905246.0,
      "step": 10265
    },
    {
      "entropy": 5.446545505523682,
      "epoch": 1.1534228112540013,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048780187998881944,
      "loss": 5.2604,
      "mean_token_accuracy": 0.18477636873722075,
      "num_tokens": 22916202.0,
      "step": 10270
    },
    {
      "entropy": 5.470248508453369,
      "epoch": 1.1539843881619587,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004877888366957486,
      "loss": 5.2578,
      "mean_token_accuracy": 0.18614826798439027,
      "num_tokens": 22926451.0,
      "step": 10275
    },
    {
      "entropy": 5.520329093933105,
      "epoch": 1.1545459650699164,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004877757866273988,
      "loss": 5.348,
      "mean_token_accuracy": 0.19069368839263917,
      "num_tokens": 22936618.0,
      "step": 10280
    },
    {
      "entropy": 5.560961484909058,
      "epoch": 1.1551075419778738,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004877627297841853,
      "loss": 5.381,
      "mean_token_accuracy": 0.17827016711235047,
      "num_tokens": 22948278.0,
      "step": 10285
    },
    {
      "entropy": 5.44008960723877,
      "epoch": 1.1556691188858315,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004877496661665241,
      "loss": 5.2742,
      "mean_token_accuracy": 0.19207292199134826,
      "num_tokens": 22959928.0,
      "step": 10290
    },
    {
      "entropy": 5.384089946746826,
      "epoch": 1.156230695793789,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000487736595774831,
      "loss": 5.3298,
      "mean_token_accuracy": 0.1838379830121994,
      "num_tokens": 22971050.0,
      "step": 10295
    },
    {
      "entropy": 5.41632490158081,
      "epoch": 1.1567922727017466,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048772351860952215,
      "loss": 5.2422,
      "mean_token_accuracy": 0.1910719946026802,
      "num_tokens": 22981471.0,
      "step": 10300
    },
    {
      "entropy": 5.601700258255005,
      "epoch": 1.157353849609704,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004877104346710139,
      "loss": 5.4813,
      "mean_token_accuracy": 0.18961743414402008,
      "num_tokens": 22992526.0,
      "step": 10305
    },
    {
      "entropy": 5.472655868530273,
      "epoch": 1.1579154265176617,
      "grad_norm": 1.015625,
      "learning_rate": 0.000487697343959723,
      "loss": 5.3413,
      "mean_token_accuracy": 0.18970987647771836,
      "num_tokens": 23003408.0,
      "step": 10310
    },
    {
      "entropy": 5.460650539398193,
      "epoch": 1.158477003425619,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004876842464760661,
      "loss": 5.2563,
      "mean_token_accuracy": 0.1915055900812149,
      "num_tokens": 23013406.0,
      "step": 10315
    },
    {
      "entropy": 5.437350940704346,
      "epoch": 1.1590385803335768,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048767114222046036,
      "loss": 5.3264,
      "mean_token_accuracy": 0.19100383818149566,
      "num_tokens": 23024960.0,
      "step": 10320
    },
    {
      "entropy": 5.54367356300354,
      "epoch": 1.1596001572415342,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000487658031193323,
      "loss": 5.3505,
      "mean_token_accuracy": 0.1825086921453476,
      "num_tokens": 23037514.0,
      "step": 10325
    },
    {
      "entropy": 5.5518410205841064,
      "epoch": 1.1601617341494919,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004876449133950714,
      "loss": 5.4485,
      "mean_token_accuracy": 0.17935476303100586,
      "num_tokens": 23048460.0,
      "step": 10330
    },
    {
      "entropy": 5.54437141418457,
      "epoch": 1.1607233110574493,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004876317888261233,
      "loss": 5.3051,
      "mean_token_accuracy": 0.1868131637573242,
      "num_tokens": 23058866.0,
      "step": 10335
    },
    {
      "entropy": 5.425964832305908,
      "epoch": 1.1612848879654067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048761865748689657,
      "loss": 5.34,
      "mean_token_accuracy": 0.1852980837225914,
      "num_tokens": 23069097.0,
      "step": 10340
    },
    {
      "entropy": 5.55105037689209,
      "epoch": 1.1618464648733644,
      "grad_norm": 1.125,
      "learning_rate": 0.00048760551937780927,
      "loss": 5.3217,
      "mean_token_accuracy": 0.18968528807163237,
      "num_tokens": 23079553.0,
      "step": 10345
    },
    {
      "entropy": 5.473508358001709,
      "epoch": 1.162408041781322,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048759237449927986,
      "loss": 5.3361,
      "mean_token_accuracy": 0.18588747680187226,
      "num_tokens": 23090554.0,
      "step": 10350
    },
    {
      "entropy": 5.41131443977356,
      "epoch": 1.1629696186892795,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048757922285172674,
      "loss": 5.2312,
      "mean_token_accuracy": 0.190905624628067,
      "num_tokens": 23101740.0,
      "step": 10355
    },
    {
      "entropy": 5.473917245864868,
      "epoch": 1.163531195597237,
      "grad_norm": 1.5,
      "learning_rate": 0.00048756606443556874,
      "loss": 5.3354,
      "mean_token_accuracy": 0.18988496959209442,
      "num_tokens": 23112842.0,
      "step": 10360
    },
    {
      "entropy": 5.434205722808838,
      "epoch": 1.1640927725051946,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048755289925122474,
      "loss": 5.3221,
      "mean_token_accuracy": 0.1870110347867012,
      "num_tokens": 23123822.0,
      "step": 10365
    },
    {
      "entropy": 5.526471328735352,
      "epoch": 1.164654349413152,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004875397272991141,
      "loss": 5.3419,
      "mean_token_accuracy": 0.18134610950946808,
      "num_tokens": 23134045.0,
      "step": 10370
    },
    {
      "entropy": 5.447522306442261,
      "epoch": 1.1652159263211097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000487526548579656,
      "loss": 5.3734,
      "mean_token_accuracy": 0.18452945798635484,
      "num_tokens": 23144200.0,
      "step": 10375
    },
    {
      "entropy": 5.455108690261841,
      "epoch": 1.1657775032290671,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048751336309327026,
      "loss": 5.307,
      "mean_token_accuracy": 0.192894583940506,
      "num_tokens": 23155671.0,
      "step": 10380
    },
    {
      "entropy": 5.567395973205566,
      "epoch": 1.1663390801370248,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004875001708403766,
      "loss": 5.4471,
      "mean_token_accuracy": 0.18633103370666504,
      "num_tokens": 23166644.0,
      "step": 10385
    },
    {
      "entropy": 5.525482654571533,
      "epoch": 1.1669006570449822,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048748697182139503,
      "loss": 5.3293,
      "mean_token_accuracy": 0.1920440375804901,
      "num_tokens": 23177074.0,
      "step": 10390
    },
    {
      "entropy": 5.534948301315308,
      "epoch": 1.16746223395294,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004874737660367459,
      "loss": 5.4461,
      "mean_token_accuracy": 0.18229351192712784,
      "num_tokens": 23188724.0,
      "step": 10395
    },
    {
      "entropy": 5.466331672668457,
      "epoch": 1.1680238108608973,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048746055348684967,
      "loss": 5.3522,
      "mean_token_accuracy": 0.1878226652741432,
      "num_tokens": 23198459.0,
      "step": 10400
    },
    {
      "entropy": 5.553414821624756,
      "epoch": 1.168585387768855,
      "grad_norm": 1.078125,
      "learning_rate": 0.000487447334172127,
      "loss": 5.361,
      "mean_token_accuracy": 0.18774810135364534,
      "num_tokens": 23208991.0,
      "step": 10405
    },
    {
      "entropy": 5.479293727874756,
      "epoch": 1.1691469646768125,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048743410809299877,
      "loss": 5.2926,
      "mean_token_accuracy": 0.18909573256969453,
      "num_tokens": 23219309.0,
      "step": 10410
    },
    {
      "entropy": 5.473764657974243,
      "epoch": 1.1697085415847701,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004874208752498862,
      "loss": 5.3993,
      "mean_token_accuracy": 0.18177039921283722,
      "num_tokens": 23230551.0,
      "step": 10415
    },
    {
      "entropy": 5.559957695007324,
      "epoch": 1.1702701184927276,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004874076356432106,
      "loss": 5.461,
      "mean_token_accuracy": 0.18303379118442537,
      "num_tokens": 23242799.0,
      "step": 10420
    },
    {
      "entropy": 5.5983470439910885,
      "epoch": 1.1708316954006852,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048739438927339346,
      "loss": 5.4458,
      "mean_token_accuracy": 0.179739548265934,
      "num_tokens": 23255112.0,
      "step": 10425
    },
    {
      "entropy": 5.47545485496521,
      "epoch": 1.1713932723086427,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048738113614085656,
      "loss": 5.3497,
      "mean_token_accuracy": 0.19257825911045073,
      "num_tokens": 23264800.0,
      "step": 10430
    },
    {
      "entropy": 5.449072074890137,
      "epoch": 1.1719548492166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000487367876246022,
      "loss": 5.2913,
      "mean_token_accuracy": 0.1865327313542366,
      "num_tokens": 23276672.0,
      "step": 10435
    },
    {
      "entropy": 5.482340717315674,
      "epoch": 1.1725164261245578,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004873546095893118,
      "loss": 5.3291,
      "mean_token_accuracy": 0.18801960945129395,
      "num_tokens": 23288282.0,
      "step": 10440
    },
    {
      "entropy": 5.491970252990723,
      "epoch": 1.1730780030325154,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004873413361711485,
      "loss": 5.3468,
      "mean_token_accuracy": 0.18586714565753937,
      "num_tokens": 23299078.0,
      "step": 10445
    },
    {
      "entropy": 5.497157001495362,
      "epoch": 1.1736395799404729,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048732805599195465,
      "loss": 5.3358,
      "mean_token_accuracy": 0.19111865609884263,
      "num_tokens": 23310036.0,
      "step": 10450
    },
    {
      "entropy": 5.4897033214569095,
      "epoch": 1.1742011568484303,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048731476905215326,
      "loss": 5.268,
      "mean_token_accuracy": 0.19594541937112808,
      "num_tokens": 23321029.0,
      "step": 10455
    },
    {
      "entropy": 5.50123085975647,
      "epoch": 1.174762733756388,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004873014753521673,
      "loss": 5.3796,
      "mean_token_accuracy": 0.1866510823369026,
      "num_tokens": 23331584.0,
      "step": 10460
    },
    {
      "entropy": 5.635983562469482,
      "epoch": 1.1753243106643454,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004872881748924199,
      "loss": 5.4467,
      "mean_token_accuracy": 0.17778573036193848,
      "num_tokens": 23343171.0,
      "step": 10465
    },
    {
      "entropy": 5.559485721588135,
      "epoch": 1.175885887572303,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004872748676733347,
      "loss": 5.3635,
      "mean_token_accuracy": 0.18856250047683715,
      "num_tokens": 23353047.0,
      "step": 10470
    },
    {
      "entropy": 5.4335010051727295,
      "epoch": 1.1764474644802605,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048726155369533547,
      "loss": 5.3611,
      "mean_token_accuracy": 0.18665113747119905,
      "num_tokens": 23364447.0,
      "step": 10475
    },
    {
      "entropy": 5.5190877437591555,
      "epoch": 1.1770090413882182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000487248232958846,
      "loss": 5.3897,
      "mean_token_accuracy": 0.18602234572172166,
      "num_tokens": 23375616.0,
      "step": 10480
    },
    {
      "entropy": 5.6089424133300785,
      "epoch": 1.1775706182961756,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004872349054642904,
      "loss": 5.4896,
      "mean_token_accuracy": 0.18161313533782958,
      "num_tokens": 23388176.0,
      "step": 10485
    },
    {
      "entropy": 5.583784341812134,
      "epoch": 1.1781321952041333,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004872215712120932,
      "loss": 5.4174,
      "mean_token_accuracy": 0.18961915075778962,
      "num_tokens": 23400563.0,
      "step": 10490
    },
    {
      "entropy": 5.504327201843262,
      "epoch": 1.1786937721120907,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004872082302026788,
      "loss": 5.3573,
      "mean_token_accuracy": 0.18537813574075698,
      "num_tokens": 23410740.0,
      "step": 10495
    },
    {
      "entropy": 5.484878921508789,
      "epoch": 1.1792553490200484,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004871948824364721,
      "loss": 5.4446,
      "mean_token_accuracy": 0.18258217871189117,
      "num_tokens": 23423086.0,
      "step": 10500
    },
    {
      "entropy": 5.565518712997436,
      "epoch": 1.1798169259280058,
      "grad_norm": 1.171875,
      "learning_rate": 0.000487181527913898,
      "loss": 5.4402,
      "mean_token_accuracy": 0.18355826139450074,
      "num_tokens": 23435869.0,
      "step": 10505
    },
    {
      "entropy": 5.514801931381226,
      "epoch": 1.1803785028359635,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048716816663538177,
      "loss": 5.2992,
      "mean_token_accuracy": 0.20019642412662506,
      "num_tokens": 23446635.0,
      "step": 10510
    },
    {
      "entropy": 5.476276874542236,
      "epoch": 1.180940079743921,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004871547986013488,
      "loss": 5.3254,
      "mean_token_accuracy": 0.18937005400657653,
      "num_tokens": 23458214.0,
      "step": 10515
    },
    {
      "entropy": 5.51998610496521,
      "epoch": 1.1815016566518786,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004871414238122248,
      "loss": 5.352,
      "mean_token_accuracy": 0.19283366650342942,
      "num_tokens": 23469383.0,
      "step": 10520
    },
    {
      "entropy": 5.449571895599365,
      "epoch": 1.182063233559836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004871280422684355,
      "loss": 5.3195,
      "mean_token_accuracy": 0.19173895865678786,
      "num_tokens": 23480841.0,
      "step": 10525
    },
    {
      "entropy": 5.428569221496582,
      "epoch": 1.1826248104677934,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048711465397040726,
      "loss": 5.2743,
      "mean_token_accuracy": 0.1885039508342743,
      "num_tokens": 23490796.0,
      "step": 10530
    },
    {
      "entropy": 5.445909738540649,
      "epoch": 1.183186387375751,
      "grad_norm": 1.25,
      "learning_rate": 0.00048710125891856604,
      "loss": 5.3153,
      "mean_token_accuracy": 0.1842427521944046,
      "num_tokens": 23501759.0,
      "step": 10535
    },
    {
      "entropy": 5.496493244171143,
      "epoch": 1.1837479642837088,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048708785711333845,
      "loss": 5.3457,
      "mean_token_accuracy": 0.18848443180322647,
      "num_tokens": 23512209.0,
      "step": 10540
    },
    {
      "entropy": 5.517921113967896,
      "epoch": 1.1843095411916662,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048707444855515124,
      "loss": 5.4205,
      "mean_token_accuracy": 0.1864883452653885,
      "num_tokens": 23524316.0,
      "step": 10545
    },
    {
      "entropy": 5.5062822818756105,
      "epoch": 1.1848711180996236,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004870610332444314,
      "loss": 5.3972,
      "mean_token_accuracy": 0.1910739853978157,
      "num_tokens": 23536199.0,
      "step": 10550
    },
    {
      "entropy": 5.4898323059082035,
      "epoch": 1.1854326950075813,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004870476111816059,
      "loss": 5.3967,
      "mean_token_accuracy": 0.18637777268886566,
      "num_tokens": 23548034.0,
      "step": 10555
    },
    {
      "entropy": 5.551316785812378,
      "epoch": 1.1859942719155387,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048703418236710234,
      "loss": 5.3548,
      "mean_token_accuracy": 0.19020506888628005,
      "num_tokens": 23559280.0,
      "step": 10560
    },
    {
      "entropy": 5.5113687992095945,
      "epoch": 1.1865558488234964,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004870207468013481,
      "loss": 5.3361,
      "mean_token_accuracy": 0.19059309214353562,
      "num_tokens": 23569163.0,
      "step": 10565
    },
    {
      "entropy": 5.510538244247437,
      "epoch": 1.1871174257314538,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004870073044847711,
      "loss": 5.396,
      "mean_token_accuracy": 0.18955078572034836,
      "num_tokens": 23580526.0,
      "step": 10570
    },
    {
      "entropy": 5.416756916046142,
      "epoch": 1.1876790026394115,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004869938554177993,
      "loss": 5.2844,
      "mean_token_accuracy": 0.18405936062335967,
      "num_tokens": 23590552.0,
      "step": 10575
    },
    {
      "entropy": 5.332808208465576,
      "epoch": 1.188240579547369,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048698039960086084,
      "loss": 5.1591,
      "mean_token_accuracy": 0.19522219747304917,
      "num_tokens": 23601889.0,
      "step": 10580
    },
    {
      "entropy": 5.490883445739746,
      "epoch": 1.1888021564553266,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004869669370343843,
      "loss": 5.4437,
      "mean_token_accuracy": 0.19266334921121597,
      "num_tokens": 23613833.0,
      "step": 10585
    },
    {
      "entropy": 5.590994930267334,
      "epoch": 1.189363733363284,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004869534677187983,
      "loss": 5.3926,
      "mean_token_accuracy": 0.19101812690496445,
      "num_tokens": 23625404.0,
      "step": 10590
    },
    {
      "entropy": 5.470618438720703,
      "epoch": 1.1899253102712417,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004869399916545317,
      "loss": 5.2762,
      "mean_token_accuracy": 0.19313479363918304,
      "num_tokens": 23635670.0,
      "step": 10595
    },
    {
      "entropy": 5.41665506362915,
      "epoch": 1.1904868871791991,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004869265088420135,
      "loss": 5.2622,
      "mean_token_accuracy": 0.1956461876630783,
      "num_tokens": 23646893.0,
      "step": 10600
    },
    {
      "entropy": 5.395202207565307,
      "epoch": 1.1910484640871568,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004869130192816731,
      "loss": 5.3825,
      "mean_token_accuracy": 0.1856496348977089,
      "num_tokens": 23657733.0,
      "step": 10605
    },
    {
      "entropy": 5.422374296188354,
      "epoch": 1.1916100409951143,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048689952297394,
      "loss": 5.2195,
      "mean_token_accuracy": 0.19550398737192154,
      "num_tokens": 23668100.0,
      "step": 10610
    },
    {
      "entropy": 5.540370607376099,
      "epoch": 1.192171617903072,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004868860199192438,
      "loss": 5.3806,
      "mean_token_accuracy": 0.1858476459980011,
      "num_tokens": 23678295.0,
      "step": 10615
    },
    {
      "entropy": 5.485654497146607,
      "epoch": 1.1927331948110294,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004868725101180146,
      "loss": 5.2324,
      "mean_token_accuracy": 0.1951396644115448,
      "num_tokens": 23689310.0,
      "step": 10620
    },
    {
      "entropy": 5.436852025985718,
      "epoch": 1.1932947717189868,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004868589935706826,
      "loss": 5.3966,
      "mean_token_accuracy": 0.18616850078105926,
      "num_tokens": 23699475.0,
      "step": 10625
    },
    {
      "entropy": 5.580835056304932,
      "epoch": 1.1938563486269445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048684547027767793,
      "loss": 5.3794,
      "mean_token_accuracy": 0.18866919577121735,
      "num_tokens": 23710186.0,
      "step": 10630
    },
    {
      "entropy": 5.396511554718018,
      "epoch": 1.1944179255349021,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004868319402394314,
      "loss": 5.1921,
      "mean_token_accuracy": 0.1933600828051567,
      "num_tokens": 23720449.0,
      "step": 10635
    },
    {
      "entropy": 5.5301899909973145,
      "epoch": 1.1949795024428596,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048681840345637375,
      "loss": 5.452,
      "mean_token_accuracy": 0.18851781636476517,
      "num_tokens": 23732295.0,
      "step": 10640
    },
    {
      "entropy": 5.452880477905273,
      "epoch": 1.195541079350817,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048680485992893597,
      "loss": 5.3516,
      "mean_token_accuracy": 0.1847606897354126,
      "num_tokens": 23743616.0,
      "step": 10645
    },
    {
      "entropy": 5.56827073097229,
      "epoch": 1.1961026562587747,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048679130965754926,
      "loss": 5.3958,
      "mean_token_accuracy": 0.1836438223719597,
      "num_tokens": 23755288.0,
      "step": 10650
    },
    {
      "entropy": 5.506766748428345,
      "epoch": 1.196664233166732,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004867777526426451,
      "loss": 5.2596,
      "mean_token_accuracy": 0.19082026481628417,
      "num_tokens": 23765916.0,
      "step": 10655
    },
    {
      "entropy": 5.3702747344970705,
      "epoch": 1.1972258100746898,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048676418888465525,
      "loss": 5.2107,
      "mean_token_accuracy": 0.19341834783554077,
      "num_tokens": 23778891.0,
      "step": 10660
    },
    {
      "entropy": 5.4559619426727295,
      "epoch": 1.1977873869826472,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004867506183840114,
      "loss": 5.3699,
      "mean_token_accuracy": 0.1918458342552185,
      "num_tokens": 23790402.0,
      "step": 10665
    },
    {
      "entropy": 5.436944055557251,
      "epoch": 1.1983489638906049,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004867370411411459,
      "loss": 5.3074,
      "mean_token_accuracy": 0.1992880582809448,
      "num_tokens": 23801928.0,
      "step": 10670
    },
    {
      "entropy": 5.55572361946106,
      "epoch": 1.1989105407985623,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048672345715649066,
      "loss": 5.3973,
      "mean_token_accuracy": 0.1797810286283493,
      "num_tokens": 23812302.0,
      "step": 10675
    },
    {
      "entropy": 5.549669694900513,
      "epoch": 1.19947211770652,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004867098664304785,
      "loss": 5.3553,
      "mean_token_accuracy": 0.1992868646979332,
      "num_tokens": 23824033.0,
      "step": 10680
    },
    {
      "entropy": 5.441835021972656,
      "epoch": 1.2000336946144774,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004866962689635422,
      "loss": 5.3074,
      "mean_token_accuracy": 0.19039777517318726,
      "num_tokens": 23834907.0,
      "step": 10685
    },
    {
      "entropy": 5.53205418586731,
      "epoch": 1.200595271522435,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048668266475611447,
      "loss": 5.4973,
      "mean_token_accuracy": 0.18706956952810289,
      "num_tokens": 23846418.0,
      "step": 10690
    },
    {
      "entropy": 5.594804286956787,
      "epoch": 1.2011568484303925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004866690538086286,
      "loss": 5.426,
      "mean_token_accuracy": 0.18486818969249724,
      "num_tokens": 23858423.0,
      "step": 10695
    },
    {
      "entropy": 5.525947713851929,
      "epoch": 1.2017184253383502,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048665543612151803,
      "loss": 5.3224,
      "mean_token_accuracy": 0.19153492003679276,
      "num_tokens": 23868340.0,
      "step": 10700
    },
    {
      "entropy": 5.427148675918579,
      "epoch": 1.2022800022463076,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048664181169521625,
      "loss": 5.2579,
      "mean_token_accuracy": 0.19293180108070374,
      "num_tokens": 23878752.0,
      "step": 10705
    },
    {
      "entropy": 5.4996744155883786,
      "epoch": 1.2028415791542653,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004866281805301571,
      "loss": 5.318,
      "mean_token_accuracy": 0.19087314158678054,
      "num_tokens": 23889374.0,
      "step": 10710
    },
    {
      "entropy": 5.530946254730225,
      "epoch": 1.2034031560622227,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004866145426267746,
      "loss": 5.374,
      "mean_token_accuracy": 0.18851903080940247,
      "num_tokens": 23901651.0,
      "step": 10715
    },
    {
      "entropy": 5.414171886444092,
      "epoch": 1.2039647329701804,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000486600897985503,
      "loss": 5.2956,
      "mean_token_accuracy": 0.1927599087357521,
      "num_tokens": 23911818.0,
      "step": 10720
    },
    {
      "entropy": 5.508558034896851,
      "epoch": 1.2045263098781378,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048658724660677674,
      "loss": 5.342,
      "mean_token_accuracy": 0.19158872216939926,
      "num_tokens": 23923840.0,
      "step": 10725
    },
    {
      "entropy": 5.47757625579834,
      "epoch": 1.2050878867860955,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004865735884910305,
      "loss": 5.2345,
      "mean_token_accuracy": 0.1947578877210617,
      "num_tokens": 23935406.0,
      "step": 10730
    },
    {
      "entropy": 5.48161211013794,
      "epoch": 1.205649463694053,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004865599236386991,
      "loss": 5.3391,
      "mean_token_accuracy": 0.1922893762588501,
      "num_tokens": 23946350.0,
      "step": 10735
    },
    {
      "entropy": 5.378890943527222,
      "epoch": 1.2062110406020103,
      "grad_norm": 1.125,
      "learning_rate": 0.0004865462520502177,
      "loss": 5.2911,
      "mean_token_accuracy": 0.19203284680843352,
      "num_tokens": 23956315.0,
      "step": 10740
    },
    {
      "entropy": 5.4803853034973145,
      "epoch": 1.206772617509968,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004865325737260216,
      "loss": 5.2019,
      "mean_token_accuracy": 0.18931492567062377,
      "num_tokens": 23966649.0,
      "step": 10745
    },
    {
      "entropy": 5.5576153755187985,
      "epoch": 1.2073341944179254,
      "grad_norm": 1.125,
      "learning_rate": 0.00048651888866654624,
      "loss": 5.4741,
      "mean_token_accuracy": 0.17762160748243333,
      "num_tokens": 23977674.0,
      "step": 10750
    },
    {
      "entropy": 5.442694044113159,
      "epoch": 1.207895771325883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004865051968722275,
      "loss": 5.3609,
      "mean_token_accuracy": 0.18723692148923873,
      "num_tokens": 23987985.0,
      "step": 10755
    },
    {
      "entropy": 5.5725297927856445,
      "epoch": 1.2084573482338405,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048649149834350126,
      "loss": 5.4555,
      "mean_token_accuracy": 0.18374131619930267,
      "num_tokens": 23998730.0,
      "step": 10760
    },
    {
      "entropy": 5.442189359664917,
      "epoch": 1.2090189251417982,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004864777930808037,
      "loss": 5.2953,
      "mean_token_accuracy": 0.18207336962223053,
      "num_tokens": 24008743.0,
      "step": 10765
    },
    {
      "entropy": 5.4576087474823,
      "epoch": 1.2095805020497556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004864640810845711,
      "loss": 5.3407,
      "mean_token_accuracy": 0.18913229405879975,
      "num_tokens": 24019805.0,
      "step": 10770
    },
    {
      "entropy": 5.501690101623535,
      "epoch": 1.2101420789577133,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004864503623552402,
      "loss": 5.3261,
      "mean_token_accuracy": 0.18727027773857116,
      "num_tokens": 24031437.0,
      "step": 10775
    },
    {
      "entropy": 5.5729974746704105,
      "epoch": 1.2107036558656707,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004864366368932478,
      "loss": 5.4505,
      "mean_token_accuracy": 0.1780677929520607,
      "num_tokens": 24043922.0,
      "step": 10780
    },
    {
      "entropy": 5.526761817932129,
      "epoch": 1.2112652327736284,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048642290469903083,
      "loss": 5.2709,
      "mean_token_accuracy": 0.1968682363629341,
      "num_tokens": 24054755.0,
      "step": 10785
    },
    {
      "entropy": 5.44307689666748,
      "epoch": 1.2118268096815858,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048640916577302664,
      "loss": 5.2823,
      "mean_token_accuracy": 0.18202337771654128,
      "num_tokens": 24067347.0,
      "step": 10790
    },
    {
      "entropy": 5.503612089157104,
      "epoch": 1.2123883865895435,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048639542011567243,
      "loss": 5.3385,
      "mean_token_accuracy": 0.19037413448095322,
      "num_tokens": 24078938.0,
      "step": 10795
    },
    {
      "entropy": 5.5415795803070065,
      "epoch": 1.212949963497501,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004863816677274062,
      "loss": 5.355,
      "mean_token_accuracy": 0.18644218295812606,
      "num_tokens": 24090275.0,
      "step": 10800
    },
    {
      "entropy": 5.482690811157227,
      "epoch": 1.2135115404054586,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048636790860866576,
      "loss": 5.3654,
      "mean_token_accuracy": 0.19203082770109176,
      "num_tokens": 24101518.0,
      "step": 10805
    },
    {
      "entropy": 5.555949640274048,
      "epoch": 1.214073117313416,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004863541427598889,
      "loss": 5.4494,
      "mean_token_accuracy": 0.18214682191610337,
      "num_tokens": 24113320.0,
      "step": 10810
    },
    {
      "entropy": 5.512860536575317,
      "epoch": 1.2146346942213737,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004863403701815144,
      "loss": 5.2884,
      "mean_token_accuracy": 0.19154454469680787,
      "num_tokens": 24124842.0,
      "step": 10815
    },
    {
      "entropy": 5.527637052536011,
      "epoch": 1.2151962711293312,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004863265908739805,
      "loss": 5.4154,
      "mean_token_accuracy": 0.19004831165075303,
      "num_tokens": 24136391.0,
      "step": 10820
    },
    {
      "entropy": 5.518775367736817,
      "epoch": 1.2157578480372888,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004863128048377259,
      "loss": 5.3768,
      "mean_token_accuracy": 0.18957658410072326,
      "num_tokens": 24147194.0,
      "step": 10825
    },
    {
      "entropy": 5.56171498298645,
      "epoch": 1.2163194249452463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004862990120731896,
      "loss": 5.4485,
      "mean_token_accuracy": 0.18261944651603698,
      "num_tokens": 24158170.0,
      "step": 10830
    },
    {
      "entropy": 5.516436910629272,
      "epoch": 1.2168810018532037,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048628521258081096,
      "loss": 5.2822,
      "mean_token_accuracy": 0.1917682930827141,
      "num_tokens": 24168671.0,
      "step": 10835
    },
    {
      "entropy": 5.451374912261963,
      "epoch": 1.2174425787611614,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048627140636102906,
      "loss": 5.3267,
      "mean_token_accuracy": 0.19494537264108658,
      "num_tokens": 24179898.0,
      "step": 10840
    },
    {
      "entropy": 5.390147399902344,
      "epoch": 1.2180041556691188,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004862575934142837,
      "loss": 5.3172,
      "mean_token_accuracy": 0.19178963899612428,
      "num_tokens": 24190622.0,
      "step": 10845
    },
    {
      "entropy": 5.666590404510498,
      "epoch": 1.2185657325770765,
      "grad_norm": 1.125,
      "learning_rate": 0.0004862437737410147,
      "loss": 5.4901,
      "mean_token_accuracy": 0.18107927292585374,
      "num_tokens": 24202526.0,
      "step": 10850
    },
    {
      "entropy": 5.502965974807739,
      "epoch": 1.219127309485034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004862299473416619,
      "loss": 5.3159,
      "mean_token_accuracy": 0.18561350256204606,
      "num_tokens": 24213186.0,
      "step": 10855
    },
    {
      "entropy": 5.423367118835449,
      "epoch": 1.2196888863929916,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048621611421666573,
      "loss": 5.3253,
      "mean_token_accuracy": 0.1877934679389,
      "num_tokens": 24225490.0,
      "step": 10860
    },
    {
      "entropy": 5.504482269287109,
      "epoch": 1.220250463300949,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004862022743664665,
      "loss": 5.3113,
      "mean_token_accuracy": 0.19293032884597777,
      "num_tokens": 24236653.0,
      "step": 10865
    },
    {
      "entropy": 5.50580325126648,
      "epoch": 1.2208120402089067,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048618842779150494,
      "loss": 5.2864,
      "mean_token_accuracy": 0.1966002181172371,
      "num_tokens": 24247843.0,
      "step": 10870
    },
    {
      "entropy": 5.400511264801025,
      "epoch": 1.221373617116864,
      "grad_norm": 1.125,
      "learning_rate": 0.00048617457449222205,
      "loss": 5.3601,
      "mean_token_accuracy": 0.19113782048225403,
      "num_tokens": 24258332.0,
      "step": 10875
    },
    {
      "entropy": 5.492725467681884,
      "epoch": 1.2219351940248218,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004861607144690587,
      "loss": 5.3457,
      "mean_token_accuracy": 0.1893364816904068,
      "num_tokens": 24268685.0,
      "step": 10880
    },
    {
      "entropy": 5.556686973571777,
      "epoch": 1.2224967709327792,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004861468477224563,
      "loss": 5.3703,
      "mean_token_accuracy": 0.18632747530937194,
      "num_tokens": 24280660.0,
      "step": 10885
    },
    {
      "entropy": 5.531473541259766,
      "epoch": 1.2230583478407369,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048613297425285635,
      "loss": 5.3023,
      "mean_token_accuracy": 0.19230025857686997,
      "num_tokens": 24291649.0,
      "step": 10890
    },
    {
      "entropy": 5.449414205551148,
      "epoch": 1.2236199247486943,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004861190940607007,
      "loss": 5.3174,
      "mean_token_accuracy": 0.19521680921316148,
      "num_tokens": 24303108.0,
      "step": 10895
    },
    {
      "entropy": 5.438139867782593,
      "epoch": 1.224181501656652,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004861052071464312,
      "loss": 5.3431,
      "mean_token_accuracy": 0.183199942111969,
      "num_tokens": 24315939.0,
      "step": 10900
    },
    {
      "entropy": 5.492931270599366,
      "epoch": 1.2247430785646094,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048609131351048994,
      "loss": 5.3215,
      "mean_token_accuracy": 0.19137313961982727,
      "num_tokens": 24326916.0,
      "step": 10905
    },
    {
      "entropy": 5.507109022140503,
      "epoch": 1.225304655472567,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004860774131533195,
      "loss": 5.3813,
      "mean_token_accuracy": 0.19142188876867294,
      "num_tokens": 24337790.0,
      "step": 10910
    },
    {
      "entropy": 5.408120346069336,
      "epoch": 1.2258662323805245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048606350607536235,
      "loss": 5.1917,
      "mean_token_accuracy": 0.19919081181287765,
      "num_tokens": 24347930.0,
      "step": 10915
    },
    {
      "entropy": 5.454737520217895,
      "epoch": 1.2264278092884822,
      "grad_norm": 1.125,
      "learning_rate": 0.0004860495922770612,
      "loss": 5.3036,
      "mean_token_accuracy": 0.18901157826185228,
      "num_tokens": 24359213.0,
      "step": 10920
    },
    {
      "entropy": 5.445453786849976,
      "epoch": 1.2269893861964396,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048603567175885926,
      "loss": 5.2538,
      "mean_token_accuracy": 0.1886252284049988,
      "num_tokens": 24371000.0,
      "step": 10925
    },
    {
      "entropy": 5.448771858215332,
      "epoch": 1.227550963104397,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004860217445211997,
      "loss": 5.1989,
      "mean_token_accuracy": 0.19589506089687347,
      "num_tokens": 24381301.0,
      "step": 10930
    },
    {
      "entropy": 5.559473848342895,
      "epoch": 1.2281125400123547,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004860078105645259,
      "loss": 5.4411,
      "mean_token_accuracy": 0.17990793585777282,
      "num_tokens": 24391808.0,
      "step": 10935
    },
    {
      "entropy": 5.6326392650604244,
      "epoch": 1.2286741169203121,
      "grad_norm": 1.0,
      "learning_rate": 0.0004859938698892816,
      "loss": 5.5473,
      "mean_token_accuracy": 0.18333044052124023,
      "num_tokens": 24404220.0,
      "step": 10940
    },
    {
      "entropy": 5.559707450866699,
      "epoch": 1.2292356938282698,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048597992249591065,
      "loss": 5.3616,
      "mean_token_accuracy": 0.1912491202354431,
      "num_tokens": 24414906.0,
      "step": 10945
    },
    {
      "entropy": 5.422472763061523,
      "epoch": 1.2297972707362272,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048596596838485704,
      "loss": 5.3387,
      "mean_token_accuracy": 0.18478153496980668,
      "num_tokens": 24426469.0,
      "step": 10950
    },
    {
      "entropy": 5.3907982349395756,
      "epoch": 1.230358847644185,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004859520075565654,
      "loss": 5.298,
      "mean_token_accuracy": 0.19148481339216233,
      "num_tokens": 24437683.0,
      "step": 10955
    },
    {
      "entropy": 5.4061651706695555,
      "epoch": 1.2309204245521423,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048593804001147986,
      "loss": 5.2058,
      "mean_token_accuracy": 0.19314789026975632,
      "num_tokens": 24449038.0,
      "step": 10960
    },
    {
      "entropy": 5.50235333442688,
      "epoch": 1.2314820014601,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004859240657500453,
      "loss": 5.3554,
      "mean_token_accuracy": 0.18964497298002242,
      "num_tokens": 24460228.0,
      "step": 10965
    },
    {
      "entropy": 5.462199687957764,
      "epoch": 1.2320435783680574,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048591008477270675,
      "loss": 5.3812,
      "mean_token_accuracy": 0.18497704714536667,
      "num_tokens": 24471222.0,
      "step": 10970
    },
    {
      "entropy": 5.413447046279908,
      "epoch": 1.232605155276015,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048589609707990925,
      "loss": 5.3304,
      "mean_token_accuracy": 0.18635518997907638,
      "num_tokens": 24482127.0,
      "step": 10975
    },
    {
      "entropy": 5.511392974853516,
      "epoch": 1.2331667321839725,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048588210267209823,
      "loss": 5.3125,
      "mean_token_accuracy": 0.1897481694817543,
      "num_tokens": 24492963.0,
      "step": 10980
    },
    {
      "entropy": 5.548450136184693,
      "epoch": 1.2337283090919302,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048586810154971935,
      "loss": 5.3381,
      "mean_token_accuracy": 0.19336990565061568,
      "num_tokens": 24502885.0,
      "step": 10985
    },
    {
      "entropy": 5.5679937362670895,
      "epoch": 1.2342898859998876,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048585409371321826,
      "loss": 5.438,
      "mean_token_accuracy": 0.18847732096910477,
      "num_tokens": 24514990.0,
      "step": 10990
    },
    {
      "entropy": 5.528681898117066,
      "epoch": 1.2348514629078453,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048584007916304106,
      "loss": 5.3529,
      "mean_token_accuracy": 0.18421241790056228,
      "num_tokens": 24525555.0,
      "step": 10995
    },
    {
      "entropy": 5.485701847076416,
      "epoch": 1.2354130398158027,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048582605789963395,
      "loss": 5.3938,
      "mean_token_accuracy": 0.18213511556386947,
      "num_tokens": 24536350.0,
      "step": 11000
    },
    {
      "entropy": 5.588993835449219,
      "epoch": 1.2359746167237604,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004858120299234434,
      "loss": 5.4034,
      "mean_token_accuracy": 0.18713052719831466,
      "num_tokens": 24547756.0,
      "step": 11005
    },
    {
      "entropy": 5.509750890731811,
      "epoch": 1.2365361936317179,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048579799523491607,
      "loss": 5.2776,
      "mean_token_accuracy": 0.19096165597438813,
      "num_tokens": 24559897.0,
      "step": 11010
    },
    {
      "entropy": 5.467511320114136,
      "epoch": 1.2370977705396755,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004857839538344988,
      "loss": 5.3089,
      "mean_token_accuracy": 0.18433302342891694,
      "num_tokens": 24570168.0,
      "step": 11015
    },
    {
      "entropy": 5.460204935073852,
      "epoch": 1.237659347447633,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048576990572263875,
      "loss": 5.2539,
      "mean_token_accuracy": 0.19639374762773515,
      "num_tokens": 24580918.0,
      "step": 11020
    },
    {
      "entropy": 5.410836982727051,
      "epoch": 1.2382209243555904,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004857558508997831,
      "loss": 5.2837,
      "mean_token_accuracy": 0.18541559875011443,
      "num_tokens": 24592378.0,
      "step": 11025
    },
    {
      "entropy": 5.5328937530517575,
      "epoch": 1.238782501263548,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004857417893663794,
      "loss": 5.3732,
      "mean_token_accuracy": 0.18891598731279374,
      "num_tokens": 24603666.0,
      "step": 11030
    },
    {
      "entropy": 5.484682369232178,
      "epoch": 1.2393440781715055,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004857277211228754,
      "loss": 5.3345,
      "mean_token_accuracy": 0.18914095163345337,
      "num_tokens": 24615299.0,
      "step": 11035
    },
    {
      "entropy": 5.496534729003907,
      "epoch": 1.2399056550794632,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004857136461697191,
      "loss": 5.3202,
      "mean_token_accuracy": 0.1921449452638626,
      "num_tokens": 24626138.0,
      "step": 11040
    },
    {
      "entropy": 5.4452039241790775,
      "epoch": 1.2404672319874206,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048569956450735853,
      "loss": 5.2981,
      "mean_token_accuracy": 0.19092509895563126,
      "num_tokens": 24636720.0,
      "step": 11045
    },
    {
      "entropy": 5.506981277465821,
      "epoch": 1.2410288088953783,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048568547613624206,
      "loss": 5.3201,
      "mean_token_accuracy": 0.18461062908172607,
      "num_tokens": 24648147.0,
      "step": 11050
    },
    {
      "entropy": 5.461820936203003,
      "epoch": 1.2415903858033357,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048567138105681833,
      "loss": 5.3638,
      "mean_token_accuracy": 0.18919588029384612,
      "num_tokens": 24659825.0,
      "step": 11055
    },
    {
      "entropy": 5.477649545669555,
      "epoch": 1.2421519627112934,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048565727926953617,
      "loss": 5.3236,
      "mean_token_accuracy": 0.1912950709462166,
      "num_tokens": 24671583.0,
      "step": 11060
    },
    {
      "entropy": 5.443061256408692,
      "epoch": 1.2427135396192508,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004856431707748445,
      "loss": 5.2595,
      "mean_token_accuracy": 0.19250702112913132,
      "num_tokens": 24682034.0,
      "step": 11065
    },
    {
      "entropy": 5.463092708587647,
      "epoch": 1.2432751165272085,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004856290555731926,
      "loss": 5.3473,
      "mean_token_accuracy": 0.18822134882211686,
      "num_tokens": 24693624.0,
      "step": 11070
    },
    {
      "entropy": 5.557461404800415,
      "epoch": 1.243836693435166,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004856149336650299,
      "loss": 5.3631,
      "mean_token_accuracy": 0.18502741307020187,
      "num_tokens": 24705340.0,
      "step": 11075
    },
    {
      "entropy": 5.620899343490601,
      "epoch": 1.2443982703431236,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004856008050508059,
      "loss": 5.4156,
      "mean_token_accuracy": 0.18347597867250443,
      "num_tokens": 24717115.0,
      "step": 11080
    },
    {
      "entropy": 5.468814563751221,
      "epoch": 1.244959847251081,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004855866697309707,
      "loss": 5.2434,
      "mean_token_accuracy": 0.19057439416646957,
      "num_tokens": 24727768.0,
      "step": 11085
    },
    {
      "entropy": 5.479076480865478,
      "epoch": 1.2455214241590387,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004855725277059742,
      "loss": 5.4109,
      "mean_token_accuracy": 0.1810232073068619,
      "num_tokens": 24740054.0,
      "step": 11090
    },
    {
      "entropy": 5.50800929069519,
      "epoch": 1.246083001066996,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004855583789762668,
      "loss": 5.283,
      "mean_token_accuracy": 0.19207758456468582,
      "num_tokens": 24750658.0,
      "step": 11095
    },
    {
      "entropy": 5.605063104629517,
      "epoch": 1.2466445779749538,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000485544223542299,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1885771170258522,
      "num_tokens": 24762730.0,
      "step": 11100
    },
    {
      "entropy": 5.432342052459717,
      "epoch": 1.2472061548829112,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004855300614045214,
      "loss": 5.2305,
      "mean_token_accuracy": 0.19524402618408204,
      "num_tokens": 24773865.0,
      "step": 11105
    },
    {
      "entropy": 5.421746683120728,
      "epoch": 1.2477677317908689,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048551589256338497,
      "loss": 5.3435,
      "mean_token_accuracy": 0.1914224237203598,
      "num_tokens": 24783030.0,
      "step": 11110
    },
    {
      "entropy": 5.414378118515015,
      "epoch": 1.2483293086988263,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048550171701934087,
      "loss": 5.2622,
      "mean_token_accuracy": 0.19432681947946548,
      "num_tokens": 24793177.0,
      "step": 11115
    },
    {
      "entropy": 5.558236885070801,
      "epoch": 1.2488908856067837,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048548753477284053,
      "loss": 5.3723,
      "mean_token_accuracy": 0.19053647071123123,
      "num_tokens": 24804796.0,
      "step": 11120
    },
    {
      "entropy": 5.559025812149048,
      "epoch": 1.2494524625147414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004854733458243354,
      "loss": 5.3767,
      "mean_token_accuracy": 0.18704819679260254,
      "num_tokens": 24816806.0,
      "step": 11125
    },
    {
      "entropy": 5.454526090621949,
      "epoch": 1.250014039422699,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004854591501742773,
      "loss": 5.3259,
      "mean_token_accuracy": 0.1941091910004616,
      "num_tokens": 24826888.0,
      "step": 11130
    },
    {
      "entropy": 5.470983648300171,
      "epoch": 1.2505756163306565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048544494782311825,
      "loss": 5.2679,
      "mean_token_accuracy": 0.19485928863286972,
      "num_tokens": 24837731.0,
      "step": 11135
    },
    {
      "entropy": 5.458429908752441,
      "epoch": 1.251137193238614,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048543073877131037,
      "loss": 5.3003,
      "mean_token_accuracy": 0.1871982917189598,
      "num_tokens": 24847900.0,
      "step": 11140
    },
    {
      "entropy": 5.462789726257324,
      "epoch": 1.2516987701465716,
      "grad_norm": 1.125,
      "learning_rate": 0.00048541652301930627,
      "loss": 5.3874,
      "mean_token_accuracy": 0.1854668065905571,
      "num_tokens": 24858878.0,
      "step": 11145
    },
    {
      "entropy": 5.493943071365356,
      "epoch": 1.252260347054529,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004854023005675585,
      "loss": 5.3234,
      "mean_token_accuracy": 0.1905045062303543,
      "num_tokens": 24870051.0,
      "step": 11150
    },
    {
      "entropy": 5.4933915615081785,
      "epoch": 1.2528219239624867,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004853880714165197,
      "loss": 5.4166,
      "mean_token_accuracy": 0.18264361917972566,
      "num_tokens": 24881671.0,
      "step": 11155
    },
    {
      "entropy": 5.537044477462769,
      "epoch": 1.2533835008704441,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048537383556664315,
      "loss": 5.2761,
      "mean_token_accuracy": 0.18795724660158158,
      "num_tokens": 24892944.0,
      "step": 11160
    },
    {
      "entropy": 5.504327917098999,
      "epoch": 1.2539450777784018,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004853595930183822,
      "loss": 5.3644,
      "mean_token_accuracy": 0.1891561046242714,
      "num_tokens": 24903346.0,
      "step": 11165
    },
    {
      "entropy": 5.45168251991272,
      "epoch": 1.2545066546863592,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004853453437721901,
      "loss": 5.338,
      "mean_token_accuracy": 0.18961320370435714,
      "num_tokens": 24913904.0,
      "step": 11170
    },
    {
      "entropy": 5.454988813400268,
      "epoch": 1.255068231594317,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004853310878285208,
      "loss": 5.2918,
      "mean_token_accuracy": 0.19362047165632248,
      "num_tokens": 24924635.0,
      "step": 11175
    },
    {
      "entropy": 5.553615617752075,
      "epoch": 1.2556298085022743,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048531682518782797,
      "loss": 5.4802,
      "mean_token_accuracy": 0.18075439780950547,
      "num_tokens": 24935466.0,
      "step": 11180
    },
    {
      "entropy": 5.584972953796386,
      "epoch": 1.256191385410232,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048530255585056594,
      "loss": 5.4355,
      "mean_token_accuracy": 0.18358389437198638,
      "num_tokens": 24948348.0,
      "step": 11185
    },
    {
      "entropy": 5.489812278747559,
      "epoch": 1.2567529623181894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004852882798171889,
      "loss": 5.2809,
      "mean_token_accuracy": 0.19066839814186096,
      "num_tokens": 24959242.0,
      "step": 11190
    },
    {
      "entropy": 5.447453498840332,
      "epoch": 1.2573145392261469,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004852739970881515,
      "loss": 5.2568,
      "mean_token_accuracy": 0.19476295560598372,
      "num_tokens": 24969855.0,
      "step": 11195
    },
    {
      "entropy": 5.429177618026733,
      "epoch": 1.2578761161341045,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048525970766390847,
      "loss": 5.2817,
      "mean_token_accuracy": 0.19262396395206452,
      "num_tokens": 24980378.0,
      "step": 11200
    },
    {
      "entropy": 5.509761619567871,
      "epoch": 1.2584376930420622,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004852454115449148,
      "loss": 5.3507,
      "mean_token_accuracy": 0.18462255895137786,
      "num_tokens": 24990646.0,
      "step": 11205
    },
    {
      "entropy": 5.61767258644104,
      "epoch": 1.2589992699500197,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004852311087316257,
      "loss": 5.4643,
      "mean_token_accuracy": 0.1814809709787369,
      "num_tokens": 25002644.0,
      "step": 11210
    },
    {
      "entropy": 5.524403715133667,
      "epoch": 1.259560846857977,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004852167992244966,
      "loss": 5.3586,
      "mean_token_accuracy": 0.1936364233493805,
      "num_tokens": 25012635.0,
      "step": 11215
    },
    {
      "entropy": 5.49802942276001,
      "epoch": 1.2601224237659348,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000485202483023983,
      "loss": 5.4706,
      "mean_token_accuracy": 0.18091963231563568,
      "num_tokens": 25024289.0,
      "step": 11220
    },
    {
      "entropy": 5.516169881820678,
      "epoch": 1.2606840006738924,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048518816013054083,
      "loss": 5.2383,
      "mean_token_accuracy": 0.18986262381076813,
      "num_tokens": 25035316.0,
      "step": 11225
    },
    {
      "entropy": 5.516177320480347,
      "epoch": 1.2612455775818499,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048517383054462617,
      "loss": 5.3003,
      "mean_token_accuracy": 0.19051390886306763,
      "num_tokens": 25046011.0,
      "step": 11230
    },
    {
      "entropy": 5.484417676925659,
      "epoch": 1.2618071544898073,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004851594942666952,
      "loss": 5.2965,
      "mean_token_accuracy": 0.1923714891076088,
      "num_tokens": 25057402.0,
      "step": 11235
    },
    {
      "entropy": 5.447501087188721,
      "epoch": 1.262368731397765,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048514515129720445,
      "loss": 5.2949,
      "mean_token_accuracy": 0.18824667632579803,
      "num_tokens": 25068287.0,
      "step": 11240
    },
    {
      "entropy": 5.436086130142212,
      "epoch": 1.2629303083057224,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004851308016366105,
      "loss": 5.2727,
      "mean_token_accuracy": 0.19113888293504716,
      "num_tokens": 25079534.0,
      "step": 11245
    },
    {
      "entropy": 5.405421876907349,
      "epoch": 1.26349188521368,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048511644528537043,
      "loss": 5.2436,
      "mean_token_accuracy": 0.19242125302553176,
      "num_tokens": 25090825.0,
      "step": 11250
    },
    {
      "entropy": 5.475717782974243,
      "epoch": 1.2640534621216375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004851020822439412,
      "loss": 5.3484,
      "mean_token_accuracy": 0.18806565552949905,
      "num_tokens": 25101859.0,
      "step": 11255
    },
    {
      "entropy": 5.460765314102173,
      "epoch": 1.2646150390295952,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048508771251278015,
      "loss": 5.2593,
      "mean_token_accuracy": 0.19475947916507722,
      "num_tokens": 25111416.0,
      "step": 11260
    },
    {
      "entropy": 5.496658134460449,
      "epoch": 1.2651766159375526,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004850733360923449,
      "loss": 5.3603,
      "mean_token_accuracy": 0.1914309784770012,
      "num_tokens": 25122129.0,
      "step": 11265
    },
    {
      "entropy": 5.514855623245239,
      "epoch": 1.2657381928455103,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004850589529830931,
      "loss": 5.3453,
      "mean_token_accuracy": 0.1911257103085518,
      "num_tokens": 25132429.0,
      "step": 11270
    },
    {
      "entropy": 5.402274036407471,
      "epoch": 1.2662997697534677,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048504456318548285,
      "loss": 5.383,
      "mean_token_accuracy": 0.19443860203027724,
      "num_tokens": 25144242.0,
      "step": 11275
    },
    {
      "entropy": 5.562759017944336,
      "epoch": 1.2668613466614254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004850301666999722,
      "loss": 5.3843,
      "mean_token_accuracy": 0.19072728753089904,
      "num_tokens": 25155332.0,
      "step": 11280
    },
    {
      "entropy": 5.5600745677948,
      "epoch": 1.2674229235693828,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004850157635270196,
      "loss": 5.3138,
      "mean_token_accuracy": 0.18983711004257203,
      "num_tokens": 25166024.0,
      "step": 11285
    },
    {
      "entropy": 5.466601705551147,
      "epoch": 1.2679845004773402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004850013536670836,
      "loss": 5.254,
      "mean_token_accuracy": 0.19609452188014984,
      "num_tokens": 25176697.0,
      "step": 11290
    },
    {
      "entropy": 5.4619403839111325,
      "epoch": 1.268546077385298,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004849869371206231,
      "loss": 5.3584,
      "mean_token_accuracy": 0.19177097380161284,
      "num_tokens": 25187421.0,
      "step": 11295
    },
    {
      "entropy": 5.467381525039673,
      "epoch": 1.2691076542932556,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048497251388809697,
      "loss": 5.3589,
      "mean_token_accuracy": 0.1935126945376396,
      "num_tokens": 25198851.0,
      "step": 11300
    },
    {
      "entropy": 5.543435716629029,
      "epoch": 1.269669231201213,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004849580839699647,
      "loss": 5.406,
      "mean_token_accuracy": 0.1843758761882782,
      "num_tokens": 25209984.0,
      "step": 11305
    },
    {
      "entropy": 5.54110951423645,
      "epoch": 1.2702308081091704,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048494364736668554,
      "loss": 5.2978,
      "mean_token_accuracy": 0.18914757221937178,
      "num_tokens": 25220672.0,
      "step": 11310
    },
    {
      "entropy": 5.520363998413086,
      "epoch": 1.270792385017128,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004849292040787192,
      "loss": 5.3291,
      "mean_token_accuracy": 0.18871739655733108,
      "num_tokens": 25231906.0,
      "step": 11315
    },
    {
      "entropy": 5.525923109054565,
      "epoch": 1.2713539619250858,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004849147541065256,
      "loss": 5.3508,
      "mean_token_accuracy": 0.18501406908035278,
      "num_tokens": 25242902.0,
      "step": 11320
    },
    {
      "entropy": 5.501807594299317,
      "epoch": 1.2719155388330432,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004849002974505649,
      "loss": 5.4056,
      "mean_token_accuracy": 0.18155012279748917,
      "num_tokens": 25254363.0,
      "step": 11325
    },
    {
      "entropy": 5.500497674942016,
      "epoch": 1.2724771157410006,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048488583411129723,
      "loss": 5.3484,
      "mean_token_accuracy": 0.1882040560245514,
      "num_tokens": 25267108.0,
      "step": 11330
    },
    {
      "entropy": 5.43377833366394,
      "epoch": 1.2730386926489583,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004848713640891832,
      "loss": 5.2984,
      "mean_token_accuracy": 0.19312814027070999,
      "num_tokens": 25278583.0,
      "step": 11335
    },
    {
      "entropy": 5.571519756317139,
      "epoch": 1.2736002695569157,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004848568873846835,
      "loss": 5.3909,
      "mean_token_accuracy": 0.1879279851913452,
      "num_tokens": 25289813.0,
      "step": 11340
    },
    {
      "entropy": 5.465385484695434,
      "epoch": 1.2741618464648734,
      "grad_norm": 1.125,
      "learning_rate": 0.00048484240399825925,
      "loss": 5.2833,
      "mean_token_accuracy": 0.19203702062368394,
      "num_tokens": 25301003.0,
      "step": 11345
    },
    {
      "entropy": 5.427354907989502,
      "epoch": 1.2747234233728308,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048482791393037134,
      "loss": 5.2768,
      "mean_token_accuracy": 0.19592732936143875,
      "num_tokens": 25312535.0,
      "step": 11350
    },
    {
      "entropy": 5.470835161209107,
      "epoch": 1.2752850002807885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048481341718148137,
      "loss": 5.3309,
      "mean_token_accuracy": 0.18613635152578353,
      "num_tokens": 25323333.0,
      "step": 11355
    },
    {
      "entropy": 5.600289726257325,
      "epoch": 1.275846577188746,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048479891375205074,
      "loss": 5.4115,
      "mean_token_accuracy": 0.1884475588798523,
      "num_tokens": 25334865.0,
      "step": 11360
    },
    {
      "entropy": 5.5824517726898195,
      "epoch": 1.2764081540967036,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048478440364254136,
      "loss": 5.3916,
      "mean_token_accuracy": 0.18760250210762025,
      "num_tokens": 25346672.0,
      "step": 11365
    },
    {
      "entropy": 5.489925146102905,
      "epoch": 1.276969731004661,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004847698868534152,
      "loss": 5.3809,
      "mean_token_accuracy": 0.1921190619468689,
      "num_tokens": 25358702.0,
      "step": 11370
    },
    {
      "entropy": 5.46632194519043,
      "epoch": 1.2775313079126187,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004847553633851345,
      "loss": 5.3192,
      "mean_token_accuracy": 0.18536871969699859,
      "num_tokens": 25369813.0,
      "step": 11375
    },
    {
      "entropy": 5.527243518829346,
      "epoch": 1.2780928848205761,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004847408332381616,
      "loss": 5.3592,
      "mean_token_accuracy": 0.18779368549585343,
      "num_tokens": 25381042.0,
      "step": 11380
    },
    {
      "entropy": 5.503709030151367,
      "epoch": 1.2786544617285336,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048472629641295933,
      "loss": 5.3471,
      "mean_token_accuracy": 0.18134421855211258,
      "num_tokens": 25391286.0,
      "step": 11385
    },
    {
      "entropy": 5.471360301971435,
      "epoch": 1.2792160386364912,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004847117529099904,
      "loss": 5.2919,
      "mean_token_accuracy": 0.19129288643598558,
      "num_tokens": 25402335.0,
      "step": 11390
    },
    {
      "entropy": 5.434795999526978,
      "epoch": 1.279777615544449,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048469720272971797,
      "loss": 5.2483,
      "mean_token_accuracy": 0.1946100503206253,
      "num_tokens": 25413484.0,
      "step": 11395
    },
    {
      "entropy": 5.5469069480896,
      "epoch": 1.2803391924524063,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004846826458726052,
      "loss": 5.3572,
      "mean_token_accuracy": 0.18792201727628707,
      "num_tokens": 25424276.0,
      "step": 11400
    },
    {
      "entropy": 5.342579078674317,
      "epoch": 1.2809007693603638,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048466808233911567,
      "loss": 5.235,
      "mean_token_accuracy": 0.19518037289381027,
      "num_tokens": 25435178.0,
      "step": 11405
    },
    {
      "entropy": 5.389122056961059,
      "epoch": 1.2814623462683215,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048465351212971313,
      "loss": 5.199,
      "mean_token_accuracy": 0.19640120565891267,
      "num_tokens": 25445792.0,
      "step": 11410
    },
    {
      "entropy": 5.437396717071533,
      "epoch": 1.2820239231762791,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004846389352448614,
      "loss": 5.2889,
      "mean_token_accuracy": 0.19388676136732103,
      "num_tokens": 25456598.0,
      "step": 11415
    },
    {
      "entropy": 5.45359468460083,
      "epoch": 1.2825855000842366,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004846243516850247,
      "loss": 5.3734,
      "mean_token_accuracy": 0.18930955827236176,
      "num_tokens": 25468311.0,
      "step": 11420
    },
    {
      "entropy": 5.552885103225708,
      "epoch": 1.283147076992194,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004846097614506673,
      "loss": 5.3832,
      "mean_token_accuracy": 0.18507194221019746,
      "num_tokens": 25480680.0,
      "step": 11425
    },
    {
      "entropy": 5.604588317871094,
      "epoch": 1.2837086539001517,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048459516454225384,
      "loss": 5.3525,
      "mean_token_accuracy": 0.19158392101526261,
      "num_tokens": 25492255.0,
      "step": 11430
    },
    {
      "entropy": 5.481392669677734,
      "epoch": 1.284270230808109,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000484580560960249,
      "loss": 5.3929,
      "mean_token_accuracy": 0.18605994880199433,
      "num_tokens": 25504530.0,
      "step": 11435
    },
    {
      "entropy": 5.42602310180664,
      "epoch": 1.2848318077160668,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004845659507051178,
      "loss": 5.3281,
      "mean_token_accuracy": 0.19711543321609498,
      "num_tokens": 25515273.0,
      "step": 11440
    },
    {
      "entropy": 5.551933288574219,
      "epoch": 1.2853933846240242,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004845513337773255,
      "loss": 5.4505,
      "mean_token_accuracy": 0.18545989394187928,
      "num_tokens": 25526050.0,
      "step": 11445
    },
    {
      "entropy": 5.565883159637451,
      "epoch": 1.2859549615319819,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048453671017733755,
      "loss": 5.3579,
      "mean_token_accuracy": 0.186768040060997,
      "num_tokens": 25538372.0,
      "step": 11450
    },
    {
      "entropy": 5.620044374465943,
      "epoch": 1.2865165384399393,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004845220799056194,
      "loss": 5.442,
      "mean_token_accuracy": 0.1926871195435524,
      "num_tokens": 25549636.0,
      "step": 11455
    },
    {
      "entropy": 5.497498512268066,
      "epoch": 1.287078115347897,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048450744296263685,
      "loss": 5.3677,
      "mean_token_accuracy": 0.18753843009471893,
      "num_tokens": 25560850.0,
      "step": 11460
    },
    {
      "entropy": 5.476309728622437,
      "epoch": 1.2876396922558544,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048449279934885625,
      "loss": 5.3039,
      "mean_token_accuracy": 0.19111284762620925,
      "num_tokens": 25572433.0,
      "step": 11465
    },
    {
      "entropy": 5.607483530044556,
      "epoch": 1.288201269163812,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004844781490647435,
      "loss": 5.3729,
      "mean_token_accuracy": 0.1838137537240982,
      "num_tokens": 25582372.0,
      "step": 11470
    },
    {
      "entropy": 5.4621470928192135,
      "epoch": 1.2887628460717695,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004844634921107654,
      "loss": 5.2084,
      "mean_token_accuracy": 0.1927051931619644,
      "num_tokens": 25592570.0,
      "step": 11475
    },
    {
      "entropy": 5.490028619766235,
      "epoch": 1.289324422979727,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048444882848738835,
      "loss": 5.3351,
      "mean_token_accuracy": 0.1845156028866768,
      "num_tokens": 25603573.0,
      "step": 11480
    },
    {
      "entropy": 5.530058574676514,
      "epoch": 1.2898859998876846,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048443415819507944,
      "loss": 5.3641,
      "mean_token_accuracy": 0.18942673951387407,
      "num_tokens": 25614233.0,
      "step": 11485
    },
    {
      "entropy": 5.527749490737915,
      "epoch": 1.2904475767956423,
      "grad_norm": 1.25,
      "learning_rate": 0.0004844194812343057,
      "loss": 5.3323,
      "mean_token_accuracy": 0.1844804048538208,
      "num_tokens": 25625077.0,
      "step": 11490
    },
    {
      "entropy": 5.434228038787841,
      "epoch": 1.2910091537035997,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004844047976055344,
      "loss": 5.3518,
      "mean_token_accuracy": 0.18301459401845932,
      "num_tokens": 25636693.0,
      "step": 11495
    },
    {
      "entropy": 5.331040525436402,
      "epoch": 1.2915707306115571,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048439010730923313,
      "loss": 5.2095,
      "mean_token_accuracy": 0.19238260984420777,
      "num_tokens": 25647644.0,
      "step": 11500
    },
    {
      "entropy": 5.435154771804809,
      "epoch": 1.2921323075195148,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048437541034586964,
      "loss": 5.2446,
      "mean_token_accuracy": 0.19497229605913163,
      "num_tokens": 25659373.0,
      "step": 11505
    },
    {
      "entropy": 5.556627988815308,
      "epoch": 1.2926938844274725,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048436070671591195,
      "loss": 5.3799,
      "mean_token_accuracy": 0.19794285893440247,
      "num_tokens": 25670479.0,
      "step": 11510
    },
    {
      "entropy": 5.544531679153442,
      "epoch": 1.29325546133543,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048434599641982806,
      "loss": 5.3877,
      "mean_token_accuracy": 0.1912296235561371,
      "num_tokens": 25682748.0,
      "step": 11515
    },
    {
      "entropy": 5.4173712730407715,
      "epoch": 1.2938170382433873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048433127945808645,
      "loss": 5.2218,
      "mean_token_accuracy": 0.19448208659887314,
      "num_tokens": 25693170.0,
      "step": 11520
    },
    {
      "entropy": 5.427676868438721,
      "epoch": 1.294378615151345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004843165558311558,
      "loss": 5.3497,
      "mean_token_accuracy": 0.1826302319765091,
      "num_tokens": 25706009.0,
      "step": 11525
    },
    {
      "entropy": 5.553053140640259,
      "epoch": 1.2949401920593024,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004843018255395048,
      "loss": 5.3782,
      "mean_token_accuracy": 0.19116197675466537,
      "num_tokens": 25716036.0,
      "step": 11530
    },
    {
      "entropy": 5.505878114700318,
      "epoch": 1.29550176896726,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048428708858360244,
      "loss": 5.1892,
      "mean_token_accuracy": 0.19445182532072067,
      "num_tokens": 25726516.0,
      "step": 11535
    },
    {
      "entropy": 5.4927653789520265,
      "epoch": 1.2960633458752175,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048427234496391816,
      "loss": 5.345,
      "mean_token_accuracy": 0.18722363859415053,
      "num_tokens": 25736340.0,
      "step": 11540
    },
    {
      "entropy": 5.356808090209961,
      "epoch": 1.2966249227831752,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048425759468092113,
      "loss": 5.2995,
      "mean_token_accuracy": 0.18733727484941481,
      "num_tokens": 25746439.0,
      "step": 11545
    },
    {
      "entropy": 5.466232585906982,
      "epoch": 1.2971864996911326,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004842428377350811,
      "loss": 5.2113,
      "mean_token_accuracy": 0.193405120074749,
      "num_tokens": 25757065.0,
      "step": 11550
    },
    {
      "entropy": 5.494842433929444,
      "epoch": 1.2977480765990903,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048422807412686796,
      "loss": 5.2876,
      "mean_token_accuracy": 0.18820154219865798,
      "num_tokens": 25768021.0,
      "step": 11555
    },
    {
      "entropy": 5.496604585647583,
      "epoch": 1.2983096535070477,
      "grad_norm": 1.125,
      "learning_rate": 0.0004842133038567519,
      "loss": 5.39,
      "mean_token_accuracy": 0.19395477026700975,
      "num_tokens": 25780197.0,
      "step": 11560
    },
    {
      "entropy": 5.477793979644775,
      "epoch": 1.2988712304150054,
      "grad_norm": 1.125,
      "learning_rate": 0.000484198526925203,
      "loss": 5.3126,
      "mean_token_accuracy": 0.18827086836099624,
      "num_tokens": 25790536.0,
      "step": 11565
    },
    {
      "entropy": 5.445866107940674,
      "epoch": 1.2994328073229628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004841837433326919,
      "loss": 5.276,
      "mean_token_accuracy": 0.19228814989328386,
      "num_tokens": 25802397.0,
      "step": 11570
    },
    {
      "entropy": 5.501689720153808,
      "epoch": 1.2999943842309205,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004841689530796893,
      "loss": 5.3892,
      "mean_token_accuracy": 0.19443353563547133,
      "num_tokens": 25813961.0,
      "step": 11575
    },
    {
      "entropy": 5.5217243194580075,
      "epoch": 1.300555961138878,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004841541561666661,
      "loss": 5.4093,
      "mean_token_accuracy": 0.18545944094657899,
      "num_tokens": 25825610.0,
      "step": 11580
    },
    {
      "entropy": 5.471556091308594,
      "epoch": 1.3011175380468356,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048413935259409343,
      "loss": 5.2128,
      "mean_token_accuracy": 0.19547210782766342,
      "num_tokens": 25836076.0,
      "step": 11585
    },
    {
      "entropy": 5.4894482612609865,
      "epoch": 1.301679114954793,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048412454236244267,
      "loss": 5.3344,
      "mean_token_accuracy": 0.19530290961265565,
      "num_tokens": 25847341.0,
      "step": 11590
    },
    {
      "entropy": 5.3722395420074465,
      "epoch": 1.3022406918627505,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048410972547218545,
      "loss": 5.2376,
      "mean_token_accuracy": 0.1964733585715294,
      "num_tokens": 25856912.0,
      "step": 11595
    },
    {
      "entropy": 5.431223630905151,
      "epoch": 1.3028022687707081,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048409490192379337,
      "loss": 5.2595,
      "mean_token_accuracy": 0.18715425878763198,
      "num_tokens": 25868370.0,
      "step": 11600
    },
    {
      "entropy": 5.4193699836730955,
      "epoch": 1.3033638456786658,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004840800717177386,
      "loss": 5.2944,
      "mean_token_accuracy": 0.19211488366127014,
      "num_tokens": 25879104.0,
      "step": 11605
    },
    {
      "entropy": 5.535993194580078,
      "epoch": 1.3039254225866233,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004840652348544931,
      "loss": 5.3484,
      "mean_token_accuracy": 0.1839719831943512,
      "num_tokens": 25889744.0,
      "step": 11610
    },
    {
      "entropy": 5.553609228134155,
      "epoch": 1.3044869994945807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004840503913345296,
      "loss": 5.4663,
      "mean_token_accuracy": 0.1850156307220459,
      "num_tokens": 25901615.0,
      "step": 11615
    },
    {
      "entropy": 5.526863670349121,
      "epoch": 1.3050485764025384,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004840355411583205,
      "loss": 5.3245,
      "mean_token_accuracy": 0.18772584199905396,
      "num_tokens": 25911418.0,
      "step": 11620
    },
    {
      "entropy": 5.549993276596069,
      "epoch": 1.3056101533104958,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048402068432633864,
      "loss": 5.4049,
      "mean_token_accuracy": 0.19036079645156861,
      "num_tokens": 25922141.0,
      "step": 11625
    },
    {
      "entropy": 5.539162349700928,
      "epoch": 1.3061717302184535,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004840058208390572,
      "loss": 5.4201,
      "mean_token_accuracy": 0.18496872782707213,
      "num_tokens": 25933897.0,
      "step": 11630
    },
    {
      "entropy": 5.5644574642181395,
      "epoch": 1.306733307126411,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004839909506969494,
      "loss": 5.3437,
      "mean_token_accuracy": 0.18744596987962722,
      "num_tokens": 25943960.0,
      "step": 11635
    },
    {
      "entropy": 5.605818891525269,
      "epoch": 1.3072948840343686,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004839760739004886,
      "loss": 5.3931,
      "mean_token_accuracy": 0.18857479095458984,
      "num_tokens": 25955104.0,
      "step": 11640
    },
    {
      "entropy": 5.493829298019409,
      "epoch": 1.307856460942326,
      "grad_norm": 1.125,
      "learning_rate": 0.00048396119045014853,
      "loss": 5.3311,
      "mean_token_accuracy": 0.193758088350296,
      "num_tokens": 25965180.0,
      "step": 11645
    },
    {
      "entropy": 5.444762754440307,
      "epoch": 1.3084180378502837,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004839463003464032,
      "loss": 5.3313,
      "mean_token_accuracy": 0.19077154099941254,
      "num_tokens": 25976600.0,
      "step": 11650
    },
    {
      "entropy": 5.396129751205445,
      "epoch": 1.308979614758241,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004839314035897265,
      "loss": 5.3005,
      "mean_token_accuracy": 0.18808194547891616,
      "num_tokens": 25987509.0,
      "step": 11655
    },
    {
      "entropy": 5.474160957336426,
      "epoch": 1.3095411916661988,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000483916500180593,
      "loss": 5.242,
      "mean_token_accuracy": 0.19153997898101807,
      "num_tokens": 25997962.0,
      "step": 11660
    },
    {
      "entropy": 5.487403345108032,
      "epoch": 1.3101027685741562,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048390159011947706,
      "loss": 5.3025,
      "mean_token_accuracy": 0.1925146296620369,
      "num_tokens": 26008932.0,
      "step": 11665
    },
    {
      "entropy": 5.488673543930053,
      "epoch": 1.3106643454821139,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004838866734068535,
      "loss": 5.3174,
      "mean_token_accuracy": 0.1836942106485367,
      "num_tokens": 26020513.0,
      "step": 11670
    },
    {
      "entropy": 5.565633153915405,
      "epoch": 1.3112259223900713,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048387175004319723,
      "loss": 5.373,
      "mean_token_accuracy": 0.18528845757246018,
      "num_tokens": 26032526.0,
      "step": 11675
    },
    {
      "entropy": 5.514460468292237,
      "epoch": 1.311787499298029,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004838568200289834,
      "loss": 5.3143,
      "mean_token_accuracy": 0.19298763275146485,
      "num_tokens": 26044216.0,
      "step": 11680
    },
    {
      "entropy": 5.417226934432984,
      "epoch": 1.3123490762059864,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004838418833646875,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18900535702705384,
      "num_tokens": 26054012.0,
      "step": 11685
    },
    {
      "entropy": 5.4516205310821535,
      "epoch": 1.3129106531139438,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048382694005078485,
      "loss": 5.3687,
      "mean_token_accuracy": 0.18900944739580156,
      "num_tokens": 26065365.0,
      "step": 11690
    },
    {
      "entropy": 5.529718637466431,
      "epoch": 1.3134722300219015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048381199008775165,
      "loss": 5.3712,
      "mean_token_accuracy": 0.1843010053038597,
      "num_tokens": 26075964.0,
      "step": 11695
    },
    {
      "entropy": 5.4653685092926025,
      "epoch": 1.3140338069298592,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004837970334760635,
      "loss": 5.3094,
      "mean_token_accuracy": 0.1927257478237152,
      "num_tokens": 26087572.0,
      "step": 11700
    },
    {
      "entropy": 5.428178977966309,
      "epoch": 1.3145953838378166,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000483782070216197,
      "loss": 5.2313,
      "mean_token_accuracy": 0.1969181165099144,
      "num_tokens": 26098612.0,
      "step": 11705
    },
    {
      "entropy": 5.512505435943604,
      "epoch": 1.315156960745774,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004837671003086283,
      "loss": 5.3931,
      "mean_token_accuracy": 0.19104406088590623,
      "num_tokens": 26109491.0,
      "step": 11710
    },
    {
      "entropy": 5.535184383392334,
      "epoch": 1.3157185376537317,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004837521237538342,
      "loss": 5.2866,
      "mean_token_accuracy": 0.19219576567411423,
      "num_tokens": 26120544.0,
      "step": 11715
    },
    {
      "entropy": 5.515174674987793,
      "epoch": 1.3162801145616891,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004837371405522915,
      "loss": 5.3864,
      "mean_token_accuracy": 0.1831790253520012,
      "num_tokens": 26131549.0,
      "step": 11720
    },
    {
      "entropy": 5.464992904663086,
      "epoch": 1.3168416914696468,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048372215070447734,
      "loss": 5.3755,
      "mean_token_accuracy": 0.19335275739431382,
      "num_tokens": 26142871.0,
      "step": 11725
    },
    {
      "entropy": 5.580825424194336,
      "epoch": 1.3174032683776042,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048370715421086884,
      "loss": 5.4526,
      "mean_token_accuracy": 0.189836722612381,
      "num_tokens": 26155293.0,
      "step": 11730
    },
    {
      "entropy": 5.571844244003296,
      "epoch": 1.317964845285562,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004836921510719438,
      "loss": 5.3931,
      "mean_token_accuracy": 0.19178189486265182,
      "num_tokens": 26167069.0,
      "step": 11735
    },
    {
      "entropy": 5.498572158813476,
      "epoch": 1.3185264221935193,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004836771412881795,
      "loss": 5.3932,
      "mean_token_accuracy": 0.18102272152900695,
      "num_tokens": 26179142.0,
      "step": 11740
    },
    {
      "entropy": 5.5183288097381595,
      "epoch": 1.319087999101477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004836621248600542,
      "loss": 5.3208,
      "mean_token_accuracy": 0.1880206748843193,
      "num_tokens": 26190059.0,
      "step": 11745
    },
    {
      "entropy": 5.520627403259278,
      "epoch": 1.3196495760094344,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004836471017880459,
      "loss": 5.3416,
      "mean_token_accuracy": 0.1924460709095001,
      "num_tokens": 26199859.0,
      "step": 11750
    },
    {
      "entropy": 5.497109222412109,
      "epoch": 1.320211152917392,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004836320720726329,
      "loss": 5.3722,
      "mean_token_accuracy": 0.18477653115987777,
      "num_tokens": 26211920.0,
      "step": 11755
    },
    {
      "entropy": 5.493609380722046,
      "epoch": 1.3207727298253495,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004836170357142939,
      "loss": 5.3307,
      "mean_token_accuracy": 0.18585283011198045,
      "num_tokens": 26222875.0,
      "step": 11760
    },
    {
      "entropy": 5.530766820907592,
      "epoch": 1.3213343067333072,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004836019927135076,
      "loss": 5.3533,
      "mean_token_accuracy": 0.18562915027141572,
      "num_tokens": 26233549.0,
      "step": 11765
    },
    {
      "entropy": 5.429447841644287,
      "epoch": 1.3218958836412646,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004835869430707529,
      "loss": 5.2386,
      "mean_token_accuracy": 0.20302475094795228,
      "num_tokens": 26243886.0,
      "step": 11770
    },
    {
      "entropy": 5.40765905380249,
      "epoch": 1.3224574605492223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048357188678650905,
      "loss": 5.28,
      "mean_token_accuracy": 0.19504840672016144,
      "num_tokens": 26254475.0,
      "step": 11775
    },
    {
      "entropy": 5.541731977462769,
      "epoch": 1.3230190374571797,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004835568238612553,
      "loss": 5.4519,
      "mean_token_accuracy": 0.18699074387550355,
      "num_tokens": 26265412.0,
      "step": 11780
    },
    {
      "entropy": 5.616350173950195,
      "epoch": 1.3235806143651372,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048354175429547155,
      "loss": 5.3941,
      "mean_token_accuracy": 0.18800213932991028,
      "num_tokens": 26276942.0,
      "step": 11785
    },
    {
      "entropy": 5.531629276275635,
      "epoch": 1.3241421912730948,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004835266780896374,
      "loss": 5.3516,
      "mean_token_accuracy": 0.18680412024259568,
      "num_tokens": 26288009.0,
      "step": 11790
    },
    {
      "entropy": 5.51987886428833,
      "epoch": 1.3247037681810525,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004835115952442329,
      "loss": 5.5002,
      "mean_token_accuracy": 0.17894561290740968,
      "num_tokens": 26301583.0,
      "step": 11795
    },
    {
      "entropy": 5.576613235473633,
      "epoch": 1.32526534508901,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048349650575973843,
      "loss": 5.3854,
      "mean_token_accuracy": 0.19318455308675767,
      "num_tokens": 26312922.0,
      "step": 11800
    },
    {
      "entropy": 5.575756168365478,
      "epoch": 1.3258269219969674,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048348140963663436,
      "loss": 5.3503,
      "mean_token_accuracy": 0.19539692103862763,
      "num_tokens": 26324108.0,
      "step": 11805
    },
    {
      "entropy": 5.520961284637451,
      "epoch": 1.326388498904925,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004834663068754012,
      "loss": 5.3991,
      "mean_token_accuracy": 0.18515405729413031,
      "num_tokens": 26335669.0,
      "step": 11810
    },
    {
      "entropy": 5.383474254608155,
      "epoch": 1.3269500758128827,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004834511974765202,
      "loss": 5.2692,
      "mean_token_accuracy": 0.1946129947900772,
      "num_tokens": 26346065.0,
      "step": 11815
    },
    {
      "entropy": 5.486476135253906,
      "epoch": 1.3275116527208402,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048343608144047194,
      "loss": 5.2465,
      "mean_token_accuracy": 0.1924546480178833,
      "num_tokens": 26357567.0,
      "step": 11820
    },
    {
      "entropy": 5.472987461090088,
      "epoch": 1.3280732296287976,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048342095876773825,
      "loss": 5.3009,
      "mean_token_accuracy": 0.18726590871810914,
      "num_tokens": 26367805.0,
      "step": 11825
    },
    {
      "entropy": 5.41779203414917,
      "epoch": 1.3286348065367553,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004834058294588003,
      "loss": 5.2161,
      "mean_token_accuracy": 0.19460426717996598,
      "num_tokens": 26379343.0,
      "step": 11830
    },
    {
      "entropy": 5.520466279983521,
      "epoch": 1.3291963834447127,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048339069351413993,
      "loss": 5.4277,
      "mean_token_accuracy": 0.18180939704179763,
      "num_tokens": 26392039.0,
      "step": 11835
    },
    {
      "entropy": 5.365854406356812,
      "epoch": 1.3297579603526704,
      "grad_norm": 1.0,
      "learning_rate": 0.000483375550934239,
      "loss": 5.2016,
      "mean_token_accuracy": 0.20187190175056458,
      "num_tokens": 26403324.0,
      "step": 11840
    },
    {
      "entropy": 5.469034719467163,
      "epoch": 1.3303195372606278,
      "grad_norm": 1.125,
      "learning_rate": 0.0004833604017195798,
      "loss": 5.3181,
      "mean_token_accuracy": 0.1933395341038704,
      "num_tokens": 26413435.0,
      "step": 11845
    },
    {
      "entropy": 5.518855142593384,
      "epoch": 1.3308811141685855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004833452458706445,
      "loss": 5.3947,
      "mean_token_accuracy": 0.18085718899965286,
      "num_tokens": 26424254.0,
      "step": 11850
    },
    {
      "entropy": 5.524479627609253,
      "epoch": 1.331442691076543,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004833300833879159,
      "loss": 5.3309,
      "mean_token_accuracy": 0.19006485342979432,
      "num_tokens": 26435658.0,
      "step": 11855
    },
    {
      "entropy": 5.539576292037964,
      "epoch": 1.3320042679845006,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004833149142718765,
      "loss": 5.373,
      "mean_token_accuracy": 0.1878064513206482,
      "num_tokens": 26447174.0,
      "step": 11860
    },
    {
      "entropy": 5.55002121925354,
      "epoch": 1.332565844892458,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004832997385230096,
      "loss": 5.4465,
      "mean_token_accuracy": 0.1872912123799324,
      "num_tokens": 26459303.0,
      "step": 11865
    },
    {
      "entropy": 5.508304023742676,
      "epoch": 1.3331274218004157,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004832845561417981,
      "loss": 5.323,
      "mean_token_accuracy": 0.18582605868577956,
      "num_tokens": 26469981.0,
      "step": 11870
    },
    {
      "entropy": 5.49251012802124,
      "epoch": 1.333688998708373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004832693671287256,
      "loss": 5.3218,
      "mean_token_accuracy": 0.19587670117616654,
      "num_tokens": 26481176.0,
      "step": 11875
    },
    {
      "entropy": 5.430894279479981,
      "epoch": 1.3342505756163305,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004832541714842757,
      "loss": 5.2728,
      "mean_token_accuracy": 0.19216970801353456,
      "num_tokens": 26492795.0,
      "step": 11880
    },
    {
      "entropy": 5.539825582504273,
      "epoch": 1.3348121525242882,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004832389692089321,
      "loss": 5.4408,
      "mean_token_accuracy": 0.1842477634549141,
      "num_tokens": 26504731.0,
      "step": 11885
    },
    {
      "entropy": 5.599880743026733,
      "epoch": 1.3353737294322459,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048322376030317905,
      "loss": 5.3162,
      "mean_token_accuracy": 0.19063470065593718,
      "num_tokens": 26516816.0,
      "step": 11890
    },
    {
      "entropy": 5.533083200454712,
      "epoch": 1.3359353063402033,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004832085447675007,
      "loss": 5.4021,
      "mean_token_accuracy": 0.1902502804994583,
      "num_tokens": 26528861.0,
      "step": 11895
    },
    {
      "entropy": 5.542850351333618,
      "epoch": 1.3364968832481607,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048319332260238153,
      "loss": 5.3766,
      "mean_token_accuracy": 0.18914323598146437,
      "num_tokens": 26538816.0,
      "step": 11900
    },
    {
      "entropy": 5.593536853790283,
      "epoch": 1.3370584601561184,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048317809380830624,
      "loss": 5.4334,
      "mean_token_accuracy": 0.18843986690044404,
      "num_tokens": 26548959.0,
      "step": 11905
    },
    {
      "entropy": 5.429722595214844,
      "epoch": 1.337620037064076,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004831628583857597,
      "loss": 5.2787,
      "mean_token_accuracy": 0.19304779022932053,
      "num_tokens": 26559189.0,
      "step": 11910
    },
    {
      "entropy": 5.376905107498169,
      "epoch": 1.3381816139720335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048314761633522696,
      "loss": 5.2377,
      "mean_token_accuracy": 0.19619487822055817,
      "num_tokens": 26570066.0,
      "step": 11915
    },
    {
      "entropy": 5.4953361511230465,
      "epoch": 1.338743190879991,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004831323676571934,
      "loss": 5.4143,
      "mean_token_accuracy": 0.1912911430001259,
      "num_tokens": 26581489.0,
      "step": 11920
    },
    {
      "entropy": 5.55296745300293,
      "epoch": 1.3393047677879486,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004831171123521446,
      "loss": 5.3355,
      "mean_token_accuracy": 0.18815976679325103,
      "num_tokens": 26592051.0,
      "step": 11925
    },
    {
      "entropy": 5.568896102905273,
      "epoch": 1.339866344695906,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004831018504205662,
      "loss": 5.3349,
      "mean_token_accuracy": 0.18929506689310074,
      "num_tokens": 26602296.0,
      "step": 11930
    },
    {
      "entropy": 5.487189292907715,
      "epoch": 1.3404279216038637,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048308658186294414,
      "loss": 5.315,
      "mean_token_accuracy": 0.18514619916677474,
      "num_tokens": 26613837.0,
      "step": 11935
    },
    {
      "entropy": 5.446844005584717,
      "epoch": 1.3409894985118211,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004830713066797646,
      "loss": 5.3757,
      "mean_token_accuracy": 0.18660722821950912,
      "num_tokens": 26624898.0,
      "step": 11940
    },
    {
      "entropy": 5.501018333435058,
      "epoch": 1.3415510754197788,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048305602487151396,
      "loss": 5.359,
      "mean_token_accuracy": 0.19223895370960237,
      "num_tokens": 26636449.0,
      "step": 11945
    },
    {
      "entropy": 5.602386617660523,
      "epoch": 1.3421126523277362,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004830407364386789,
      "loss": 5.4509,
      "mean_token_accuracy": 0.18780952990055083,
      "num_tokens": 26649192.0,
      "step": 11950
    },
    {
      "entropy": 5.484031391143799,
      "epoch": 1.342674229235694,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000483025441381746,
      "loss": 5.2621,
      "mean_token_accuracy": 0.19373395442962646,
      "num_tokens": 26660005.0,
      "step": 11955
    },
    {
      "entropy": 5.59054012298584,
      "epoch": 1.3432358061436513,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048301013970120235,
      "loss": 5.4115,
      "mean_token_accuracy": 0.18633730560541154,
      "num_tokens": 26672165.0,
      "step": 11960
    },
    {
      "entropy": 5.466653633117676,
      "epoch": 1.343797383051609,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048299483139753517,
      "loss": 5.359,
      "mean_token_accuracy": 0.1914936989545822,
      "num_tokens": 26683135.0,
      "step": 11965
    },
    {
      "entropy": 5.461401414871216,
      "epoch": 1.3443589599595664,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004829795164712319,
      "loss": 5.241,
      "mean_token_accuracy": 0.1981671690940857,
      "num_tokens": 26693724.0,
      "step": 11970
    },
    {
      "entropy": 5.5157993793487545,
      "epoch": 1.3449205368675239,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048296419492278023,
      "loss": 5.4293,
      "mean_token_accuracy": 0.18887880593538284,
      "num_tokens": 26705880.0,
      "step": 11975
    },
    {
      "entropy": 5.547498750686645,
      "epoch": 1.3454821137754815,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004829488667526679,
      "loss": 5.419,
      "mean_token_accuracy": 0.19185321778059006,
      "num_tokens": 26716978.0,
      "step": 11980
    },
    {
      "entropy": 5.626371669769287,
      "epoch": 1.3460436906834392,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048293353196138294,
      "loss": 5.3592,
      "mean_token_accuracy": 0.19536844044923782,
      "num_tokens": 26728380.0,
      "step": 11985
    },
    {
      "entropy": 5.436957788467407,
      "epoch": 1.3466052675913966,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048291819054941375,
      "loss": 5.3076,
      "mean_token_accuracy": 0.1934093102812767,
      "num_tokens": 26739034.0,
      "step": 11990
    },
    {
      "entropy": 5.292926263809204,
      "epoch": 1.347166844499354,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004829028425172487,
      "loss": 5.114,
      "mean_token_accuracy": 0.2030069813132286,
      "num_tokens": 26750123.0,
      "step": 11995
    },
    {
      "entropy": 5.428149938583374,
      "epoch": 1.3477284214073117,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048288748786537656,
      "loss": 5.2874,
      "mean_token_accuracy": 0.20094300508499147,
      "num_tokens": 26760411.0,
      "step": 12000
    },
    {
      "epoch": 1.3477284214073117,
      "eval_entropy": 5.320394583513419,
      "eval_loss": 5.423473834991455,
      "eval_mean_token_accuracy": 0.1944776664912962,
      "eval_num_tokens": 26760411.0,
      "eval_runtime": 59.5728,
      "eval_samples_per_second": 520.59,
      "eval_steps_per_second": 65.08,
      "step": 12000
    },
    {
      "entropy": 5.478197383880615,
      "epoch": 1.3482899983152694,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004828721265942861,
      "loss": 5.3597,
      "mean_token_accuracy": 0.19468082040548323,
      "num_tokens": 26770821.0,
      "step": 12005
    },
    {
      "entropy": 5.5496132373809814,
      "epoch": 1.3488515752232269,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048285675870446653,
      "loss": 5.3429,
      "mean_token_accuracy": 0.19109699130058289,
      "num_tokens": 26782099.0,
      "step": 12010
    },
    {
      "entropy": 5.573636913299561,
      "epoch": 1.3494131521311843,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048284138419640724,
      "loss": 5.4507,
      "mean_token_accuracy": 0.18952461928129197,
      "num_tokens": 26794056.0,
      "step": 12015
    },
    {
      "entropy": 5.463382816314697,
      "epoch": 1.349974729039142,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004828260030705976,
      "loss": 5.3013,
      "mean_token_accuracy": 0.18759446889162062,
      "num_tokens": 26804501.0,
      "step": 12020
    },
    {
      "entropy": 5.553075218200684,
      "epoch": 1.3505363059470994,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048281061532752734,
      "loss": 5.3504,
      "mean_token_accuracy": 0.19351499825716018,
      "num_tokens": 26814996.0,
      "step": 12025
    },
    {
      "entropy": 5.499054431915283,
      "epoch": 1.351097882855057,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004827952209676866,
      "loss": 5.2946,
      "mean_token_accuracy": 0.1943981885910034,
      "num_tokens": 26825838.0,
      "step": 12030
    },
    {
      "entropy": 5.5289122581481935,
      "epoch": 1.3516594597630145,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048277981999156534,
      "loss": 5.3142,
      "mean_token_accuracy": 0.19527603238821029,
      "num_tokens": 26836320.0,
      "step": 12035
    },
    {
      "entropy": 5.513550424575806,
      "epoch": 1.3522210366709722,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048276441239965414,
      "loss": 5.3089,
      "mean_token_accuracy": 0.19219874292612077,
      "num_tokens": 26846485.0,
      "step": 12040
    },
    {
      "entropy": 5.592586374282837,
      "epoch": 1.3527826135789296,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004827489981924434,
      "loss": 5.4285,
      "mean_token_accuracy": 0.1832916557788849,
      "num_tokens": 26858554.0,
      "step": 12045
    },
    {
      "entropy": 5.51733169555664,
      "epoch": 1.3533441904868873,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048273357737042393,
      "loss": 5.3427,
      "mean_token_accuracy": 0.19549855589866638,
      "num_tokens": 26869538.0,
      "step": 12050
    },
    {
      "entropy": 5.557057523727417,
      "epoch": 1.3539057673948447,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048271814993408694,
      "loss": 5.4092,
      "mean_token_accuracy": 0.18891253620386123,
      "num_tokens": 26880284.0,
      "step": 12055
    },
    {
      "entropy": 5.488908147811889,
      "epoch": 1.3544673443028024,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004827027158839233,
      "loss": 5.3995,
      "mean_token_accuracy": 0.1821691557765007,
      "num_tokens": 26893180.0,
      "step": 12060
    },
    {
      "entropy": 5.480938911437988,
      "epoch": 1.3550289212107598,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004826872752204248,
      "loss": 5.251,
      "mean_token_accuracy": 0.194026155769825,
      "num_tokens": 26904213.0,
      "step": 12065
    },
    {
      "entropy": 5.504014730453491,
      "epoch": 1.3555904981187172,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048267182794408276,
      "loss": 5.3422,
      "mean_token_accuracy": 0.19684326350688935,
      "num_tokens": 26915566.0,
      "step": 12070
    },
    {
      "entropy": 5.395453691482544,
      "epoch": 1.356152075026675,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004826563740553893,
      "loss": 5.2932,
      "mean_token_accuracy": 0.19817951172590256,
      "num_tokens": 26926253.0,
      "step": 12075
    },
    {
      "entropy": 5.429937934875488,
      "epoch": 1.3567136519346326,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004826409135548363,
      "loss": 5.292,
      "mean_token_accuracy": 0.19771064221858978,
      "num_tokens": 26936723.0,
      "step": 12080
    },
    {
      "entropy": 5.461929750442505,
      "epoch": 1.35727522884259,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004826254464429161,
      "loss": 5.2832,
      "mean_token_accuracy": 0.19243800789117813,
      "num_tokens": 26947551.0,
      "step": 12085
    },
    {
      "entropy": 5.534619951248169,
      "epoch": 1.3578368057505474,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048260997272012114,
      "loss": 5.4036,
      "mean_token_accuracy": 0.19200349599123,
      "num_tokens": 26959969.0,
      "step": 12090
    },
    {
      "entropy": 5.553624153137207,
      "epoch": 1.358398382658505,
      "grad_norm": 1.125,
      "learning_rate": 0.0004825944923869441,
      "loss": 5.4682,
      "mean_token_accuracy": 0.18287914842367173,
      "num_tokens": 26972511.0,
      "step": 12095
    },
    {
      "entropy": 5.565522480010986,
      "epoch": 1.3589599595664628,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048257900544387795,
      "loss": 5.4405,
      "mean_token_accuracy": 0.18577493578195572,
      "num_tokens": 26983827.0,
      "step": 12100
    },
    {
      "entropy": 5.592643737792969,
      "epoch": 1.3595215364744202,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048256351189141574,
      "loss": 5.404,
      "mean_token_accuracy": 0.18419968634843825,
      "num_tokens": 26994668.0,
      "step": 12105
    },
    {
      "entropy": 5.437271165847778,
      "epoch": 1.3600831133823776,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004825480117300508,
      "loss": 5.2605,
      "mean_token_accuracy": 0.1999809518456459,
      "num_tokens": 27006313.0,
      "step": 12110
    },
    {
      "entropy": 5.464850616455078,
      "epoch": 1.3606446902903353,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048253250496027675,
      "loss": 5.3272,
      "mean_token_accuracy": 0.1898820087313652,
      "num_tokens": 27017628.0,
      "step": 12115
    },
    {
      "entropy": 5.449024724960327,
      "epoch": 1.3612062671982927,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048251699158258714,
      "loss": 5.2192,
      "mean_token_accuracy": 0.1939452037215233,
      "num_tokens": 27028303.0,
      "step": 12120
    },
    {
      "entropy": 5.558975076675415,
      "epoch": 1.3617678441062504,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004825014715974761,
      "loss": 5.3899,
      "mean_token_accuracy": 0.18635841608047485,
      "num_tokens": 27039307.0,
      "step": 12125
    },
    {
      "entropy": 5.4659029006958,
      "epoch": 1.3623294210142078,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048248594500543767,
      "loss": 5.2662,
      "mean_token_accuracy": 0.1897073581814766,
      "num_tokens": 27049271.0,
      "step": 12130
    },
    {
      "entropy": 5.365239048004151,
      "epoch": 1.3628909979221655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048247041180696633,
      "loss": 5.179,
      "mean_token_accuracy": 0.19675081372261047,
      "num_tokens": 27059510.0,
      "step": 12135
    },
    {
      "entropy": 5.606520462036133,
      "epoch": 1.363452574830123,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004824548720025566,
      "loss": 5.523,
      "mean_token_accuracy": 0.17754722088575364,
      "num_tokens": 27071685.0,
      "step": 12140
    },
    {
      "entropy": 5.416141366958618,
      "epoch": 1.3640141517380806,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004824393255927033,
      "loss": 5.2462,
      "mean_token_accuracy": 0.1859876334667206,
      "num_tokens": 27081957.0,
      "step": 12145
    },
    {
      "entropy": 5.390149831771851,
      "epoch": 1.364575728646038,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048242377257790145,
      "loss": 5.254,
      "mean_token_accuracy": 0.19522633850574495,
      "num_tokens": 27091609.0,
      "step": 12150
    },
    {
      "entropy": 5.48789267539978,
      "epoch": 1.3651373055539957,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048240821295864614,
      "loss": 5.4331,
      "mean_token_accuracy": 0.1839677393436432,
      "num_tokens": 27103889.0,
      "step": 12155
    },
    {
      "entropy": 5.594719934463501,
      "epoch": 1.3656988824619531,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000482392646735433,
      "loss": 5.3644,
      "mean_token_accuracy": 0.19539450258016586,
      "num_tokens": 27114186.0,
      "step": 12160
    },
    {
      "entropy": 5.3895775318145756,
      "epoch": 1.3662604593699106,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004823770739087574,
      "loss": 5.2358,
      "mean_token_accuracy": 0.1960143566131592,
      "num_tokens": 27124677.0,
      "step": 12165
    },
    {
      "entropy": 5.543695735931396,
      "epoch": 1.3668220362778682,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004823614944791154,
      "loss": 5.4065,
      "mean_token_accuracy": 0.19060794711112977,
      "num_tokens": 27134873.0,
      "step": 12170
    },
    {
      "entropy": 5.506030416488647,
      "epoch": 1.367383613185826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004823459084470031,
      "loss": 5.4083,
      "mean_token_accuracy": 0.1852779746055603,
      "num_tokens": 27146107.0,
      "step": 12175
    },
    {
      "entropy": 5.439063644409179,
      "epoch": 1.3679451900937833,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004823303158129165,
      "loss": 5.2579,
      "mean_token_accuracy": 0.19871385544538497,
      "num_tokens": 27158821.0,
      "step": 12180
    },
    {
      "entropy": 5.448226499557495,
      "epoch": 1.3685067670017408,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004823147165773523,
      "loss": 5.2843,
      "mean_token_accuracy": 0.1965023010969162,
      "num_tokens": 27169513.0,
      "step": 12185
    },
    {
      "entropy": 5.390142393112183,
      "epoch": 1.3690683439096984,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048229911074080715,
      "loss": 5.2786,
      "mean_token_accuracy": 0.1985243782401085,
      "num_tokens": 27180059.0,
      "step": 12190
    },
    {
      "entropy": 5.485644292831421,
      "epoch": 1.369629920817656,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048228349830377787,
      "loss": 5.3711,
      "mean_token_accuracy": 0.1902673915028572,
      "num_tokens": 27190738.0,
      "step": 12195
    },
    {
      "entropy": 5.505376386642456,
      "epoch": 1.3701914977256135,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048226787926676165,
      "loss": 5.2515,
      "mean_token_accuracy": 0.1893245533108711,
      "num_tokens": 27200664.0,
      "step": 12200
    },
    {
      "entropy": 5.458845520019532,
      "epoch": 1.370753074633571,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048225225363025576,
      "loss": 5.3105,
      "mean_token_accuracy": 0.19332458823919296,
      "num_tokens": 27211011.0,
      "step": 12205
    },
    {
      "entropy": 5.4974650859832765,
      "epoch": 1.3713146515415287,
      "grad_norm": 1.125,
      "learning_rate": 0.00048223662139475773,
      "loss": 5.3715,
      "mean_token_accuracy": 0.1991906002163887,
      "num_tokens": 27221888.0,
      "step": 12210
    },
    {
      "entropy": 5.569535923004151,
      "epoch": 1.371876228449486,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048222098256076523,
      "loss": 5.3575,
      "mean_token_accuracy": 0.18815023750066756,
      "num_tokens": 27234671.0,
      "step": 12215
    },
    {
      "entropy": 5.406680059432984,
      "epoch": 1.3724378053574438,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004822053371287763,
      "loss": 5.1589,
      "mean_token_accuracy": 0.20078462660312651,
      "num_tokens": 27245283.0,
      "step": 12220
    },
    {
      "entropy": 5.483945226669311,
      "epoch": 1.3729993822654012,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004821896850992891,
      "loss": 5.3197,
      "mean_token_accuracy": 0.19311304688453673,
      "num_tokens": 27255056.0,
      "step": 12225
    },
    {
      "entropy": 5.567039108276367,
      "epoch": 1.3735609591733589,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000482174026472802,
      "loss": 5.3615,
      "mean_token_accuracy": 0.1872144341468811,
      "num_tokens": 27266893.0,
      "step": 12230
    },
    {
      "entropy": 5.468055295944214,
      "epoch": 1.3741225360813163,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004821583612498135,
      "loss": 5.392,
      "mean_token_accuracy": 0.18777517080307007,
      "num_tokens": 27279504.0,
      "step": 12235
    },
    {
      "entropy": 5.524121522903442,
      "epoch": 1.374684112989274,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048214268943082246,
      "loss": 5.4465,
      "mean_token_accuracy": 0.18385884016752244,
      "num_tokens": 27290525.0,
      "step": 12240
    },
    {
      "entropy": 5.534918260574341,
      "epoch": 1.3752456898972314,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004821270110163278,
      "loss": 5.3009,
      "mean_token_accuracy": 0.1905997797846794,
      "num_tokens": 27301404.0,
      "step": 12245
    },
    {
      "entropy": 5.474128341674804,
      "epoch": 1.375807266805189,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004821113260068288,
      "loss": 5.244,
      "mean_token_accuracy": 0.18948123157024382,
      "num_tokens": 27312635.0,
      "step": 12250
    },
    {
      "entropy": 5.376252698898315,
      "epoch": 1.3763688437131465,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004820956344028249,
      "loss": 5.2927,
      "mean_token_accuracy": 0.19547967612743378,
      "num_tokens": 27322485.0,
      "step": 12255
    },
    {
      "entropy": 5.478473758697509,
      "epoch": 1.376930420621104,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048207993620481566,
      "loss": 5.3524,
      "mean_token_accuracy": 0.18763597011566163,
      "num_tokens": 27332568.0,
      "step": 12260
    },
    {
      "entropy": 5.604016065597534,
      "epoch": 1.3774919975290616,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004820642314133009,
      "loss": 5.3856,
      "mean_token_accuracy": 0.18369714319705963,
      "num_tokens": 27343058.0,
      "step": 12265
    },
    {
      "entropy": 5.5251256942749025,
      "epoch": 1.3780535744370193,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048204852002878073,
      "loss": 5.2777,
      "mean_token_accuracy": 0.1988302230834961,
      "num_tokens": 27355316.0,
      "step": 12270
    },
    {
      "entropy": 5.479616785049439,
      "epoch": 1.3786151513449767,
      "grad_norm": 1.25,
      "learning_rate": 0.00048203280205175544,
      "loss": 5.2206,
      "mean_token_accuracy": 0.1981513559818268,
      "num_tokens": 27366098.0,
      "step": 12275
    },
    {
      "entropy": 5.421827983856201,
      "epoch": 1.3791767282529341,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004820170774827253,
      "loss": 5.2369,
      "mean_token_accuracy": 0.1967311054468155,
      "num_tokens": 27376923.0,
      "step": 12280
    },
    {
      "entropy": 5.4623127460479735,
      "epoch": 1.3797383051608918,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004820013463221912,
      "loss": 5.3739,
      "mean_token_accuracy": 0.18950267285108566,
      "num_tokens": 27386789.0,
      "step": 12285
    },
    {
      "entropy": 5.4712121963500975,
      "epoch": 1.3802998820688495,
      "grad_norm": 1.078125,
      "learning_rate": 0.000481985608570654,
      "loss": 5.2972,
      "mean_token_accuracy": 0.18938076943159105,
      "num_tokens": 27399241.0,
      "step": 12290
    },
    {
      "entropy": 5.493840456008911,
      "epoch": 1.380861458976807,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004819698642286147,
      "loss": 5.1998,
      "mean_token_accuracy": 0.20267933756113052,
      "num_tokens": 27409123.0,
      "step": 12295
    },
    {
      "entropy": 5.411140584945679,
      "epoch": 1.3814230358847643,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048195411329657464,
      "loss": 5.2602,
      "mean_token_accuracy": 0.189808289706707,
      "num_tokens": 27419920.0,
      "step": 12300
    },
    {
      "entropy": 5.349962997436523,
      "epoch": 1.381984612792722,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004819383557750353,
      "loss": 5.1474,
      "mean_token_accuracy": 0.19934788793325425,
      "num_tokens": 27429552.0,
      "step": 12305
    },
    {
      "entropy": 5.522129106521606,
      "epoch": 1.3825461897006794,
      "grad_norm": 1.125,
      "learning_rate": 0.00048192259166449845,
      "loss": 5.3666,
      "mean_token_accuracy": 0.19459759145975114,
      "num_tokens": 27440404.0,
      "step": 12310
    },
    {
      "entropy": 5.489263820648193,
      "epoch": 1.383107766608637,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048190682096546606,
      "loss": 5.3942,
      "mean_token_accuracy": 0.1918272614479065,
      "num_tokens": 27452231.0,
      "step": 12315
    },
    {
      "entropy": 5.490846252441406,
      "epoch": 1.3836693435165945,
      "grad_norm": 1.125,
      "learning_rate": 0.00048189104367844034,
      "loss": 5.2863,
      "mean_token_accuracy": 0.1911655768752098,
      "num_tokens": 27463658.0,
      "step": 12320
    },
    {
      "entropy": 5.5051093101501465,
      "epoch": 1.3842309204245522,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048187525980392343,
      "loss": 5.3354,
      "mean_token_accuracy": 0.18559889048337935,
      "num_tokens": 27475403.0,
      "step": 12325
    },
    {
      "entropy": 5.559214401245117,
      "epoch": 1.3847924973325096,
      "grad_norm": 1.125,
      "learning_rate": 0.000481859469342418,
      "loss": 5.406,
      "mean_token_accuracy": 0.18680911213159562,
      "num_tokens": 27487056.0,
      "step": 12330
    },
    {
      "entropy": 5.427961730957032,
      "epoch": 1.3853540742404673,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004818436722944269,
      "loss": 5.269,
      "mean_token_accuracy": 0.1931942269206047,
      "num_tokens": 27498659.0,
      "step": 12335
    },
    {
      "entropy": 5.604191732406616,
      "epoch": 1.3859156511484247,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004818278686604531,
      "loss": 5.4099,
      "mean_token_accuracy": 0.1852207824587822,
      "num_tokens": 27510577.0,
      "step": 12340
    },
    {
      "entropy": 5.513951683044434,
      "epoch": 1.3864772280563824,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004818120584409996,
      "loss": 5.3258,
      "mean_token_accuracy": 0.19546688497066497,
      "num_tokens": 27521441.0,
      "step": 12345
    },
    {
      "entropy": 5.481344890594483,
      "epoch": 1.3870388049643398,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004817962416365701,
      "loss": 5.2698,
      "mean_token_accuracy": 0.1940573751926422,
      "num_tokens": 27533443.0,
      "step": 12350
    },
    {
      "entropy": 5.492029619216919,
      "epoch": 1.3876003818722973,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048178041824766796,
      "loss": 5.3216,
      "mean_token_accuracy": 0.19381196796894073,
      "num_tokens": 27543872.0,
      "step": 12355
    },
    {
      "entropy": 5.63539752960205,
      "epoch": 1.388161958780255,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004817645882747971,
      "loss": 5.513,
      "mean_token_accuracy": 0.18065459430217742,
      "num_tokens": 27556556.0,
      "step": 12360
    },
    {
      "entropy": 5.505446863174439,
      "epoch": 1.3887235356882126,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004817487517184616,
      "loss": 5.3759,
      "mean_token_accuracy": 0.18792864829301834,
      "num_tokens": 27570739.0,
      "step": 12365
    },
    {
      "entropy": 5.530737257003784,
      "epoch": 1.38928511259617,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004817329085791657,
      "loss": 5.4155,
      "mean_token_accuracy": 0.18117647469043732,
      "num_tokens": 27582438.0,
      "step": 12370
    },
    {
      "entropy": 5.536424160003662,
      "epoch": 1.3898466895041275,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004817170588574138,
      "loss": 5.3438,
      "mean_token_accuracy": 0.19049696773290634,
      "num_tokens": 27593854.0,
      "step": 12375
    },
    {
      "entropy": 5.562060594558716,
      "epoch": 1.3904082664120851,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004817012025537105,
      "loss": 5.2976,
      "mean_token_accuracy": 0.19260254949331285,
      "num_tokens": 27604459.0,
      "step": 12380
    },
    {
      "entropy": 5.590554714202881,
      "epoch": 1.3909698433200428,
      "grad_norm": 1.125,
      "learning_rate": 0.0004816853396685608,
      "loss": 5.4944,
      "mean_token_accuracy": 0.17838713377714158,
      "num_tokens": 27617901.0,
      "step": 12385
    },
    {
      "entropy": 5.562151765823364,
      "epoch": 1.3915314202280002,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048166947020246973,
      "loss": 5.4018,
      "mean_token_accuracy": 0.18608833253383636,
      "num_tokens": 27628574.0,
      "step": 12390
    },
    {
      "entropy": 5.520782613754273,
      "epoch": 1.3920929971359577,
      "grad_norm": 1.125,
      "learning_rate": 0.00048165359415594245,
      "loss": 5.271,
      "mean_token_accuracy": 0.19218895584344864,
      "num_tokens": 27639195.0,
      "step": 12395
    },
    {
      "entropy": 5.433777475357056,
      "epoch": 1.3926545740439153,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048163771152948473,
      "loss": 5.203,
      "mean_token_accuracy": 0.2015116885304451,
      "num_tokens": 27649898.0,
      "step": 12400
    },
    {
      "entropy": 5.486884546279907,
      "epoch": 1.3932161509518728,
      "grad_norm": 1.09375,
      "learning_rate": 0.000481621822323602,
      "loss": 5.351,
      "mean_token_accuracy": 0.1842460259795189,
      "num_tokens": 27663617.0,
      "step": 12405
    },
    {
      "entropy": 5.380657386779785,
      "epoch": 1.3937777278598305,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004816059265388004,
      "loss": 5.2187,
      "mean_token_accuracy": 0.19785332977771758,
      "num_tokens": 27674823.0,
      "step": 12410
    },
    {
      "entropy": 5.4988696575164795,
      "epoch": 1.394339304767788,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048159002417558596,
      "loss": 5.3503,
      "mean_token_accuracy": 0.19219503402709961,
      "num_tokens": 27686483.0,
      "step": 12415
    },
    {
      "entropy": 5.531269931793213,
      "epoch": 1.3949008816757456,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004815741152344651,
      "loss": 5.2916,
      "mean_token_accuracy": 0.19547995328903198,
      "num_tokens": 27697081.0,
      "step": 12420
    },
    {
      "entropy": 5.444240856170654,
      "epoch": 1.395462458583703,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004815581997159442,
      "loss": 5.3507,
      "mean_token_accuracy": 0.19486741870641708,
      "num_tokens": 27706710.0,
      "step": 12425
    },
    {
      "entropy": 5.48530011177063,
      "epoch": 1.3960240354916607,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048154227762053003,
      "loss": 5.3902,
      "mean_token_accuracy": 0.18007151484489442,
      "num_tokens": 27718213.0,
      "step": 12430
    },
    {
      "entropy": 5.476761341094971,
      "epoch": 1.396585612399618,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004815263489487298,
      "loss": 5.3101,
      "mean_token_accuracy": 0.1885454311966896,
      "num_tokens": 27729064.0,
      "step": 12435
    },
    {
      "entropy": 5.53508186340332,
      "epoch": 1.3971471893075758,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004815104137010504,
      "loss": 5.3047,
      "mean_token_accuracy": 0.19287462383508683,
      "num_tokens": 27740121.0,
      "step": 12440
    },
    {
      "entropy": 5.447685861587525,
      "epoch": 1.3977087662155332,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004814944718779993,
      "loss": 5.3476,
      "mean_token_accuracy": 0.1958514094352722,
      "num_tokens": 27751597.0,
      "step": 12445
    },
    {
      "entropy": 5.409804916381836,
      "epoch": 1.3982703431234906,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004814785234800843,
      "loss": 5.2541,
      "mean_token_accuracy": 0.19108182787895203,
      "num_tokens": 27761539.0,
      "step": 12450
    },
    {
      "entropy": 5.4863481521606445,
      "epoch": 1.3988319200314483,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048146256850781287,
      "loss": 5.2594,
      "mean_token_accuracy": 0.1919708341360092,
      "num_tokens": 27772904.0,
      "step": 12455
    },
    {
      "entropy": 5.51110053062439,
      "epoch": 1.399393496939406,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048144660696169327,
      "loss": 5.266,
      "mean_token_accuracy": 0.19515890181064605,
      "num_tokens": 27783706.0,
      "step": 12460
    },
    {
      "entropy": 5.458293867111206,
      "epoch": 1.3999550738473634,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048143063884223365,
      "loss": 5.2888,
      "mean_token_accuracy": 0.18573704957962037,
      "num_tokens": 27794503.0,
      "step": 12465
    },
    {
      "entropy": 5.452811193466187,
      "epoch": 1.4005166507553208,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004814146641499424,
      "loss": 5.283,
      "mean_token_accuracy": 0.19184025824069978,
      "num_tokens": 27805952.0,
      "step": 12470
    },
    {
      "entropy": 5.502015399932861,
      "epoch": 1.4010782276632785,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048139868288532816,
      "loss": 5.2885,
      "mean_token_accuracy": 0.1947245940566063,
      "num_tokens": 27817530.0,
      "step": 12475
    },
    {
      "entropy": 5.519699287414551,
      "epoch": 1.4016398045712362,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004813826950488998,
      "loss": 5.3536,
      "mean_token_accuracy": 0.18735283315181733,
      "num_tokens": 27829932.0,
      "step": 12480
    },
    {
      "entropy": 5.51725401878357,
      "epoch": 1.4022013814791936,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048136670064116645,
      "loss": 5.3415,
      "mean_token_accuracy": 0.19383803606033326,
      "num_tokens": 27840993.0,
      "step": 12485
    },
    {
      "entropy": 5.514503240585327,
      "epoch": 1.402762958387151,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004813506996626373,
      "loss": 5.2684,
      "mean_token_accuracy": 0.19149025678634643,
      "num_tokens": 27852727.0,
      "step": 12490
    },
    {
      "entropy": 5.466609477996826,
      "epoch": 1.4033245352951087,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004813346921138218,
      "loss": 5.4397,
      "mean_token_accuracy": 0.1922093689441681,
      "num_tokens": 27865183.0,
      "step": 12495
    },
    {
      "entropy": 5.410083341598511,
      "epoch": 1.4038861122030661,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004813186779952297,
      "loss": 5.1917,
      "mean_token_accuracy": 0.20270057320594786,
      "num_tokens": 27877266.0,
      "step": 12500
    },
    {
      "entropy": 5.548344945907592,
      "epoch": 1.4044476891110238,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004813026573073708,
      "loss": 5.3368,
      "mean_token_accuracy": 0.18815984725952148,
      "num_tokens": 27888863.0,
      "step": 12505
    },
    {
      "entropy": 5.452338647842407,
      "epoch": 1.4050092660189812,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004812866300507554,
      "loss": 5.274,
      "mean_token_accuracy": 0.19844181090593338,
      "num_tokens": 27900443.0,
      "step": 12510
    },
    {
      "entropy": 5.4826854228973385,
      "epoch": 1.405570842926939,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004812705962258936,
      "loss": 5.2625,
      "mean_token_accuracy": 0.19299466013908387,
      "num_tokens": 27911241.0,
      "step": 12515
    },
    {
      "entropy": 5.488195991516113,
      "epoch": 1.4061324198348963,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004812545558332961,
      "loss": 5.2858,
      "mean_token_accuracy": 0.19225993007421494,
      "num_tokens": 27922171.0,
      "step": 12520
    },
    {
      "entropy": 5.38026533126831,
      "epoch": 1.406693996742854,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048123850887347346,
      "loss": 5.2409,
      "mean_token_accuracy": 0.1974858194589615,
      "num_tokens": 27933627.0,
      "step": 12525
    },
    {
      "entropy": 5.318044519424438,
      "epoch": 1.4072555736508114,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048122245534693676,
      "loss": 5.2029,
      "mean_token_accuracy": 0.18923496454954147,
      "num_tokens": 27943585.0,
      "step": 12530
    },
    {
      "entropy": 5.532860851287841,
      "epoch": 1.407817150558769,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004812063952541971,
      "loss": 5.317,
      "mean_token_accuracy": 0.1894823729991913,
      "num_tokens": 27955554.0,
      "step": 12535
    },
    {
      "entropy": 5.52153058052063,
      "epoch": 1.4083787274667265,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048119032859576577,
      "loss": 5.1996,
      "mean_token_accuracy": 0.1972256824374199,
      "num_tokens": 27966150.0,
      "step": 12540
    },
    {
      "entropy": 5.426303052902222,
      "epoch": 1.408940304374684,
      "grad_norm": 1.125,
      "learning_rate": 0.0004811742553721545,
      "loss": 5.2422,
      "mean_token_accuracy": 0.19363798648118974,
      "num_tokens": 27977436.0,
      "step": 12545
    },
    {
      "entropy": 5.5245991230010985,
      "epoch": 1.4095018812826416,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048115817558387494,
      "loss": 5.4071,
      "mean_token_accuracy": 0.18410052061080934,
      "num_tokens": 27988325.0,
      "step": 12550
    },
    {
      "entropy": 5.561803531646729,
      "epoch": 1.4100634581905993,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004811420892314391,
      "loss": 5.2413,
      "mean_token_accuracy": 0.1947012275457382,
      "num_tokens": 27999760.0,
      "step": 12555
    },
    {
      "entropy": 5.545596790313721,
      "epoch": 1.4106250350985567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004811259963153592,
      "loss": 5.409,
      "mean_token_accuracy": 0.1847024142742157,
      "num_tokens": 28011333.0,
      "step": 12560
    },
    {
      "entropy": 5.42786660194397,
      "epoch": 1.4111866120065142,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004811098968361476,
      "loss": 5.3376,
      "mean_token_accuracy": 0.19361564218997956,
      "num_tokens": 28023464.0,
      "step": 12565
    },
    {
      "entropy": 5.443120670318604,
      "epoch": 1.4117481889144718,
      "grad_norm": 1.171875,
      "learning_rate": 0.000481093790794317,
      "loss": 5.2982,
      "mean_token_accuracy": 0.184749773144722,
      "num_tokens": 28034205.0,
      "step": 12570
    },
    {
      "entropy": 5.538404178619385,
      "epoch": 1.4123097658224295,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004810776781903801,
      "loss": 5.288,
      "mean_token_accuracy": 0.18942586928606034,
      "num_tokens": 28045846.0,
      "step": 12575
    },
    {
      "entropy": 5.497859764099121,
      "epoch": 1.412871342730387,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048106155902485003,
      "loss": 5.3002,
      "mean_token_accuracy": 0.19742057472467422,
      "num_tokens": 28057365.0,
      "step": 12580
    },
    {
      "entropy": 5.482359743118286,
      "epoch": 1.4134329196383444,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004810454332982401,
      "loss": 5.3851,
      "mean_token_accuracy": 0.1854017272591591,
      "num_tokens": 28069098.0,
      "step": 12585
    },
    {
      "entropy": 5.507374238967896,
      "epoch": 1.413994496546302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004810293010110635,
      "loss": 5.3201,
      "mean_token_accuracy": 0.1919554203748703,
      "num_tokens": 28080008.0,
      "step": 12590
    },
    {
      "entropy": 5.577986812591552,
      "epoch": 1.4145560734542595,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004810131621638342,
      "loss": 5.3612,
      "mean_token_accuracy": 0.18817386478185655,
      "num_tokens": 28091657.0,
      "step": 12595
    },
    {
      "entropy": 5.4135431289672855,
      "epoch": 1.4151176503622171,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048099701675706583,
      "loss": 5.2356,
      "mean_token_accuracy": 0.1927874505519867,
      "num_tokens": 28102586.0,
      "step": 12600
    },
    {
      "entropy": 5.466880178451538,
      "epoch": 1.4156792272701746,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004809808647912725,
      "loss": 5.3369,
      "mean_token_accuracy": 0.1913236677646637,
      "num_tokens": 28113066.0,
      "step": 12605
    },
    {
      "entropy": 5.525369787216187,
      "epoch": 1.4162408041781323,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004809647062669686,
      "loss": 5.3299,
      "mean_token_accuracy": 0.18975775837898254,
      "num_tokens": 28124814.0,
      "step": 12610
    },
    {
      "entropy": 5.601166725158691,
      "epoch": 1.4168023810860897,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048094854118466856,
      "loss": 5.3664,
      "mean_token_accuracy": 0.18623321056365966,
      "num_tokens": 28138009.0,
      "step": 12615
    },
    {
      "entropy": 5.39503927230835,
      "epoch": 1.4173639579940474,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048093236954488715,
      "loss": 5.2687,
      "mean_token_accuracy": 0.19271760582923889,
      "num_tokens": 28149222.0,
      "step": 12620
    },
    {
      "entropy": 5.469629001617432,
      "epoch": 1.4179255349020048,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004809161913481392,
      "loss": 5.33,
      "mean_token_accuracy": 0.19065048396587372,
      "num_tokens": 28161323.0,
      "step": 12625
    },
    {
      "entropy": 5.497536325454712,
      "epoch": 1.4184871118099625,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048090000659493985,
      "loss": 5.3072,
      "mean_token_accuracy": 0.19020311385393143,
      "num_tokens": 28173242.0,
      "step": 12630
    },
    {
      "entropy": 5.459263563156128,
      "epoch": 1.41904868871792,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048088381528580445,
      "loss": 5.2768,
      "mean_token_accuracy": 0.19995663762092591,
      "num_tokens": 28183927.0,
      "step": 12635
    },
    {
      "entropy": 5.451407861709595,
      "epoch": 1.4196102656258773,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048086761742124846,
      "loss": 5.2704,
      "mean_token_accuracy": 0.19252085089683532,
      "num_tokens": 28194375.0,
      "step": 12640
    },
    {
      "entropy": 5.494913482666016,
      "epoch": 1.420171842533835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048085141300178774,
      "loss": 5.3253,
      "mean_token_accuracy": 0.19481477290391921,
      "num_tokens": 28206492.0,
      "step": 12645
    },
    {
      "entropy": 5.480983066558838,
      "epoch": 1.4207334194417927,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048083520202793815,
      "loss": 5.3536,
      "mean_token_accuracy": 0.18314705938100814,
      "num_tokens": 28217795.0,
      "step": 12650
    },
    {
      "entropy": 5.431267499923706,
      "epoch": 1.42129499634975,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004808189845002159,
      "loss": 5.2972,
      "mean_token_accuracy": 0.1943140760064125,
      "num_tokens": 28228796.0,
      "step": 12655
    },
    {
      "entropy": 5.507163000106812,
      "epoch": 1.4218565732577075,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004808027604191374,
      "loss": 5.3315,
      "mean_token_accuracy": 0.19252949357032775,
      "num_tokens": 28238873.0,
      "step": 12660
    },
    {
      "entropy": 5.46311354637146,
      "epoch": 1.4224181501656652,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048078652978521914,
      "loss": 5.1383,
      "mean_token_accuracy": 0.205023892223835,
      "num_tokens": 28249832.0,
      "step": 12665
    },
    {
      "entropy": 5.472410678863525,
      "epoch": 1.4229797270736229,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048077029259897797,
      "loss": 5.3395,
      "mean_token_accuracy": 0.18793419599533082,
      "num_tokens": 28261532.0,
      "step": 12670
    },
    {
      "entropy": 5.557413625717163,
      "epoch": 1.4235413039815803,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048075404886093094,
      "loss": 5.3666,
      "mean_token_accuracy": 0.19007611721754075,
      "num_tokens": 28272008.0,
      "step": 12675
    },
    {
      "entropy": 5.501730108261109,
      "epoch": 1.4241028808895377,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048073779857159504,
      "loss": 5.3299,
      "mean_token_accuracy": 0.19588117748498918,
      "num_tokens": 28283239.0,
      "step": 12680
    },
    {
      "entropy": 5.544003295898437,
      "epoch": 1.4246644577974954,
      "grad_norm": 1.0625,
      "learning_rate": 0.000480721541731488,
      "loss": 5.3596,
      "mean_token_accuracy": 0.18821366429328917,
      "num_tokens": 28293597.0,
      "step": 12685
    },
    {
      "entropy": 5.4740571022033695,
      "epoch": 1.4252260347054528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048070527834112727,
      "loss": 5.2833,
      "mean_token_accuracy": 0.1938060313463211,
      "num_tokens": 28305124.0,
      "step": 12690
    },
    {
      "entropy": 5.422293663024902,
      "epoch": 1.4257876116134105,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004806890084010306,
      "loss": 5.2945,
      "mean_token_accuracy": 0.19523887783288957,
      "num_tokens": 28315875.0,
      "step": 12695
    },
    {
      "entropy": 5.480016803741455,
      "epoch": 1.426349188521368,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048067273191171615,
      "loss": 5.2741,
      "mean_token_accuracy": 0.1949415236711502,
      "num_tokens": 28327705.0,
      "step": 12700
    },
    {
      "entropy": 5.4407182216644285,
      "epoch": 1.4269107654293256,
      "grad_norm": 1.0,
      "learning_rate": 0.0004806564488737022,
      "loss": 5.2614,
      "mean_token_accuracy": 0.1960237666964531,
      "num_tokens": 28338300.0,
      "step": 12705
    },
    {
      "entropy": 5.4076940536499025,
      "epoch": 1.427472342337283,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004806401592875071,
      "loss": 5.2238,
      "mean_token_accuracy": 0.1996994972229004,
      "num_tokens": 28348485.0,
      "step": 12710
    },
    {
      "entropy": 5.481090927124024,
      "epoch": 1.4280339192452407,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048062386315364964,
      "loss": 5.2562,
      "mean_token_accuracy": 0.19686545729637145,
      "num_tokens": 28359118.0,
      "step": 12715
    },
    {
      "entropy": 5.447401285171509,
      "epoch": 1.4285954961531981,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004806075604726485,
      "loss": 5.2881,
      "mean_token_accuracy": 0.19529957473278045,
      "num_tokens": 28369608.0,
      "step": 12720
    },
    {
      "entropy": 5.492588376998901,
      "epoch": 1.4291570730611558,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048059125124502304,
      "loss": 5.2821,
      "mean_token_accuracy": 0.1956749364733696,
      "num_tokens": 28379709.0,
      "step": 12725
    },
    {
      "entropy": 5.471444416046142,
      "epoch": 1.4297186499691132,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004805749354712924,
      "loss": 5.3086,
      "mean_token_accuracy": 0.19217366576194764,
      "num_tokens": 28390758.0,
      "step": 12730
    },
    {
      "entropy": 5.665639638900757,
      "epoch": 1.4302802268770707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000480558613151976,
      "loss": 5.5601,
      "mean_token_accuracy": 0.1810941606760025,
      "num_tokens": 28403485.0,
      "step": 12735
    },
    {
      "entropy": 5.611013650894165,
      "epoch": 1.4308418037850283,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004805422842875936,
      "loss": 5.3605,
      "mean_token_accuracy": 0.1880544364452362,
      "num_tokens": 28413627.0,
      "step": 12740
    },
    {
      "entropy": 5.499139595031738,
      "epoch": 1.431403380692986,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048052594887866524,
      "loss": 5.2441,
      "mean_token_accuracy": 0.1897425964474678,
      "num_tokens": 28424202.0,
      "step": 12745
    },
    {
      "entropy": 5.356061124801636,
      "epoch": 1.4319649576009434,
      "grad_norm": 1.125,
      "learning_rate": 0.00048050960692571086,
      "loss": 5.2211,
      "mean_token_accuracy": 0.1956148400902748,
      "num_tokens": 28435136.0,
      "step": 12750
    },
    {
      "entropy": 5.407910776138306,
      "epoch": 1.4325265345089009,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000480493258429251,
      "loss": 5.235,
      "mean_token_accuracy": 0.19518414586782457,
      "num_tokens": 28444915.0,
      "step": 12755
    },
    {
      "entropy": 5.464826679229736,
      "epoch": 1.4330881114168585,
      "grad_norm": 1.03125,
      "learning_rate": 0.000480476903389806,
      "loss": 5.3008,
      "mean_token_accuracy": 0.1882382333278656,
      "num_tokens": 28458295.0,
      "step": 12760
    },
    {
      "entropy": 5.542707633972168,
      "epoch": 1.4336496883248162,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004804605418078968,
      "loss": 5.349,
      "mean_token_accuracy": 0.1964672848582268,
      "num_tokens": 28469587.0,
      "step": 12765
    },
    {
      "entropy": 5.559011840820313,
      "epoch": 1.4342112652327736,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048044417368404426,
      "loss": 5.3453,
      "mean_token_accuracy": 0.1865287646651268,
      "num_tokens": 28481498.0,
      "step": 12770
    },
    {
      "entropy": 5.524534511566162,
      "epoch": 1.434772842140731,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048042779901876946,
      "loss": 5.4694,
      "mean_token_accuracy": 0.18070788830518722,
      "num_tokens": 28492626.0,
      "step": 12775
    },
    {
      "entropy": 5.497053003311157,
      "epoch": 1.4353344190486887,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004804114178125939,
      "loss": 5.3346,
      "mean_token_accuracy": 0.19229722023010254,
      "num_tokens": 28503882.0,
      "step": 12780
    },
    {
      "entropy": 5.532601833343506,
      "epoch": 1.4358959959566462,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004803950300660391,
      "loss": 5.2667,
      "mean_token_accuracy": 0.20189556181430818,
      "num_tokens": 28514786.0,
      "step": 12785
    },
    {
      "entropy": 5.491387796401978,
      "epoch": 1.4364575728646038,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004803786357796269,
      "loss": 5.3003,
      "mean_token_accuracy": 0.19508771896362304,
      "num_tokens": 28526045.0,
      "step": 12790
    },
    {
      "entropy": 5.531581211090088,
      "epoch": 1.4370191497725613,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048036223495387934,
      "loss": 5.3128,
      "mean_token_accuracy": 0.1926025003194809,
      "num_tokens": 28536773.0,
      "step": 12795
    },
    {
      "entropy": 5.442726755142212,
      "epoch": 1.437580726680519,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004803458275893185,
      "loss": 5.2426,
      "mean_token_accuracy": 0.19309162199497223,
      "num_tokens": 28548099.0,
      "step": 12800
    },
    {
      "entropy": 5.45169415473938,
      "epoch": 1.4381423035884764,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004803294136864669,
      "loss": 5.3134,
      "mean_token_accuracy": 0.19566312730312346,
      "num_tokens": 28559372.0,
      "step": 12805
    },
    {
      "entropy": 5.4630919933319095,
      "epoch": 1.438703880496434,
      "grad_norm": 1.25,
      "learning_rate": 0.00048031299324584716,
      "loss": 5.3085,
      "mean_token_accuracy": 0.1951773449778557,
      "num_tokens": 28570106.0,
      "step": 12810
    },
    {
      "entropy": 5.457654142379761,
      "epoch": 1.4392654574043915,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000480296566267982,
      "loss": 5.3312,
      "mean_token_accuracy": 0.19525408297777175,
      "num_tokens": 28582417.0,
      "step": 12815
    },
    {
      "entropy": 5.549852800369263,
      "epoch": 1.4398270343123492,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004802801327533947,
      "loss": 5.4421,
      "mean_token_accuracy": 0.19046890884637832,
      "num_tokens": 28594341.0,
      "step": 12820
    },
    {
      "entropy": 5.497651624679565,
      "epoch": 1.4403886112203066,
      "grad_norm": 1.125,
      "learning_rate": 0.00048026369270260814,
      "loss": 5.3253,
      "mean_token_accuracy": 0.19379577934741973,
      "num_tokens": 28605580.0,
      "step": 12825
    },
    {
      "entropy": 5.535317039489746,
      "epoch": 1.440950188128264,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004802472461161462,
      "loss": 5.3568,
      "mean_token_accuracy": 0.1853962793946266,
      "num_tokens": 28617779.0,
      "step": 12830
    },
    {
      "entropy": 5.543725109100341,
      "epoch": 1.4415117650362217,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004802307929945322,
      "loss": 5.3502,
      "mean_token_accuracy": 0.18570167124271392,
      "num_tokens": 28629714.0,
      "step": 12835
    },
    {
      "entropy": 5.555856466293335,
      "epoch": 1.4420733419441794,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048021433333829015,
      "loss": 5.3589,
      "mean_token_accuracy": 0.18780214488506317,
      "num_tokens": 28641381.0,
      "step": 12840
    },
    {
      "entropy": 5.518057870864868,
      "epoch": 1.4426349188521368,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048019786714794424,
      "loss": 5.3499,
      "mean_token_accuracy": 0.1917356953024864,
      "num_tokens": 28653650.0,
      "step": 12845
    },
    {
      "entropy": 5.477000761032104,
      "epoch": 1.4431964957600942,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004801813944240186,
      "loss": 5.2695,
      "mean_token_accuracy": 0.19412501752376557,
      "num_tokens": 28665252.0,
      "step": 12850
    },
    {
      "entropy": 5.452428817749023,
      "epoch": 1.443758072668052,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004801649151670378,
      "loss": 5.3091,
      "mean_token_accuracy": 0.19653584510087968,
      "num_tokens": 28677072.0,
      "step": 12855
    },
    {
      "entropy": 5.423432111740112,
      "epoch": 1.4443196495760096,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004801484293775265,
      "loss": 5.2378,
      "mean_token_accuracy": 0.19026581794023514,
      "num_tokens": 28688767.0,
      "step": 12860
    },
    {
      "entropy": 5.538924264907837,
      "epoch": 1.444881226483967,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004801319370560097,
      "loss": 5.5004,
      "mean_token_accuracy": 0.18820954412221907,
      "num_tokens": 28699773.0,
      "step": 12865
    },
    {
      "entropy": 5.571140956878662,
      "epoch": 1.4454428033919244,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004801154382030124,
      "loss": 5.3737,
      "mean_token_accuracy": 0.18863944411277772,
      "num_tokens": 28710971.0,
      "step": 12870
    },
    {
      "entropy": 5.489381122589111,
      "epoch": 1.446004380299882,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048009893281906004,
      "loss": 5.2367,
      "mean_token_accuracy": 0.19461744278669357,
      "num_tokens": 28721544.0,
      "step": 12875
    },
    {
      "entropy": 5.541806602478028,
      "epoch": 1.4465659572078395,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048008242090467803,
      "loss": 5.4494,
      "mean_token_accuracy": 0.18645317256450653,
      "num_tokens": 28733790.0,
      "step": 12880
    },
    {
      "entropy": 5.582160615921021,
      "epoch": 1.4471275341157972,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004800659024603923,
      "loss": 5.3539,
      "mean_token_accuracy": 0.19444509148597716,
      "num_tokens": 28745218.0,
      "step": 12885
    },
    {
      "entropy": 5.6067163944244385,
      "epoch": 1.4476891110237546,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048004937748672864,
      "loss": 5.4006,
      "mean_token_accuracy": 0.18395177125930787,
      "num_tokens": 28758061.0,
      "step": 12890
    },
    {
      "entropy": 5.444657278060913,
      "epoch": 1.4482506879317123,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048003284598421327,
      "loss": 5.2122,
      "mean_token_accuracy": 0.20201971381902695,
      "num_tokens": 28770037.0,
      "step": 12895
    },
    {
      "entropy": 5.561487770080566,
      "epoch": 1.4488122648396697,
      "grad_norm": 1.125,
      "learning_rate": 0.0004800163079533726,
      "loss": 5.3851,
      "mean_token_accuracy": 0.18889002948999406,
      "num_tokens": 28781052.0,
      "step": 12900
    },
    {
      "entropy": 5.435133695602417,
      "epoch": 1.4493738417476274,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004799997633947332,
      "loss": 5.2454,
      "mean_token_accuracy": 0.19051423966884612,
      "num_tokens": 28791225.0,
      "step": 12905
    },
    {
      "entropy": 5.48399248123169,
      "epoch": 1.4499354186555848,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047998321230882174,
      "loss": 5.2937,
      "mean_token_accuracy": 0.19318269491195678,
      "num_tokens": 28803022.0,
      "step": 12910
    },
    {
      "entropy": 5.42376823425293,
      "epoch": 1.4504969955635425,
      "grad_norm": 1.125,
      "learning_rate": 0.00047996665469616536,
      "loss": 5.2714,
      "mean_token_accuracy": 0.1893945261836052,
      "num_tokens": 28813789.0,
      "step": 12915
    },
    {
      "entropy": 5.540867328643799,
      "epoch": 1.4510585724715,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047995009055729114,
      "loss": 5.3309,
      "mean_token_accuracy": 0.19015136808156968,
      "num_tokens": 28825015.0,
      "step": 12920
    },
    {
      "entropy": 5.6369280338287355,
      "epoch": 1.4516201493794574,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004799335198927266,
      "loss": 5.4289,
      "mean_token_accuracy": 0.18817705661058426,
      "num_tokens": 28835852.0,
      "step": 12925
    },
    {
      "entropy": 5.461359071731567,
      "epoch": 1.452181726287415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004799169427029993,
      "loss": 5.3229,
      "mean_token_accuracy": 0.19205658584833146,
      "num_tokens": 28846684.0,
      "step": 12930
    },
    {
      "entropy": 5.46303915977478,
      "epoch": 1.4527433031953727,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047990035898863706,
      "loss": 5.3489,
      "mean_token_accuracy": 0.18716940730810167,
      "num_tokens": 28857402.0,
      "step": 12935
    },
    {
      "entropy": 5.437236499786377,
      "epoch": 1.4533048801033301,
      "grad_norm": 1.125,
      "learning_rate": 0.0004798837687501679,
      "loss": 5.1814,
      "mean_token_accuracy": 0.19869955033063888,
      "num_tokens": 28867866.0,
      "step": 12940
    },
    {
      "entropy": 5.452563762664795,
      "epoch": 1.4538664570112876,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004798671719881201,
      "loss": 5.2944,
      "mean_token_accuracy": 0.19044434130191804,
      "num_tokens": 28878102.0,
      "step": 12945
    },
    {
      "entropy": 5.587057209014892,
      "epoch": 1.4544280339192452,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004798505687030221,
      "loss": 5.393,
      "mean_token_accuracy": 0.1933970719575882,
      "num_tokens": 28888981.0,
      "step": 12950
    },
    {
      "entropy": 5.460827445983886,
      "epoch": 1.454989610827203,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004798339588954025,
      "loss": 5.2587,
      "mean_token_accuracy": 0.19057833701372145,
      "num_tokens": 28899733.0,
      "step": 12955
    },
    {
      "entropy": 5.42146863937378,
      "epoch": 1.4555511877351603,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004798173425657903,
      "loss": 5.2516,
      "mean_token_accuracy": 0.1928494781255722,
      "num_tokens": 28910701.0,
      "step": 12960
    },
    {
      "entropy": 5.516459035873413,
      "epoch": 1.4561127646431178,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004798007197147144,
      "loss": 5.3255,
      "mean_token_accuracy": 0.18599175214767455,
      "num_tokens": 28922627.0,
      "step": 12965
    },
    {
      "entropy": 5.539343929290771,
      "epoch": 1.4566743415510754,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004797840903427042,
      "loss": 5.3582,
      "mean_token_accuracy": 0.19016606211662293,
      "num_tokens": 28933292.0,
      "step": 12970
    },
    {
      "entropy": 5.425318622589112,
      "epoch": 1.4572359184590329,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047976745445028906,
      "loss": 5.2292,
      "mean_token_accuracy": 0.1934402659535408,
      "num_tokens": 28943898.0,
      "step": 12975
    },
    {
      "entropy": 5.461353921890259,
      "epoch": 1.4577974953669905,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047975081203799883,
      "loss": 5.3243,
      "mean_token_accuracy": 0.19086166322231293,
      "num_tokens": 28955248.0,
      "step": 12980
    },
    {
      "entropy": 5.464456987380982,
      "epoch": 1.458359072274948,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047973416310636327,
      "loss": 5.2137,
      "mean_token_accuracy": 0.19882305711507797,
      "num_tokens": 28966905.0,
      "step": 12985
    },
    {
      "entropy": 5.506450462341308,
      "epoch": 1.4589206491829056,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004797175076559126,
      "loss": 5.3439,
      "mean_token_accuracy": 0.1928555890917778,
      "num_tokens": 28978515.0,
      "step": 12990
    },
    {
      "entropy": 5.426952695846557,
      "epoch": 1.459482226090863,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047970084568717703,
      "loss": 5.3111,
      "mean_token_accuracy": 0.18737949877977372,
      "num_tokens": 28990185.0,
      "step": 12995
    },
    {
      "entropy": 5.498533773422241,
      "epoch": 1.4600438029988208,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004796841772006872,
      "loss": 5.3377,
      "mean_token_accuracy": 0.1894269436597824,
      "num_tokens": 29000779.0,
      "step": 13000
    },
    {
      "entropy": 5.567147445678711,
      "epoch": 1.4606053799067782,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047966750219697376,
      "loss": 5.4434,
      "mean_token_accuracy": 0.1878877818584442,
      "num_tokens": 29012066.0,
      "step": 13005
    },
    {
      "entropy": 5.660548448562622,
      "epoch": 1.4611669568147359,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004796508206765677,
      "loss": 5.4279,
      "mean_token_accuracy": 0.18517484068870543,
      "num_tokens": 29023416.0,
      "step": 13010
    },
    {
      "entropy": 5.489505100250244,
      "epoch": 1.4617285337226933,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047963413264000005,
      "loss": 5.2847,
      "mean_token_accuracy": 0.19129559099674226,
      "num_tokens": 29034469.0,
      "step": 13015
    },
    {
      "entropy": 5.468514680862427,
      "epoch": 1.4622901106306507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004796174380878023,
      "loss": 5.2571,
      "mean_token_accuracy": 0.18737759739160537,
      "num_tokens": 29045306.0,
      "step": 13020
    },
    {
      "entropy": 5.424688482284546,
      "epoch": 1.4628516875386084,
      "grad_norm": 1.046875,
      "learning_rate": 0.000479600737020506,
      "loss": 5.2514,
      "mean_token_accuracy": 0.19463146477937698,
      "num_tokens": 29057114.0,
      "step": 13025
    },
    {
      "entropy": 5.459911918640136,
      "epoch": 1.463413264446566,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004795840294386429,
      "loss": 5.2317,
      "mean_token_accuracy": 0.18986470401287078,
      "num_tokens": 29067250.0,
      "step": 13030
    },
    {
      "entropy": 5.441561079025268,
      "epoch": 1.4639748413545235,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047956731534274483,
      "loss": 5.2798,
      "mean_token_accuracy": 0.19596178978681564,
      "num_tokens": 29078582.0,
      "step": 13035
    },
    {
      "entropy": 5.460959815979004,
      "epoch": 1.464536418262481,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047955059473334425,
      "loss": 5.3071,
      "mean_token_accuracy": 0.19390060752630234,
      "num_tokens": 29090113.0,
      "step": 13040
    },
    {
      "entropy": 5.419605445861817,
      "epoch": 1.4650979951704386,
      "grad_norm": 1.125,
      "learning_rate": 0.0004795338676109733,
      "loss": 5.2842,
      "mean_token_accuracy": 0.1960938185453415,
      "num_tokens": 29101117.0,
      "step": 13045
    },
    {
      "entropy": 5.534759855270385,
      "epoch": 1.4656595720783963,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004795171339761647,
      "loss": 5.3071,
      "mean_token_accuracy": 0.19129250943660736,
      "num_tokens": 29112569.0,
      "step": 13050
    },
    {
      "entropy": 5.562998580932617,
      "epoch": 1.4662211489863537,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047950039382945126,
      "loss": 5.4059,
      "mean_token_accuracy": 0.19032600075006484,
      "num_tokens": 29124462.0,
      "step": 13055
    },
    {
      "entropy": 5.4725611209869385,
      "epoch": 1.4667827258943111,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047948364717136587,
      "loss": 5.2544,
      "mean_token_accuracy": 0.19727838188409805,
      "num_tokens": 29135401.0,
      "step": 13060
    },
    {
      "entropy": 5.46421971321106,
      "epoch": 1.4673443028022688,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047946689400244196,
      "loss": 5.2914,
      "mean_token_accuracy": 0.18822217732667923,
      "num_tokens": 29146601.0,
      "step": 13065
    },
    {
      "entropy": 5.461003684997559,
      "epoch": 1.4679058797102262,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047945013432321284,
      "loss": 5.2827,
      "mean_token_accuracy": 0.19355401992797852,
      "num_tokens": 29157299.0,
      "step": 13070
    },
    {
      "entropy": 5.557260990142822,
      "epoch": 1.468467456618184,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047943336813421205,
      "loss": 5.3788,
      "mean_token_accuracy": 0.18798348754644395,
      "num_tokens": 29169101.0,
      "step": 13075
    },
    {
      "entropy": 5.402161026000977,
      "epoch": 1.4690290335261413,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004794165954359736,
      "loss": 5.1954,
      "mean_token_accuracy": 0.19632158428430557,
      "num_tokens": 29181286.0,
      "step": 13080
    },
    {
      "entropy": 5.496028709411621,
      "epoch": 1.469590610434099,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004793998162290315,
      "loss": 5.3725,
      "mean_token_accuracy": 0.18671411275863647,
      "num_tokens": 29191181.0,
      "step": 13085
    },
    {
      "entropy": 5.458876609802246,
      "epoch": 1.4701521873420564,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004793830305139199,
      "loss": 5.2256,
      "mean_token_accuracy": 0.1940981224179268,
      "num_tokens": 29202426.0,
      "step": 13090
    },
    {
      "entropy": 5.44552435874939,
      "epoch": 1.470713764250014,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004793662382911734,
      "loss": 5.2395,
      "mean_token_accuracy": 0.19542424231767655,
      "num_tokens": 29212817.0,
      "step": 13095
    },
    {
      "entropy": 5.447294235229492,
      "epoch": 1.4712753411579715,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004793494395613265,
      "loss": 5.2564,
      "mean_token_accuracy": 0.19038020670413972,
      "num_tokens": 29224644.0,
      "step": 13100
    },
    {
      "entropy": 5.465914535522461,
      "epoch": 1.4718369180659292,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004793326343249144,
      "loss": 5.278,
      "mean_token_accuracy": 0.1904280200600624,
      "num_tokens": 29236443.0,
      "step": 13105
    },
    {
      "entropy": 5.559574317932129,
      "epoch": 1.4723984949738866,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004793158225824718,
      "loss": 5.4181,
      "mean_token_accuracy": 0.18571652621030807,
      "num_tokens": 29247687.0,
      "step": 13110
    },
    {
      "entropy": 5.531558847427368,
      "epoch": 1.4729600718818443,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047929900433453423,
      "loss": 5.3187,
      "mean_token_accuracy": 0.19203899800777435,
      "num_tokens": 29259739.0,
      "step": 13115
    },
    {
      "entropy": 5.515462970733642,
      "epoch": 1.4735216487898017,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047928217958163703,
      "loss": 5.3534,
      "mean_token_accuracy": 0.19106705933809282,
      "num_tokens": 29270834.0,
      "step": 13120
    },
    {
      "entropy": 5.5195942401885985,
      "epoch": 1.4740832256977594,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047926534832431615,
      "loss": 5.3097,
      "mean_token_accuracy": 0.1902782216668129,
      "num_tokens": 29282291.0,
      "step": 13125
    },
    {
      "entropy": 5.539123392105102,
      "epoch": 1.4746448026057168,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004792485105631073,
      "loss": 5.3587,
      "mean_token_accuracy": 0.18881076127290725,
      "num_tokens": 29293836.0,
      "step": 13130
    },
    {
      "entropy": 5.499612712860108,
      "epoch": 1.4752063795136743,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004792316662985467,
      "loss": 5.3694,
      "mean_token_accuracy": 0.18158402442932128,
      "num_tokens": 29305550.0,
      "step": 13135
    },
    {
      "entropy": 5.549182415008545,
      "epoch": 1.475767956421632,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047921481553117053,
      "loss": 5.3601,
      "mean_token_accuracy": 0.18821000307798386,
      "num_tokens": 29316973.0,
      "step": 13140
    },
    {
      "entropy": 5.491867399215698,
      "epoch": 1.4763295333295896,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047919795826151544,
      "loss": 5.2121,
      "mean_token_accuracy": 0.20074503421783446,
      "num_tokens": 29327743.0,
      "step": 13145
    },
    {
      "entropy": 5.446516513824463,
      "epoch": 1.476891110237547,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004791810944901183,
      "loss": 5.2028,
      "mean_token_accuracy": 0.18895393311977388,
      "num_tokens": 29338286.0,
      "step": 13150
    },
    {
      "entropy": 5.45173192024231,
      "epoch": 1.4774526871455045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004791642242175158,
      "loss": 5.3388,
      "mean_token_accuracy": 0.1874464675784111,
      "num_tokens": 29348877.0,
      "step": 13155
    },
    {
      "entropy": 5.388352346420288,
      "epoch": 1.4780142640534621,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047914734744424517,
      "loss": 5.1175,
      "mean_token_accuracy": 0.20348843485116958,
      "num_tokens": 29360072.0,
      "step": 13160
    },
    {
      "entropy": 5.493221950531006,
      "epoch": 1.4785758409614196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047913046417084393,
      "loss": 5.3624,
      "mean_token_accuracy": 0.19179106056690215,
      "num_tokens": 29370744.0,
      "step": 13165
    },
    {
      "entropy": 5.541601181030273,
      "epoch": 1.4791374178693772,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047911357439784947,
      "loss": 5.3143,
      "mean_token_accuracy": 0.18578248023986815,
      "num_tokens": 29381463.0,
      "step": 13170
    },
    {
      "entropy": 5.493764734268188,
      "epoch": 1.4796989947773347,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047909667812579964,
      "loss": 5.3297,
      "mean_token_accuracy": 0.19343615770339967,
      "num_tokens": 29394227.0,
      "step": 13175
    },
    {
      "entropy": 5.433274221420288,
      "epoch": 1.4802605716852923,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047907977535523253,
      "loss": 5.2525,
      "mean_token_accuracy": 0.1978638008236885,
      "num_tokens": 29404754.0,
      "step": 13180
    },
    {
      "entropy": 5.422193717956543,
      "epoch": 1.4808221485932498,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004790628660866861,
      "loss": 5.2931,
      "mean_token_accuracy": 0.19936129003763198,
      "num_tokens": 29416161.0,
      "step": 13185
    },
    {
      "entropy": 5.415967893600464,
      "epoch": 1.4813837255012074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004790459503206989,
      "loss": 5.1809,
      "mean_token_accuracy": 0.20436224192380906,
      "num_tokens": 29426587.0,
      "step": 13190
    },
    {
      "entropy": 5.449272394180298,
      "epoch": 1.4819453024091649,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004790290280578095,
      "loss": 5.3465,
      "mean_token_accuracy": 0.18565037846565247,
      "num_tokens": 29437488.0,
      "step": 13195
    },
    {
      "entropy": 5.500899744033814,
      "epoch": 1.4825068793171226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047901209929855675,
      "loss": 5.322,
      "mean_token_accuracy": 0.19296833872795105,
      "num_tokens": 29448244.0,
      "step": 13200
    },
    {
      "entropy": 5.587065601348877,
      "epoch": 1.48306845622508,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004789951640434796,
      "loss": 5.3906,
      "mean_token_accuracy": 0.18762186765670777,
      "num_tokens": 29458260.0,
      "step": 13205
    },
    {
      "entropy": 5.605184030532837,
      "epoch": 1.4836300331330377,
      "grad_norm": 1.0,
      "learning_rate": 0.0004789782222931173,
      "loss": 5.3437,
      "mean_token_accuracy": 0.1924204185605049,
      "num_tokens": 29470874.0,
      "step": 13210
    },
    {
      "entropy": 5.439106416702271,
      "epoch": 1.484191610040995,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047896127404800935,
      "loss": 5.251,
      "mean_token_accuracy": 0.19099540412425994,
      "num_tokens": 29481442.0,
      "step": 13215
    },
    {
      "entropy": 5.515398836135864,
      "epoch": 1.4847531869489528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004789443193086953,
      "loss": 5.3977,
      "mean_token_accuracy": 0.1918204203248024,
      "num_tokens": 29492727.0,
      "step": 13220
    },
    {
      "entropy": 5.5656898021698,
      "epoch": 1.4853147638569102,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047892735807571505,
      "loss": 5.3295,
      "mean_token_accuracy": 0.1868348702788353,
      "num_tokens": 29502729.0,
      "step": 13225
    },
    {
      "entropy": 5.583171081542969,
      "epoch": 1.4858763407648676,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004789103903496086,
      "loss": 5.2811,
      "mean_token_accuracy": 0.19567034244537354,
      "num_tokens": 29514315.0,
      "step": 13230
    },
    {
      "entropy": 5.482412528991699,
      "epoch": 1.4864379176728253,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004788934161309161,
      "loss": 5.2285,
      "mean_token_accuracy": 0.19366237223148347,
      "num_tokens": 29526030.0,
      "step": 13235
    },
    {
      "entropy": 5.40715765953064,
      "epoch": 1.486999494580783,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047887643542017836,
      "loss": 5.2592,
      "mean_token_accuracy": 0.19387319236993789,
      "num_tokens": 29536972.0,
      "step": 13240
    },
    {
      "entropy": 5.432439184188842,
      "epoch": 1.4875610714887404,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047885944821793574,
      "loss": 5.2998,
      "mean_token_accuracy": 0.19370868653059006,
      "num_tokens": 29547325.0,
      "step": 13245
    },
    {
      "entropy": 5.506995725631714,
      "epoch": 1.4881226483966978,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047884245452472926,
      "loss": 5.3174,
      "mean_token_accuracy": 0.19362844675779342,
      "num_tokens": 29557944.0,
      "step": 13250
    },
    {
      "entropy": 5.386491537094116,
      "epoch": 1.4886842253046555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004788254543410999,
      "loss": 5.2346,
      "mean_token_accuracy": 0.20049775540828704,
      "num_tokens": 29568618.0,
      "step": 13255
    },
    {
      "entropy": 5.5095103740692135,
      "epoch": 1.489245802212613,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047880844766758914,
      "loss": 5.3153,
      "mean_token_accuracy": 0.1876966953277588,
      "num_tokens": 29580682.0,
      "step": 13260
    },
    {
      "entropy": 5.556739950180054,
      "epoch": 1.4898073791205706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047879143450473824,
      "loss": 5.3558,
      "mean_token_accuracy": 0.188737690448761,
      "num_tokens": 29592302.0,
      "step": 13265
    },
    {
      "entropy": 5.525094366073608,
      "epoch": 1.490368956028528,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047877441485308917,
      "loss": 5.3769,
      "mean_token_accuracy": 0.18692763149738312,
      "num_tokens": 29603591.0,
      "step": 13270
    },
    {
      "entropy": 5.480088710784912,
      "epoch": 1.4909305329364857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047875738871318356,
      "loss": 5.3068,
      "mean_token_accuracy": 0.19129645675420762,
      "num_tokens": 29615542.0,
      "step": 13275
    },
    {
      "entropy": 5.581087923049926,
      "epoch": 1.4914921098444431,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004787403560855637,
      "loss": 5.3518,
      "mean_token_accuracy": 0.19800916761159898,
      "num_tokens": 29626650.0,
      "step": 13280
    },
    {
      "entropy": 5.442483901977539,
      "epoch": 1.4920536867524008,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004787233169707719,
      "loss": 5.2925,
      "mean_token_accuracy": 0.19341592937707902,
      "num_tokens": 29637135.0,
      "step": 13285
    },
    {
      "entropy": 5.460017681121826,
      "epoch": 1.4926152636603582,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004787062713693507,
      "loss": 5.3307,
      "mean_token_accuracy": 0.18432952463626862,
      "num_tokens": 29648862.0,
      "step": 13290
    },
    {
      "entropy": 5.4734086990356445,
      "epoch": 1.493176840568316,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047868921928184276,
      "loss": 5.3015,
      "mean_token_accuracy": 0.19206209033727645,
      "num_tokens": 29661015.0,
      "step": 13295
    },
    {
      "entropy": 5.575405168533325,
      "epoch": 1.4937384174762733,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004786721607087911,
      "loss": 5.2945,
      "mean_token_accuracy": 0.1862231731414795,
      "num_tokens": 29673521.0,
      "step": 13300
    },
    {
      "entropy": 5.560371828079224,
      "epoch": 1.494299994384231,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004786550956507389,
      "loss": 5.3667,
      "mean_token_accuracy": 0.19226312190294265,
      "num_tokens": 29685379.0,
      "step": 13305
    },
    {
      "entropy": 5.430146837234497,
      "epoch": 1.4948615712921884,
      "grad_norm": 1.125,
      "learning_rate": 0.0004786380241082294,
      "loss": 5.2869,
      "mean_token_accuracy": 0.19056206047534943,
      "num_tokens": 29696372.0,
      "step": 13310
    },
    {
      "entropy": 5.481883001327515,
      "epoch": 1.495423148200146,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047862094608180623,
      "loss": 5.2985,
      "mean_token_accuracy": 0.1957177847623825,
      "num_tokens": 29707401.0,
      "step": 13315
    },
    {
      "entropy": 5.429145479202271,
      "epoch": 1.4959847251081035,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004786038615720132,
      "loss": 5.2746,
      "mean_token_accuracy": 0.19296327531337737,
      "num_tokens": 29717524.0,
      "step": 13320
    },
    {
      "entropy": 5.508581018447876,
      "epoch": 1.496546302016061,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047858677057939427,
      "loss": 5.3769,
      "mean_token_accuracy": 0.18857701867818832,
      "num_tokens": 29728235.0,
      "step": 13325
    },
    {
      "entropy": 5.580225276947021,
      "epoch": 1.4971078789240186,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047856967310449356,
      "loss": 5.3777,
      "mean_token_accuracy": 0.18802305459976196,
      "num_tokens": 29738800.0,
      "step": 13330
    },
    {
      "entropy": 5.5447821617126465,
      "epoch": 1.4976694558319763,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047855256914785556,
      "loss": 5.3214,
      "mean_token_accuracy": 0.19085756093263626,
      "num_tokens": 29750346.0,
      "step": 13335
    },
    {
      "entropy": 5.464204978942871,
      "epoch": 1.4982310327399337,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047853545871002476,
      "loss": 5.2588,
      "mean_token_accuracy": 0.19634692519903182,
      "num_tokens": 29761931.0,
      "step": 13340
    },
    {
      "entropy": 5.406882429122925,
      "epoch": 1.4987926096478912,
      "grad_norm": 1.0,
      "learning_rate": 0.00047851834179154595,
      "loss": 5.2309,
      "mean_token_accuracy": 0.19622450917959214,
      "num_tokens": 29773333.0,
      "step": 13345
    },
    {
      "entropy": 5.440138578414917,
      "epoch": 1.4993541865558488,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004785012183929643,
      "loss": 5.2149,
      "mean_token_accuracy": 0.19750139564275743,
      "num_tokens": 29785125.0,
      "step": 13350
    },
    {
      "entropy": 5.43392653465271,
      "epoch": 1.4999157634638065,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004784840885148249,
      "loss": 5.3337,
      "mean_token_accuracy": 0.1961698442697525,
      "num_tokens": 29796288.0,
      "step": 13355
    },
    {
      "entropy": 5.52727780342102,
      "epoch": 1.500477340371764,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004784669521576732,
      "loss": 5.3559,
      "mean_token_accuracy": 0.18841904401779175,
      "num_tokens": 29806348.0,
      "step": 13360
    },
    {
      "entropy": 5.5186844825744625,
      "epoch": 1.5010389172797214,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004784498093220549,
      "loss": 5.2493,
      "mean_token_accuracy": 0.19519648402929307,
      "num_tokens": 29816342.0,
      "step": 13365
    },
    {
      "entropy": 5.414639472961426,
      "epoch": 1.501600494187679,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004784326600085157,
      "loss": 5.2348,
      "mean_token_accuracy": 0.19205697178840636,
      "num_tokens": 29826789.0,
      "step": 13370
    },
    {
      "entropy": 5.472879409790039,
      "epoch": 1.5021620710956367,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047841550421760164,
      "loss": 5.3375,
      "mean_token_accuracy": 0.19075942933559417,
      "num_tokens": 29837633.0,
      "step": 13375
    },
    {
      "entropy": 5.540190505981445,
      "epoch": 1.502723648003594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047839834194985916,
      "loss": 5.3984,
      "mean_token_accuracy": 0.1893691673874855,
      "num_tokens": 29848498.0,
      "step": 13380
    },
    {
      "entropy": 5.5391168117523195,
      "epoch": 1.5032852249115516,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004783811732058345,
      "loss": 5.3596,
      "mean_token_accuracy": 0.18855499476194382,
      "num_tokens": 29859777.0,
      "step": 13385
    },
    {
      "entropy": 5.739963531494141,
      "epoch": 1.5038468018195092,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047836399798607446,
      "loss": 5.5347,
      "mean_token_accuracy": 0.1819843664765358,
      "num_tokens": 29873373.0,
      "step": 13390
    },
    {
      "entropy": 5.542487335205078,
      "epoch": 1.5044083787274667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047834681629112583,
      "loss": 5.301,
      "mean_token_accuracy": 0.19327073991298677,
      "num_tokens": 29884749.0,
      "step": 13395
    },
    {
      "entropy": 5.38848180770874,
      "epoch": 1.5049699556354241,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004783296281215357,
      "loss": 5.2235,
      "mean_token_accuracy": 0.19183147847652435,
      "num_tokens": 29895059.0,
      "step": 13400
    },
    {
      "entropy": 5.400285625457764,
      "epoch": 1.5055315325433818,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004783124334778513,
      "loss": 5.2841,
      "mean_token_accuracy": 0.1841924697160721,
      "num_tokens": 29906611.0,
      "step": 13405
    },
    {
      "entropy": 5.365358114242554,
      "epoch": 1.5060931094513395,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004782952323606202,
      "loss": 5.1492,
      "mean_token_accuracy": 0.20012490600347518,
      "num_tokens": 29918179.0,
      "step": 13410
    },
    {
      "entropy": 5.48297004699707,
      "epoch": 1.506654686359297,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004782780247703901,
      "loss": 5.3434,
      "mean_token_accuracy": 0.18956200480461122,
      "num_tokens": 29929612.0,
      "step": 13415
    },
    {
      "entropy": 5.483995485305786,
      "epoch": 1.5072162632672543,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004782608107077088,
      "loss": 5.2149,
      "mean_token_accuracy": 0.20151948630809785,
      "num_tokens": 29939569.0,
      "step": 13420
    },
    {
      "entropy": 5.46502571105957,
      "epoch": 1.507777840175212,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047824359017312445,
      "loss": 5.2477,
      "mean_token_accuracy": 0.19439188688993453,
      "num_tokens": 29950344.0,
      "step": 13425
    },
    {
      "entropy": 5.446194648742676,
      "epoch": 1.5083394170831697,
      "grad_norm": 1.125,
      "learning_rate": 0.0004782263631671854,
      "loss": 5.2439,
      "mean_token_accuracy": 0.20005802065134048,
      "num_tokens": 29960748.0,
      "step": 13430
    },
    {
      "entropy": 5.604156732559204,
      "epoch": 1.508900993991127,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047820912969044014,
      "loss": 5.4388,
      "mean_token_accuracy": 0.18356028199195862,
      "num_tokens": 29971678.0,
      "step": 13435
    },
    {
      "entropy": 5.423716306686401,
      "epoch": 1.5094625708990845,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047819188974343734,
      "loss": 5.2819,
      "mean_token_accuracy": 0.19111263155937194,
      "num_tokens": 29982477.0,
      "step": 13440
    },
    {
      "entropy": 5.465841484069824,
      "epoch": 1.5100241478070422,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004781746433267259,
      "loss": 5.3103,
      "mean_token_accuracy": 0.19847854226827621,
      "num_tokens": 29994539.0,
      "step": 13445
    },
    {
      "entropy": 5.5075523853302,
      "epoch": 1.5105857247149999,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047815739044085515,
      "loss": 5.2557,
      "mean_token_accuracy": 0.19905335009098052,
      "num_tokens": 30004821.0,
      "step": 13450
    },
    {
      "entropy": 5.5712090015411375,
      "epoch": 1.5111473016229573,
      "grad_norm": 1.125,
      "learning_rate": 0.00047814013108637423,
      "loss": 5.3214,
      "mean_token_accuracy": 0.1914004534482956,
      "num_tokens": 30016343.0,
      "step": 13455
    },
    {
      "entropy": 5.456988000869751,
      "epoch": 1.5117088785309147,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047812286526383274,
      "loss": 5.2216,
      "mean_token_accuracy": 0.20098562389612198,
      "num_tokens": 30027719.0,
      "step": 13460
    },
    {
      "entropy": 5.44575252532959,
      "epoch": 1.5122704554388724,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004781055929737804,
      "loss": 5.2425,
      "mean_token_accuracy": 0.19920252710580827,
      "num_tokens": 30037988.0,
      "step": 13465
    },
    {
      "entropy": 5.4822839260101315,
      "epoch": 1.51283203234683,
      "grad_norm": 1.125,
      "learning_rate": 0.00047808831421676724,
      "loss": 5.2336,
      "mean_token_accuracy": 0.19513847082853317,
      "num_tokens": 30049736.0,
      "step": 13470
    },
    {
      "entropy": 5.507877826690674,
      "epoch": 1.5133936092547875,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004780710289933435,
      "loss": 5.3559,
      "mean_token_accuracy": 0.1882481500506401,
      "num_tokens": 30061963.0,
      "step": 13475
    },
    {
      "entropy": 5.536148500442505,
      "epoch": 1.513955186162745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004780537373040593,
      "loss": 5.3019,
      "mean_token_accuracy": 0.19694520235061647,
      "num_tokens": 30072256.0,
      "step": 13480
    },
    {
      "entropy": 5.430942010879517,
      "epoch": 1.5145167630707026,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047803643914946537,
      "loss": 5.3161,
      "mean_token_accuracy": 0.19644870907068251,
      "num_tokens": 30084804.0,
      "step": 13485
    },
    {
      "entropy": 5.489186716079712,
      "epoch": 1.51507833997866,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004780191345301125,
      "loss": 5.3,
      "mean_token_accuracy": 0.1934249296784401,
      "num_tokens": 30096938.0,
      "step": 13490
    },
    {
      "entropy": 5.5410013675689695,
      "epoch": 1.5156399168866175,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047800182344655165,
      "loss": 5.2759,
      "mean_token_accuracy": 0.18755671232938767,
      "num_tokens": 30108233.0,
      "step": 13495
    },
    {
      "entropy": 5.491359090805053,
      "epoch": 1.5162014937945751,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477984505899334,
      "loss": 5.313,
      "mean_token_accuracy": 0.19041243493556975,
      "num_tokens": 30120246.0,
      "step": 13500
    },
    {
      "entropy": 5.469485569000244,
      "epoch": 1.5167630707025328,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047796718188901095,
      "loss": 5.135,
      "mean_token_accuracy": 0.20159560590982437,
      "num_tokens": 30131213.0,
      "step": 13505
    },
    {
      "entropy": 5.372662925720215,
      "epoch": 1.5173246476104902,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047794985141613413,
      "loss": 5.162,
      "mean_token_accuracy": 0.20038333684206008,
      "num_tokens": 30143300.0,
      "step": 13510
    },
    {
      "entropy": 5.432745695114136,
      "epoch": 1.5178862245184477,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047793251448125526,
      "loss": 5.2379,
      "mean_token_accuracy": 0.19180796593427657,
      "num_tokens": 30153794.0,
      "step": 13515
    },
    {
      "entropy": 5.483161783218383,
      "epoch": 1.5184478014264053,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004779151710849265,
      "loss": 5.3581,
      "mean_token_accuracy": 0.1866893783211708,
      "num_tokens": 30165152.0,
      "step": 13520
    },
    {
      "entropy": 5.492881536483765,
      "epoch": 1.519009378334363,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004778978212276999,
      "loss": 5.2931,
      "mean_token_accuracy": 0.19126631766557695,
      "num_tokens": 30176338.0,
      "step": 13525
    },
    {
      "entropy": 5.481426525115967,
      "epoch": 1.5195709552423204,
      "grad_norm": 1.125,
      "learning_rate": 0.0004778804649101281,
      "loss": 5.3264,
      "mean_token_accuracy": 0.18730773776769638,
      "num_tokens": 30187666.0,
      "step": 13530
    },
    {
      "entropy": 5.484125757217408,
      "epoch": 1.5201325321502779,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004778631021327636,
      "loss": 5.3525,
      "mean_token_accuracy": 0.18887080997228622,
      "num_tokens": 30198892.0,
      "step": 13535
    },
    {
      "entropy": 5.444531393051148,
      "epoch": 1.5206941090582355,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004778457328961592,
      "loss": 5.3102,
      "mean_token_accuracy": 0.18783968985080718,
      "num_tokens": 30210677.0,
      "step": 13540
    },
    {
      "entropy": 5.513345432281494,
      "epoch": 1.5212556859661932,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047782835720086797,
      "loss": 5.2577,
      "mean_token_accuracy": 0.2003664642572403,
      "num_tokens": 30221217.0,
      "step": 13545
    },
    {
      "entropy": 5.511929988861084,
      "epoch": 1.5218172628741506,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004778109750474432,
      "loss": 5.3866,
      "mean_token_accuracy": 0.18828341215848923,
      "num_tokens": 30233576.0,
      "step": 13550
    },
    {
      "entropy": 5.43354434967041,
      "epoch": 1.522378839782108,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004777935864364383,
      "loss": 5.215,
      "mean_token_accuracy": 0.19677537828683853,
      "num_tokens": 30244787.0,
      "step": 13555
    },
    {
      "entropy": 5.497850465774536,
      "epoch": 1.5229404166900657,
      "grad_norm": 1.09375,
      "learning_rate": 0.000477776191368407,
      "loss": 5.3034,
      "mean_token_accuracy": 0.19393382221460342,
      "num_tokens": 30256656.0,
      "step": 13560
    },
    {
      "entropy": 5.48765115737915,
      "epoch": 1.5235019935980234,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004777587898439031,
      "loss": 5.3694,
      "mean_token_accuracy": 0.19351017475128174,
      "num_tokens": 30267715.0,
      "step": 13565
    },
    {
      "entropy": 5.522074365615845,
      "epoch": 1.5240635705059808,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047774138186348073,
      "loss": 5.3456,
      "mean_token_accuracy": 0.18663860112428665,
      "num_tokens": 30278119.0,
      "step": 13570
    },
    {
      "entropy": 5.53045244216919,
      "epoch": 1.5246251474139383,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004777239674276941,
      "loss": 5.2563,
      "mean_token_accuracy": 0.1942450299859047,
      "num_tokens": 30289899.0,
      "step": 13575
    },
    {
      "entropy": 5.475974655151367,
      "epoch": 1.525186724321896,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004777065465370977,
      "loss": 5.2725,
      "mean_token_accuracy": 0.19403212517499924,
      "num_tokens": 30300799.0,
      "step": 13580
    },
    {
      "entropy": 5.604724645614624,
      "epoch": 1.5257483012298534,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004776891191922463,
      "loss": 5.4902,
      "mean_token_accuracy": 0.18343800008296968,
      "num_tokens": 30312955.0,
      "step": 13585
    },
    {
      "entropy": 5.53210973739624,
      "epoch": 1.5263098781378108,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004776716853936947,
      "loss": 5.266,
      "mean_token_accuracy": 0.1911618486046791,
      "num_tokens": 30323089.0,
      "step": 13590
    },
    {
      "entropy": 5.426593255996704,
      "epoch": 1.5268714550457685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000477654245141998,
      "loss": 5.2106,
      "mean_token_accuracy": 0.19920119941234588,
      "num_tokens": 30333926.0,
      "step": 13595
    },
    {
      "entropy": 5.377472257614135,
      "epoch": 1.5274330319537262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004776367984377116,
      "loss": 5.3193,
      "mean_token_accuracy": 0.19035524278879165,
      "num_tokens": 30346347.0,
      "step": 13600
    },
    {
      "entropy": 5.471315193176269,
      "epoch": 1.5279946088616836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047761934528139086,
      "loss": 5.2786,
      "mean_token_accuracy": 0.18975663930177689,
      "num_tokens": 30357757.0,
      "step": 13605
    },
    {
      "entropy": 5.624136877059937,
      "epoch": 1.528556185769641,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004776018856735916,
      "loss": 5.3609,
      "mean_token_accuracy": 0.19210373163223265,
      "num_tokens": 30369323.0,
      "step": 13610
    },
    {
      "entropy": 5.445489358901978,
      "epoch": 1.5291177626775987,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004775844196148698,
      "loss": 5.206,
      "mean_token_accuracy": 0.19743847995996475,
      "num_tokens": 30380335.0,
      "step": 13615
    },
    {
      "entropy": 5.415156698226928,
      "epoch": 1.5296793395855564,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047756694710578146,
      "loss": 5.2212,
      "mean_token_accuracy": 0.19508120268583298,
      "num_tokens": 30391338.0,
      "step": 13620
    },
    {
      "entropy": 5.4708761215209964,
      "epoch": 1.5302409164935138,
      "grad_norm": 1.015625,
      "learning_rate": 0.000477549468146883,
      "loss": 5.3326,
      "mean_token_accuracy": 0.18398072123527526,
      "num_tokens": 30402917.0,
      "step": 13625
    },
    {
      "entropy": 5.409539222717285,
      "epoch": 1.5308024934014712,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004775319827387309,
      "loss": 5.1283,
      "mean_token_accuracy": 0.1965908482670784,
      "num_tokens": 30412690.0,
      "step": 13630
    },
    {
      "entropy": 5.469643211364746,
      "epoch": 1.531364070309429,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004775144908818818,
      "loss": 5.3176,
      "mean_token_accuracy": 0.1923662856221199,
      "num_tokens": 30423020.0,
      "step": 13635
    },
    {
      "entropy": 5.639208745956421,
      "epoch": 1.5319256472173866,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047749699257689296,
      "loss": 5.54,
      "mean_token_accuracy": 0.1781605027616024,
      "num_tokens": 30434039.0,
      "step": 13640
    },
    {
      "entropy": 5.605772399902344,
      "epoch": 1.532487224125344,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004774794878243212,
      "loss": 5.3114,
      "mean_token_accuracy": 0.1892308309674263,
      "num_tokens": 30446743.0,
      "step": 13645
    },
    {
      "entropy": 5.502392721176148,
      "epoch": 1.5330488010333014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000477461976624724,
      "loss": 5.2872,
      "mean_token_accuracy": 0.19604190289974213,
      "num_tokens": 30456969.0,
      "step": 13650
    },
    {
      "entropy": 5.387057065963745,
      "epoch": 1.533610377941259,
      "grad_norm": 1.125,
      "learning_rate": 0.00047744445897865905,
      "loss": 5.3414,
      "mean_token_accuracy": 0.18829695284366607,
      "num_tokens": 30467497.0,
      "step": 13655
    },
    {
      "entropy": 5.50472354888916,
      "epoch": 1.5341719548492168,
      "grad_norm": 1.171875,
      "learning_rate": 0.000477426934886684,
      "loss": 5.3023,
      "mean_token_accuracy": 0.1921074241399765,
      "num_tokens": 30479428.0,
      "step": 13660
    },
    {
      "entropy": 5.499163389205933,
      "epoch": 1.5347335317571742,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047740940434935677,
      "loss": 5.2562,
      "mean_token_accuracy": 0.19616036117076874,
      "num_tokens": 30490193.0,
      "step": 13665
    },
    {
      "entropy": 5.537275123596191,
      "epoch": 1.5352951086651316,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004773918673672356,
      "loss": 5.3104,
      "mean_token_accuracy": 0.18778176605701447,
      "num_tokens": 30500311.0,
      "step": 13670
    },
    {
      "entropy": 5.450046491622925,
      "epoch": 1.5358566855730893,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047737432394087887,
      "loss": 5.23,
      "mean_token_accuracy": 0.19670809358358382,
      "num_tokens": 30512485.0,
      "step": 13675
    },
    {
      "entropy": 5.453216409683227,
      "epoch": 1.5364182624810467,
      "grad_norm": 1.125,
      "learning_rate": 0.00047735677407084525,
      "loss": 5.214,
      "mean_token_accuracy": 0.194830584526062,
      "num_tokens": 30523495.0,
      "step": 13680
    },
    {
      "entropy": 5.428583288192749,
      "epoch": 1.5369798393890042,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004773392177576934,
      "loss": 5.2787,
      "mean_token_accuracy": 0.19782847613096238,
      "num_tokens": 30533964.0,
      "step": 13685
    },
    {
      "entropy": 5.384706258773804,
      "epoch": 1.5375414162969618,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004773216550019824,
      "loss": 5.1667,
      "mean_token_accuracy": 0.19759590327739715,
      "num_tokens": 30544798.0,
      "step": 13690
    },
    {
      "entropy": 5.478674459457397,
      "epoch": 1.5381029932049195,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047730408580427136,
      "loss": 5.3466,
      "mean_token_accuracy": 0.19161437302827836,
      "num_tokens": 30555991.0,
      "step": 13695
    },
    {
      "entropy": 5.47773928642273,
      "epoch": 1.538664570112877,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047728651016511994,
      "loss": 5.1952,
      "mean_token_accuracy": 0.19581881463527678,
      "num_tokens": 30566848.0,
      "step": 13700
    },
    {
      "entropy": 5.470114231109619,
      "epoch": 1.5392261470208344,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047726892808508744,
      "loss": 5.2138,
      "mean_token_accuracy": 0.2048891678452492,
      "num_tokens": 30577361.0,
      "step": 13705
    },
    {
      "entropy": 5.453024816513062,
      "epoch": 1.539787723928792,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004772513395647339,
      "loss": 5.2808,
      "mean_token_accuracy": 0.1933975875377655,
      "num_tokens": 30587685.0,
      "step": 13710
    },
    {
      "entropy": 5.454155158996582,
      "epoch": 1.5403493008367497,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004772337446046192,
      "loss": 5.3298,
      "mean_token_accuracy": 0.19296918511390687,
      "num_tokens": 30599864.0,
      "step": 13715
    },
    {
      "entropy": 5.480147457122802,
      "epoch": 1.5409108777447071,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047721614320530366,
      "loss": 5.32,
      "mean_token_accuracy": 0.19130200743675232,
      "num_tokens": 30610885.0,
      "step": 13720
    },
    {
      "entropy": 5.495508050918579,
      "epoch": 1.5414724546526646,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047719853536734764,
      "loss": 5.2618,
      "mean_token_accuracy": 0.1925037369132042,
      "num_tokens": 30622537.0,
      "step": 13725
    },
    {
      "entropy": 5.572053241729736,
      "epoch": 1.5420340315606222,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004771809210913119,
      "loss": 5.4054,
      "mean_token_accuracy": 0.18633068799972535,
      "num_tokens": 30635045.0,
      "step": 13730
    },
    {
      "entropy": 5.450768852233887,
      "epoch": 1.54259560846858,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047716330037775724,
      "loss": 5.2344,
      "mean_token_accuracy": 0.18950365036725997,
      "num_tokens": 30645998.0,
      "step": 13735
    },
    {
      "entropy": 5.490896511077881,
      "epoch": 1.5431571853765373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004771456732272446,
      "loss": 5.2954,
      "mean_token_accuracy": 0.18911807537078856,
      "num_tokens": 30658744.0,
      "step": 13740
    },
    {
      "entropy": 5.440768098831176,
      "epoch": 1.5437187622844948,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047712803964033545,
      "loss": 5.1626,
      "mean_token_accuracy": 0.20175869762897491,
      "num_tokens": 30669420.0,
      "step": 13745
    },
    {
      "entropy": 5.343130588531494,
      "epoch": 1.5442803391924524,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047711039961759107,
      "loss": 5.2422,
      "mean_token_accuracy": 0.1955384060740471,
      "num_tokens": 30680037.0,
      "step": 13750
    },
    {
      "entropy": 5.526803350448608,
      "epoch": 1.54484191610041,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004770927531595731,
      "loss": 5.2926,
      "mean_token_accuracy": 0.1943487599492073,
      "num_tokens": 30691650.0,
      "step": 13755
    },
    {
      "entropy": 5.454730749130249,
      "epoch": 1.5454034930083675,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047707510026684353,
      "loss": 5.2969,
      "mean_token_accuracy": 0.19266333431005478,
      "num_tokens": 30701876.0,
      "step": 13760
    },
    {
      "entropy": 5.634919834136963,
      "epoch": 1.545965069916325,
      "grad_norm": 1.25,
      "learning_rate": 0.00047705744093996447,
      "loss": 5.5084,
      "mean_token_accuracy": 0.18438412100076676,
      "num_tokens": 30714484.0,
      "step": 13765
    },
    {
      "entropy": 5.493253707885742,
      "epoch": 1.5465266468242826,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000477039775179498,
      "loss": 5.2796,
      "mean_token_accuracy": 0.18781240284442902,
      "num_tokens": 30725185.0,
      "step": 13770
    },
    {
      "entropy": 5.450872564315796,
      "epoch": 1.54708822373224,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047702210298600684,
      "loss": 5.2825,
      "mean_token_accuracy": 0.19344881623983384,
      "num_tokens": 30735522.0,
      "step": 13775
    },
    {
      "entropy": 5.437634229660034,
      "epoch": 1.5476498006401975,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047700442436005356,
      "loss": 5.2977,
      "mean_token_accuracy": 0.19602542221546174,
      "num_tokens": 30746353.0,
      "step": 13780
    },
    {
      "entropy": 5.497179985046387,
      "epoch": 1.5482113775481552,
      "grad_norm": 1.140625,
      "learning_rate": 0.000476986739302201,
      "loss": 5.2472,
      "mean_token_accuracy": 0.19434531778097153,
      "num_tokens": 30756245.0,
      "step": 13785
    },
    {
      "entropy": 5.527978515625,
      "epoch": 1.5487729544561128,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047696904781301233,
      "loss": 5.3709,
      "mean_token_accuracy": 0.19312793463468553,
      "num_tokens": 30767387.0,
      "step": 13790
    },
    {
      "entropy": 5.480736446380615,
      "epoch": 1.5493345313640703,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004769513498930508,
      "loss": 5.3263,
      "mean_token_accuracy": 0.19317321628332138,
      "num_tokens": 30780140.0,
      "step": 13795
    },
    {
      "entropy": 5.465695142745972,
      "epoch": 1.5498961082720277,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047693364554288,
      "loss": 5.221,
      "mean_token_accuracy": 0.2014308288693428,
      "num_tokens": 30791137.0,
      "step": 13800
    },
    {
      "entropy": 5.399987125396729,
      "epoch": 1.5504576851799854,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047691593476306373,
      "loss": 5.2444,
      "mean_token_accuracy": 0.1990022599697113,
      "num_tokens": 30801708.0,
      "step": 13805
    },
    {
      "entropy": 5.493716144561768,
      "epoch": 1.551019262087943,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004768982175541656,
      "loss": 5.3142,
      "mean_token_accuracy": 0.190645994246006,
      "num_tokens": 30813519.0,
      "step": 13810
    },
    {
      "entropy": 5.555122327804566,
      "epoch": 1.5515808389959005,
      "grad_norm": 1.125,
      "learning_rate": 0.00047688049391675006,
      "loss": 5.3754,
      "mean_token_accuracy": 0.19263965040445327,
      "num_tokens": 30823786.0,
      "step": 13815
    },
    {
      "entropy": 5.435871315002442,
      "epoch": 1.552142415903858,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004768627638513813,
      "loss": 5.2836,
      "mean_token_accuracy": 0.1947065681219101,
      "num_tokens": 30834381.0,
      "step": 13820
    },
    {
      "entropy": 5.494338035583496,
      "epoch": 1.5527039928118156,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004768450273586237,
      "loss": 5.2442,
      "mean_token_accuracy": 0.1950327128171921,
      "num_tokens": 30844865.0,
      "step": 13825
    },
    {
      "entropy": 5.587636661529541,
      "epoch": 1.5532655697197733,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047682728443904223,
      "loss": 5.3394,
      "mean_token_accuracy": 0.19499900937080383,
      "num_tokens": 30857088.0,
      "step": 13830
    },
    {
      "entropy": 5.475209426879883,
      "epoch": 1.5538271466277307,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047680953509320176,
      "loss": 5.3698,
      "mean_token_accuracy": 0.18178119659423828,
      "num_tokens": 30868320.0,
      "step": 13835
    },
    {
      "entropy": 5.496590232849121,
      "epoch": 1.5543887235356881,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004767917793216674,
      "loss": 5.2778,
      "mean_token_accuracy": 0.19671133607625962,
      "num_tokens": 30878598.0,
      "step": 13840
    },
    {
      "entropy": 5.479313516616822,
      "epoch": 1.5549503004436458,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047677401712500455,
      "loss": 5.2127,
      "mean_token_accuracy": 0.19437508583068847,
      "num_tokens": 30889254.0,
      "step": 13845
    },
    {
      "entropy": 5.476692724227905,
      "epoch": 1.5555118773516035,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004767562485037787,
      "loss": 5.3502,
      "mean_token_accuracy": 0.1947739079594612,
      "num_tokens": 30901164.0,
      "step": 13850
    },
    {
      "entropy": 5.432692766189575,
      "epoch": 1.556073454259561,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047673847345855566,
      "loss": 5.2461,
      "mean_token_accuracy": 0.19266577064990997,
      "num_tokens": 30911828.0,
      "step": 13855
    },
    {
      "entropy": 5.457504558563232,
      "epoch": 1.5566350311675183,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004767206919899014,
      "loss": 5.1374,
      "mean_token_accuracy": 0.20487585067749023,
      "num_tokens": 30922050.0,
      "step": 13860
    },
    {
      "entropy": 5.384016752243042,
      "epoch": 1.557196608075476,
      "grad_norm": 1.09375,
      "learning_rate": 0.000476702904098382,
      "loss": 5.2087,
      "mean_token_accuracy": 0.19853302389383315,
      "num_tokens": 30932723.0,
      "step": 13865
    },
    {
      "entropy": 5.432240676879883,
      "epoch": 1.5577581849834334,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000476685109784564,
      "loss": 5.2371,
      "mean_token_accuracy": 0.19728236943483352,
      "num_tokens": 30943699.0,
      "step": 13870
    },
    {
      "entropy": 5.456026458740235,
      "epoch": 1.5583197618913909,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047666730904901377,
      "loss": 5.2934,
      "mean_token_accuracy": 0.1911129340529442,
      "num_tokens": 30954315.0,
      "step": 13875
    },
    {
      "entropy": 5.5283287525177,
      "epoch": 1.5588813387993485,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047664950189229833,
      "loss": 5.4669,
      "mean_token_accuracy": 0.18846064060926437,
      "num_tokens": 30965840.0,
      "step": 13880
    },
    {
      "entropy": 5.599387454986572,
      "epoch": 1.5594429157073062,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004766316883149845,
      "loss": 5.3613,
      "mean_token_accuracy": 0.19338658899068834,
      "num_tokens": 30977051.0,
      "step": 13885
    },
    {
      "entropy": 5.436714315414429,
      "epoch": 1.5600044926152636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047661386831763955,
      "loss": 5.2007,
      "mean_token_accuracy": 0.19587831646203996,
      "num_tokens": 30988730.0,
      "step": 13890
    },
    {
      "entropy": 5.416201066970825,
      "epoch": 1.560566069523221,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047659604190083076,
      "loss": 5.2325,
      "mean_token_accuracy": 0.19640145897865297,
      "num_tokens": 31000571.0,
      "step": 13895
    },
    {
      "entropy": 5.421653079986572,
      "epoch": 1.5611276464311787,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047657820906512585,
      "loss": 5.2311,
      "mean_token_accuracy": 0.2024218738079071,
      "num_tokens": 31011003.0,
      "step": 13900
    },
    {
      "entropy": 5.508617877960205,
      "epoch": 1.5616892233391364,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047656036981109256,
      "loss": 5.259,
      "mean_token_accuracy": 0.19444389045238494,
      "num_tokens": 31022510.0,
      "step": 13905
    },
    {
      "entropy": 5.512660360336303,
      "epoch": 1.5622508002470938,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047654252413929896,
      "loss": 5.3107,
      "mean_token_accuracy": 0.1909449055790901,
      "num_tokens": 31033149.0,
      "step": 13910
    },
    {
      "entropy": 5.562669801712036,
      "epoch": 1.5628123771550513,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047652467205031317,
      "loss": 5.3444,
      "mean_token_accuracy": 0.19989275485277175,
      "num_tokens": 31043626.0,
      "step": 13915
    },
    {
      "entropy": 5.5153059482574465,
      "epoch": 1.563373954063009,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047650681354470363,
      "loss": 5.3888,
      "mean_token_accuracy": 0.19250257909297944,
      "num_tokens": 31055083.0,
      "step": 13920
    },
    {
      "entropy": 5.455924987792969,
      "epoch": 1.5639355309709666,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047648894862303907,
      "loss": 5.2346,
      "mean_token_accuracy": 0.19278991222381592,
      "num_tokens": 31066763.0,
      "step": 13925
    },
    {
      "entropy": 5.484093093872071,
      "epoch": 1.564497107878924,
      "grad_norm": 1.125,
      "learning_rate": 0.00047647107728588826,
      "loss": 5.2875,
      "mean_token_accuracy": 0.1909972682595253,
      "num_tokens": 31078722.0,
      "step": 13930
    },
    {
      "entropy": 5.493951749801636,
      "epoch": 1.5650586847868815,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047645319953382024,
      "loss": 5.3446,
      "mean_token_accuracy": 0.19772130846977234,
      "num_tokens": 31090207.0,
      "step": 13935
    },
    {
      "entropy": 5.5285381317138675,
      "epoch": 1.5656202616948391,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047643531536740405,
      "loss": 5.3663,
      "mean_token_accuracy": 0.19416908472776412,
      "num_tokens": 31101943.0,
      "step": 13940
    },
    {
      "entropy": 5.508974838256836,
      "epoch": 1.5661818386027968,
      "grad_norm": 1.125,
      "learning_rate": 0.0004764174247872094,
      "loss": 5.3699,
      "mean_token_accuracy": 0.18728849440813064,
      "num_tokens": 31115032.0,
      "step": 13945
    },
    {
      "entropy": 5.56921796798706,
      "epoch": 1.5667434155107542,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047639952779380586,
      "loss": 5.2785,
      "mean_token_accuracy": 0.19570743590593337,
      "num_tokens": 31124965.0,
      "step": 13950
    },
    {
      "entropy": 5.532122325897217,
      "epoch": 1.5673049924187117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004763816243877632,
      "loss": 5.2999,
      "mean_token_accuracy": 0.19396822452545165,
      "num_tokens": 31136692.0,
      "step": 13955
    },
    {
      "entropy": 5.481381416320801,
      "epoch": 1.5678665693266693,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004763637145696515,
      "loss": 5.293,
      "mean_token_accuracy": 0.18791648298501967,
      "num_tokens": 31148352.0,
      "step": 13960
    },
    {
      "entropy": 5.4693190574646,
      "epoch": 1.5684281462346268,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000476345798340041,
      "loss": 5.2338,
      "mean_token_accuracy": 0.19450350254774093,
      "num_tokens": 31159892.0,
      "step": 13965
    },
    {
      "entropy": 5.555439901351929,
      "epoch": 1.5689897231425842,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004763278756995022,
      "loss": 5.4015,
      "mean_token_accuracy": 0.18517508506774902,
      "num_tokens": 31170087.0,
      "step": 13970
    },
    {
      "entropy": 5.374926376342773,
      "epoch": 1.5695513000505419,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047630994664860576,
      "loss": 5.1999,
      "mean_token_accuracy": 0.19553141742944719,
      "num_tokens": 31180687.0,
      "step": 13975
    },
    {
      "entropy": 5.544255304336548,
      "epoch": 1.5701128769584995,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004762920111879226,
      "loss": 5.3401,
      "mean_token_accuracy": 0.1890310525894165,
      "num_tokens": 31192513.0,
      "step": 13980
    },
    {
      "entropy": 5.459659147262573,
      "epoch": 1.570674453866457,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004762740693180237,
      "loss": 5.2655,
      "mean_token_accuracy": 0.19225750267505645,
      "num_tokens": 31203432.0,
      "step": 13985
    },
    {
      "entropy": 5.568687295913696,
      "epoch": 1.5712360307744144,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047625612103948035,
      "loss": 5.2506,
      "mean_token_accuracy": 0.1983028993010521,
      "num_tokens": 31214474.0,
      "step": 13990
    },
    {
      "entropy": 5.50894193649292,
      "epoch": 1.571797607682372,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004762381663528641,
      "loss": 5.3206,
      "mean_token_accuracy": 0.1880207508802414,
      "num_tokens": 31226489.0,
      "step": 13995
    },
    {
      "entropy": 5.356471872329712,
      "epoch": 1.5723591845903298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004762202052587466,
      "loss": 5.1894,
      "mean_token_accuracy": 0.19300511181354524,
      "num_tokens": 31237568.0,
      "step": 14000
    },
    {
      "entropy": 5.421355724334717,
      "epoch": 1.5729207614982872,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004762022377576996,
      "loss": 5.2111,
      "mean_token_accuracy": 0.20428184866905214,
      "num_tokens": 31248587.0,
      "step": 14005
    },
    {
      "entropy": 5.501969671249389,
      "epoch": 1.5734823384062446,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004761842638502954,
      "loss": 5.3303,
      "mean_token_accuracy": 0.18805623501539231,
      "num_tokens": 31259513.0,
      "step": 14010
    },
    {
      "entropy": 5.521801328659057,
      "epoch": 1.5740439153142023,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004761662835371062,
      "loss": 5.3141,
      "mean_token_accuracy": 0.18917859792709352,
      "num_tokens": 31270984.0,
      "step": 14015
    },
    {
      "entropy": 5.44963321685791,
      "epoch": 1.57460549222216,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004761482968187045,
      "loss": 5.2638,
      "mean_token_accuracy": 0.19760411977767944,
      "num_tokens": 31281907.0,
      "step": 14020
    },
    {
      "entropy": 5.3613365650177,
      "epoch": 1.5751670691301174,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004761303036956631,
      "loss": 5.3019,
      "mean_token_accuracy": 0.19272556900978088,
      "num_tokens": 31292903.0,
      "step": 14025
    },
    {
      "entropy": 5.535918760299682,
      "epoch": 1.5757286460380748,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004761123041685547,
      "loss": 5.312,
      "mean_token_accuracy": 0.19466183632612227,
      "num_tokens": 31304185.0,
      "step": 14030
    },
    {
      "entropy": 5.525847482681274,
      "epoch": 1.5762902229460325,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004760942982379526,
      "loss": 5.3092,
      "mean_token_accuracy": 0.1920395791530609,
      "num_tokens": 31314806.0,
      "step": 14035
    },
    {
      "entropy": 5.479031562805176,
      "epoch": 1.5768517998539902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047607628590443004,
      "loss": 5.3115,
      "mean_token_accuracy": 0.1921336606144905,
      "num_tokens": 31326361.0,
      "step": 14040
    },
    {
      "entropy": 5.527260398864746,
      "epoch": 1.5774133767619476,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047605826716856056,
      "loss": 5.2866,
      "mean_token_accuracy": 0.1968590497970581,
      "num_tokens": 31336549.0,
      "step": 14045
    },
    {
      "entropy": 5.565581274032593,
      "epoch": 1.577974953669905,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004760402420309178,
      "loss": 5.3934,
      "mean_token_accuracy": 0.1887447103857994,
      "num_tokens": 31346994.0,
      "step": 14050
    },
    {
      "entropy": 5.513060712814331,
      "epoch": 1.5785365305778627,
      "grad_norm": 1.125,
      "learning_rate": 0.0004760222104920759,
      "loss": 5.3006,
      "mean_token_accuracy": 0.19068933874368668,
      "num_tokens": 31358449.0,
      "step": 14055
    },
    {
      "entropy": 5.538665771484375,
      "epoch": 1.5790981074858201,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004760041725526088,
      "loss": 5.3417,
      "mean_token_accuracy": 0.1894978925585747,
      "num_tokens": 31368964.0,
      "step": 14060
    },
    {
      "entropy": 5.506728410720825,
      "epoch": 1.5796596843937776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047598612821309077,
      "loss": 5.269,
      "mean_token_accuracy": 0.19515904784202576,
      "num_tokens": 31379388.0,
      "step": 14065
    },
    {
      "entropy": 5.44234070777893,
      "epoch": 1.5802212613017352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004759680774740966,
      "loss": 5.3204,
      "mean_token_accuracy": 0.19756012856960298,
      "num_tokens": 31390757.0,
      "step": 14070
    },
    {
      "entropy": 5.432030057907104,
      "epoch": 1.580782838209693,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004759500203362008,
      "loss": 5.2781,
      "mean_token_accuracy": 0.1921206921339035,
      "num_tokens": 31402309.0,
      "step": 14075
    },
    {
      "entropy": 5.526041746139526,
      "epoch": 1.5813444151176503,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004759319567999785,
      "loss": 5.289,
      "mean_token_accuracy": 0.1956588089466095,
      "num_tokens": 31412559.0,
      "step": 14080
    },
    {
      "entropy": 5.511046600341797,
      "epoch": 1.5819059920256078,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004759138868660048,
      "loss": 5.4156,
      "mean_token_accuracy": 0.18994818180799483,
      "num_tokens": 31424982.0,
      "step": 14085
    },
    {
      "entropy": 5.478686332702637,
      "epoch": 1.5824675689335654,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047589581053485494,
      "loss": 5.1467,
      "mean_token_accuracy": 0.1976684331893921,
      "num_tokens": 31434472.0,
      "step": 14090
    },
    {
      "entropy": 5.516457509994507,
      "epoch": 1.583029145841523,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047587772780710455,
      "loss": 5.3694,
      "mean_token_accuracy": 0.19179917573928834,
      "num_tokens": 31445458.0,
      "step": 14095
    },
    {
      "entropy": 5.471166610717773,
      "epoch": 1.5835907227494805,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047585963868332945,
      "loss": 5.2447,
      "mean_token_accuracy": 0.20258166491985322,
      "num_tokens": 31455851.0,
      "step": 14100
    },
    {
      "entropy": 5.371727085113525,
      "epoch": 1.584152299657438,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004758415431641055,
      "loss": 5.209,
      "mean_token_accuracy": 0.19591001570224761,
      "num_tokens": 31466279.0,
      "step": 14105
    },
    {
      "entropy": 5.527924585342407,
      "epoch": 1.5847138765653956,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004758234412500089,
      "loss": 5.2871,
      "mean_token_accuracy": 0.19851753860712051,
      "num_tokens": 31477933.0,
      "step": 14110
    },
    {
      "entropy": 5.533932399749756,
      "epoch": 1.5852754534733533,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047580533294161616,
      "loss": 5.3294,
      "mean_token_accuracy": 0.1907680407166481,
      "num_tokens": 31487640.0,
      "step": 14115
    },
    {
      "entropy": 5.529735708236695,
      "epoch": 1.5858370303813107,
      "grad_norm": 1.25,
      "learning_rate": 0.0004757872182395036,
      "loss": 5.2721,
      "mean_token_accuracy": 0.19871703535318375,
      "num_tokens": 31499059.0,
      "step": 14120
    },
    {
      "entropy": 5.496701622009278,
      "epoch": 1.5863986072892682,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047576909714424823,
      "loss": 5.2879,
      "mean_token_accuracy": 0.19345563650131226,
      "num_tokens": 31510264.0,
      "step": 14125
    },
    {
      "entropy": 5.493211460113526,
      "epoch": 1.5869601841972258,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047575096965642687,
      "loss": 5.2804,
      "mean_token_accuracy": 0.19585746228694917,
      "num_tokens": 31520873.0,
      "step": 14130
    },
    {
      "entropy": 5.481922245025634,
      "epoch": 1.5875217611051835,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047573283577661684,
      "loss": 5.2984,
      "mean_token_accuracy": 0.19228129982948303,
      "num_tokens": 31533836.0,
      "step": 14135
    },
    {
      "entropy": 5.43515305519104,
      "epoch": 1.588083338013141,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004757146955053955,
      "loss": 5.301,
      "mean_token_accuracy": 0.19055052399635314,
      "num_tokens": 31545091.0,
      "step": 14140
    },
    {
      "entropy": 5.472206687927246,
      "epoch": 1.5886449149210984,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047569654884334036,
      "loss": 5.208,
      "mean_token_accuracy": 0.1969258964061737,
      "num_tokens": 31557139.0,
      "step": 14145
    },
    {
      "entropy": 5.504077005386352,
      "epoch": 1.589206491829056,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004756783957910292,
      "loss": 5.3078,
      "mean_token_accuracy": 0.1930576056241989,
      "num_tokens": 31567928.0,
      "step": 14150
    },
    {
      "entropy": 5.419829034805298,
      "epoch": 1.5897680687370135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047566023634904015,
      "loss": 5.3135,
      "mean_token_accuracy": 0.19434235394001007,
      "num_tokens": 31578081.0,
      "step": 14155
    },
    {
      "entropy": 5.4468381881713865,
      "epoch": 1.590329645644971,
      "grad_norm": 1.125,
      "learning_rate": 0.0004756420705179513,
      "loss": 5.226,
      "mean_token_accuracy": 0.19410833418369294,
      "num_tokens": 31589648.0,
      "step": 14160
    },
    {
      "entropy": 5.523640060424805,
      "epoch": 1.5908912225529286,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004756238982983412,
      "loss": 5.2843,
      "mean_token_accuracy": 0.1986084297299385,
      "num_tokens": 31601085.0,
      "step": 14165
    },
    {
      "entropy": 5.5377849578857425,
      "epoch": 1.5914527994608862,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004756057196907883,
      "loss": 5.3494,
      "mean_token_accuracy": 0.19505300372838974,
      "num_tokens": 31612005.0,
      "step": 14170
    },
    {
      "entropy": 5.507797145843506,
      "epoch": 1.5920143763688437,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047558753469587154,
      "loss": 5.319,
      "mean_token_accuracy": 0.19059222042560578,
      "num_tokens": 31623427.0,
      "step": 14175
    },
    {
      "entropy": 5.606068420410156,
      "epoch": 1.5925759532768011,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047556934331416986,
      "loss": 5.4492,
      "mean_token_accuracy": 0.18563432544469832,
      "num_tokens": 31635086.0,
      "step": 14180
    },
    {
      "entropy": 5.5014121532440186,
      "epoch": 1.5931375301847588,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004755511455462624,
      "loss": 5.2893,
      "mean_token_accuracy": 0.19231295138597487,
      "num_tokens": 31646707.0,
      "step": 14185
    },
    {
      "entropy": 5.371788024902344,
      "epoch": 1.5936991070927164,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004755329413927287,
      "loss": 5.1594,
      "mean_token_accuracy": 0.19806861728429795,
      "num_tokens": 31657327.0,
      "step": 14190
    },
    {
      "entropy": 5.447383260726928,
      "epoch": 1.5942606840006739,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004755147308541485,
      "loss": 5.1941,
      "mean_token_accuracy": 0.194957172870636,
      "num_tokens": 31668075.0,
      "step": 14195
    },
    {
      "entropy": 5.505075407028198,
      "epoch": 1.5948222609086313,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047549651393110145,
      "loss": 5.3298,
      "mean_token_accuracy": 0.18746793121099473,
      "num_tokens": 31680270.0,
      "step": 14200
    },
    {
      "entropy": 5.453749942779541,
      "epoch": 1.595383837816589,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047547829062416757,
      "loss": 5.1678,
      "mean_token_accuracy": 0.20338918566703795,
      "num_tokens": 31691271.0,
      "step": 14205
    },
    {
      "entropy": 5.3994364738464355,
      "epoch": 1.5959454147245467,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047546006093392726,
      "loss": 5.1951,
      "mean_token_accuracy": 0.19713371396064758,
      "num_tokens": 31702982.0,
      "step": 14210
    },
    {
      "entropy": 5.453500843048095,
      "epoch": 1.596506991632504,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004754418248609608,
      "loss": 5.3336,
      "mean_token_accuracy": 0.19095061868429183,
      "num_tokens": 31715129.0,
      "step": 14215
    },
    {
      "entropy": 5.560067462921142,
      "epoch": 1.5970685685404615,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004754235824058489,
      "loss": 5.3785,
      "mean_token_accuracy": 0.19212473332881927,
      "num_tokens": 31726392.0,
      "step": 14220
    },
    {
      "entropy": 5.587955665588379,
      "epoch": 1.5976301454484192,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004754053335691724,
      "loss": 5.3388,
      "mean_token_accuracy": 0.19493759274482728,
      "num_tokens": 31738143.0,
      "step": 14225
    },
    {
      "entropy": 5.557267475128174,
      "epoch": 1.5981917223563769,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047538707835151237,
      "loss": 5.3105,
      "mean_token_accuracy": 0.18792663365602494,
      "num_tokens": 31748510.0,
      "step": 14230
    },
    {
      "entropy": 5.346973037719726,
      "epoch": 1.5987532992643343,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047536881675345,
      "loss": 5.1956,
      "mean_token_accuracy": 0.19649087637662888,
      "num_tokens": 31759378.0,
      "step": 14235
    },
    {
      "entropy": 5.52656397819519,
      "epoch": 1.5993148761722917,
      "grad_norm": 1.125,
      "learning_rate": 0.0004753505487755669,
      "loss": 5.4153,
      "mean_token_accuracy": 0.18594098836183548,
      "num_tokens": 31771627.0,
      "step": 14240
    },
    {
      "entropy": 5.472970390319825,
      "epoch": 1.5998764530802494,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047533227441844453,
      "loss": 5.2136,
      "mean_token_accuracy": 0.20129866302013397,
      "num_tokens": 31781990.0,
      "step": 14245
    },
    {
      "entropy": 5.496505451202393,
      "epoch": 1.6004380299882068,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047531399368266485,
      "loss": 5.2773,
      "mean_token_accuracy": 0.19124405235052108,
      "num_tokens": 31792144.0,
      "step": 14250
    },
    {
      "entropy": 5.490231895446778,
      "epoch": 1.6009996068961643,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004752957065688099,
      "loss": 5.3072,
      "mean_token_accuracy": 0.19637819677591323,
      "num_tokens": 31804022.0,
      "step": 14255
    },
    {
      "entropy": 5.553983306884765,
      "epoch": 1.601561183804122,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000475277413077462,
      "loss": 5.3417,
      "mean_token_accuracy": 0.1905074432492256,
      "num_tokens": 31815108.0,
      "step": 14260
    },
    {
      "entropy": 5.424648475646973,
      "epoch": 1.6021227607120796,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047525911320920363,
      "loss": 5.3117,
      "mean_token_accuracy": 0.19563350677490235,
      "num_tokens": 31825727.0,
      "step": 14265
    },
    {
      "entropy": 5.419054317474365,
      "epoch": 1.602684337620037,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004752408069646174,
      "loss": 5.2736,
      "mean_token_accuracy": 0.20331913679838182,
      "num_tokens": 31836168.0,
      "step": 14270
    },
    {
      "entropy": 5.577020692825317,
      "epoch": 1.6032459145279945,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047522249434428623,
      "loss": 5.3639,
      "mean_token_accuracy": 0.19066408425569534,
      "num_tokens": 31847383.0,
      "step": 14275
    },
    {
      "entropy": 5.492323684692383,
      "epoch": 1.6038074914359521,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004752041753487931,
      "loss": 5.2019,
      "mean_token_accuracy": 0.2013877347111702,
      "num_tokens": 31858679.0,
      "step": 14280
    },
    {
      "entropy": 5.437363386154175,
      "epoch": 1.6043690683439098,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004751858499787215,
      "loss": 5.2859,
      "mean_token_accuracy": 0.19410343170166017,
      "num_tokens": 31870195.0,
      "step": 14285
    },
    {
      "entropy": 5.523854541778564,
      "epoch": 1.6049306452518672,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004751675182346547,
      "loss": 5.2942,
      "mean_token_accuracy": 0.19254668056964874,
      "num_tokens": 31881175.0,
      "step": 14290
    },
    {
      "entropy": 5.435376453399658,
      "epoch": 1.6054922221598247,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047514918011717656,
      "loss": 5.2271,
      "mean_token_accuracy": 0.20184947550296783,
      "num_tokens": 31891199.0,
      "step": 14295
    },
    {
      "entropy": 5.619966793060303,
      "epoch": 1.6060537990677823,
      "grad_norm": 1.125,
      "learning_rate": 0.00047513083562687085,
      "loss": 5.4093,
      "mean_token_accuracy": 0.1852930799126625,
      "num_tokens": 31903379.0,
      "step": 14300
    },
    {
      "entropy": 5.543605899810791,
      "epoch": 1.60661537597574,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004751124847643217,
      "loss": 5.2089,
      "mean_token_accuracy": 0.19692232012748717,
      "num_tokens": 31913958.0,
      "step": 14305
    },
    {
      "entropy": 5.388770532608032,
      "epoch": 1.6071769528836974,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004750941275301135,
      "loss": 5.1874,
      "mean_token_accuracy": 0.1972719430923462,
      "num_tokens": 31924547.0,
      "step": 14310
    },
    {
      "entropy": 5.561536169052124,
      "epoch": 1.6077385297916549,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004750757639248306,
      "loss": 5.3863,
      "mean_token_accuracy": 0.1887125328183174,
      "num_tokens": 31936987.0,
      "step": 14315
    },
    {
      "entropy": 5.4675726890563965,
      "epoch": 1.6083001066996125,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047505739394905776,
      "loss": 5.2395,
      "mean_token_accuracy": 0.19213483482599258,
      "num_tokens": 31948165.0,
      "step": 14320
    },
    {
      "entropy": 5.514142751693726,
      "epoch": 1.6088616836075702,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047503901760338003,
      "loss": 5.3111,
      "mean_token_accuracy": 0.19153206348419188,
      "num_tokens": 31959206.0,
      "step": 14325
    },
    {
      "entropy": 5.5301055908203125,
      "epoch": 1.6094232605155276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047502063488838225,
      "loss": 5.294,
      "mean_token_accuracy": 0.19198443591594697,
      "num_tokens": 31970804.0,
      "step": 14330
    },
    {
      "entropy": 5.441630125045776,
      "epoch": 1.609984837423485,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047500224580464994,
      "loss": 5.2058,
      "mean_token_accuracy": 0.19461580365896225,
      "num_tokens": 31982961.0,
      "step": 14335
    },
    {
      "entropy": 5.503299951553345,
      "epoch": 1.6105464143314427,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047498385035276856,
      "loss": 5.2575,
      "mean_token_accuracy": 0.1995654433965683,
      "num_tokens": 31993812.0,
      "step": 14340
    },
    {
      "entropy": 5.431442356109619,
      "epoch": 1.6111079912394002,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004749654485333238,
      "loss": 5.3128,
      "mean_token_accuracy": 0.19547715336084365,
      "num_tokens": 32004688.0,
      "step": 14345
    },
    {
      "entropy": 5.433307743072509,
      "epoch": 1.6116695681473576,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047494704034690163,
      "loss": 5.2032,
      "mean_token_accuracy": 0.19612354934215545,
      "num_tokens": 32014653.0,
      "step": 14350
    },
    {
      "entropy": 5.449719381332398,
      "epoch": 1.6122311450553153,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047492862579408816,
      "loss": 5.2579,
      "mean_token_accuracy": 0.19029328227043152,
      "num_tokens": 32025638.0,
      "step": 14355
    },
    {
      "entropy": 5.34260835647583,
      "epoch": 1.612792721963273,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047491020487546966,
      "loss": 5.1635,
      "mean_token_accuracy": 0.19991593956947326,
      "num_tokens": 32036316.0,
      "step": 14360
    },
    {
      "entropy": 5.438609409332275,
      "epoch": 1.6133542988712304,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004748917775916327,
      "loss": 5.1995,
      "mean_token_accuracy": 0.19836467802524566,
      "num_tokens": 32047355.0,
      "step": 14365
    },
    {
      "entropy": 5.4777916431427,
      "epoch": 1.6139158757791878,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047487334394316405,
      "loss": 5.3065,
      "mean_token_accuracy": 0.18453036099672318,
      "num_tokens": 32058897.0,
      "step": 14370
    },
    {
      "entropy": 5.487823629379273,
      "epoch": 1.6144774526871455,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004748549039306506,
      "loss": 5.3024,
      "mean_token_accuracy": 0.19205862283706665,
      "num_tokens": 32069932.0,
      "step": 14375
    },
    {
      "entropy": 5.49405083656311,
      "epoch": 1.6150390295951031,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004748364575546795,
      "loss": 5.2169,
      "mean_token_accuracy": 0.1968003213405609,
      "num_tokens": 32081359.0,
      "step": 14380
    },
    {
      "entropy": 5.389449834823608,
      "epoch": 1.6156006065030606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047481800481583806,
      "loss": 5.1819,
      "mean_token_accuracy": 0.20150613486766816,
      "num_tokens": 32093050.0,
      "step": 14385
    },
    {
      "entropy": 5.4959498882293705,
      "epoch": 1.616162183411018,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004747995457147139,
      "loss": 5.3132,
      "mean_token_accuracy": 0.19254093915224074,
      "num_tokens": 32104623.0,
      "step": 14390
    },
    {
      "entropy": 5.464283275604248,
      "epoch": 1.6167237603189757,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047478108025189467,
      "loss": 5.283,
      "mean_token_accuracy": 0.18424515426158905,
      "num_tokens": 32115845.0,
      "step": 14395
    },
    {
      "entropy": 5.477752304077148,
      "epoch": 1.6172853372269334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047476260842796837,
      "loss": 5.2261,
      "mean_token_accuracy": 0.19243161231279374,
      "num_tokens": 32127066.0,
      "step": 14400
    },
    {
      "entropy": 5.46799669265747,
      "epoch": 1.6178469141348908,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004747441302435231,
      "loss": 5.3624,
      "mean_token_accuracy": 0.1913568213582039,
      "num_tokens": 32138344.0,
      "step": 14405
    },
    {
      "entropy": 5.5497472286224365,
      "epoch": 1.6184084910428482,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047472564569914726,
      "loss": 5.3297,
      "mean_token_accuracy": 0.18632324635982514,
      "num_tokens": 32149054.0,
      "step": 14410
    },
    {
      "entropy": 5.494712781906128,
      "epoch": 1.618970067950806,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047470715479542945,
      "loss": 5.2114,
      "mean_token_accuracy": 0.18590016961097716,
      "num_tokens": 32160385.0,
      "step": 14415
    },
    {
      "entropy": 5.524468421936035,
      "epoch": 1.6195316448587636,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004746886575329584,
      "loss": 5.3714,
      "mean_token_accuracy": 0.18828795105218887,
      "num_tokens": 32171856.0,
      "step": 14420
    },
    {
      "entropy": 5.449940872192383,
      "epoch": 1.620093221766721,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047467015391232305,
      "loss": 5.2858,
      "mean_token_accuracy": 0.19384409189224244,
      "num_tokens": 32182690.0,
      "step": 14425
    },
    {
      "entropy": 5.452857494354248,
      "epoch": 1.6206547986746784,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047465164393411245,
      "loss": 5.2593,
      "mean_token_accuracy": 0.1913243591785431,
      "num_tokens": 32194045.0,
      "step": 14430
    },
    {
      "entropy": 5.511925172805786,
      "epoch": 1.621216375582636,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047463312759891615,
      "loss": 5.3286,
      "mean_token_accuracy": 0.18562723100185394,
      "num_tokens": 32205699.0,
      "step": 14435
    },
    {
      "entropy": 5.581693935394287,
      "epoch": 1.6217779524905935,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004746146049073236,
      "loss": 5.3264,
      "mean_token_accuracy": 0.18597105294466018,
      "num_tokens": 32217974.0,
      "step": 14440
    },
    {
      "entropy": 5.365362930297851,
      "epoch": 1.622339529398551,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047459607585992466,
      "loss": 5.213,
      "mean_token_accuracy": 0.19434062540531158,
      "num_tokens": 32228723.0,
      "step": 14445
    },
    {
      "entropy": 5.451690435409546,
      "epoch": 1.6229011063065086,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047457754045730923,
      "loss": 5.3759,
      "mean_token_accuracy": 0.1877187356352806,
      "num_tokens": 32241048.0,
      "step": 14450
    },
    {
      "entropy": 5.4257649898529055,
      "epoch": 1.6234626832144663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047455899870006745,
      "loss": 5.2322,
      "mean_token_accuracy": 0.19688658267259598,
      "num_tokens": 32251255.0,
      "step": 14455
    },
    {
      "entropy": 5.471079158782959,
      "epoch": 1.6240242601224237,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004745404505887899,
      "loss": 5.276,
      "mean_token_accuracy": 0.18840143382549285,
      "num_tokens": 32262954.0,
      "step": 14460
    },
    {
      "entropy": 5.49017071723938,
      "epoch": 1.6245858370303812,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004745218961240669,
      "loss": 5.2787,
      "mean_token_accuracy": 0.1973302334547043,
      "num_tokens": 32275445.0,
      "step": 14465
    },
    {
      "entropy": 5.48036036491394,
      "epoch": 1.6251474139383388,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004745033353064893,
      "loss": 5.2261,
      "mean_token_accuracy": 0.20021368861198424,
      "num_tokens": 32285976.0,
      "step": 14470
    },
    {
      "entropy": 5.419240140914917,
      "epoch": 1.6257089908462965,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004744847681366482,
      "loss": 5.2253,
      "mean_token_accuracy": 0.19624136984348298,
      "num_tokens": 32296304.0,
      "step": 14475
    },
    {
      "entropy": 5.466180610656738,
      "epoch": 1.626270567754254,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047446619461513466,
      "loss": 5.2573,
      "mean_token_accuracy": 0.202895824611187,
      "num_tokens": 32307085.0,
      "step": 14480
    },
    {
      "entropy": 5.437974071502685,
      "epoch": 1.6268321446622114,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004744476147425402,
      "loss": 5.1813,
      "mean_token_accuracy": 0.2056938111782074,
      "num_tokens": 32316814.0,
      "step": 14485
    },
    {
      "entropy": 5.413638639450073,
      "epoch": 1.627393721570169,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047442902851945634,
      "loss": 5.2688,
      "mean_token_accuracy": 0.19101787954568863,
      "num_tokens": 32327820.0,
      "step": 14490
    },
    {
      "entropy": 5.552432203292847,
      "epoch": 1.6279552984781267,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047441043594647486,
      "loss": 5.2797,
      "mean_token_accuracy": 0.19531487971544265,
      "num_tokens": 32338185.0,
      "step": 14495
    },
    {
      "entropy": 5.474474716186523,
      "epoch": 1.6285168753860841,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047439183702418765,
      "loss": 5.1826,
      "mean_token_accuracy": 0.19765564799308777,
      "num_tokens": 32348537.0,
      "step": 14500
    },
    {
      "entropy": 5.459687805175781,
      "epoch": 1.6290784522940416,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047437323175318716,
      "loss": 5.312,
      "mean_token_accuracy": 0.19006200581789018,
      "num_tokens": 32359094.0,
      "step": 14505
    },
    {
      "entropy": 5.324436140060425,
      "epoch": 1.6296400292019992,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047435462013406554,
      "loss": 5.1447,
      "mean_token_accuracy": 0.20569638013839722,
      "num_tokens": 32370811.0,
      "step": 14510
    },
    {
      "entropy": 5.520537900924682,
      "epoch": 1.630201606109957,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004743360021674155,
      "loss": 5.343,
      "mean_token_accuracy": 0.19003819078207015,
      "num_tokens": 32381522.0,
      "step": 14515
    },
    {
      "entropy": 5.472776699066162,
      "epoch": 1.6307631830179143,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047431737785382996,
      "loss": 5.305,
      "mean_token_accuracy": 0.19053868055343628,
      "num_tokens": 32393221.0,
      "step": 14520
    },
    {
      "entropy": 5.555664110183716,
      "epoch": 1.6313247599258718,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047429874719390165,
      "loss": 5.2553,
      "mean_token_accuracy": 0.186443330347538,
      "num_tokens": 32404007.0,
      "step": 14525
    },
    {
      "entropy": 5.52168836593628,
      "epoch": 1.6318863368338294,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047428011018822404,
      "loss": 5.3659,
      "mean_token_accuracy": 0.18848371654748916,
      "num_tokens": 32414318.0,
      "step": 14530
    },
    {
      "entropy": 5.453628826141357,
      "epoch": 1.6324479137417869,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047426146683739043,
      "loss": 5.2959,
      "mean_token_accuracy": 0.1955850303173065,
      "num_tokens": 32425493.0,
      "step": 14535
    },
    {
      "entropy": 5.4751489639282225,
      "epoch": 1.6330094906497443,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047424281714199436,
      "loss": 5.2545,
      "mean_token_accuracy": 0.19708335846662522,
      "num_tokens": 32435765.0,
      "step": 14540
    },
    {
      "entropy": 5.570137643814087,
      "epoch": 1.633571067557702,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047422416110262975,
      "loss": 5.322,
      "mean_token_accuracy": 0.19859974682331086,
      "num_tokens": 32446478.0,
      "step": 14545
    },
    {
      "entropy": 5.433111381530762,
      "epoch": 1.6341326444656596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047420549871989053,
      "loss": 5.3233,
      "mean_token_accuracy": 0.19823819696903228,
      "num_tokens": 32456866.0,
      "step": 14550
    },
    {
      "entropy": 5.507346677780151,
      "epoch": 1.634694221373617,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000474186829994371,
      "loss": 5.2098,
      "mean_token_accuracy": 0.20223504304885864,
      "num_tokens": 32466973.0,
      "step": 14555
    },
    {
      "entropy": 5.591839790344238,
      "epoch": 1.6352557982815745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047416815492666564,
      "loss": 5.3921,
      "mean_token_accuracy": 0.18848806172609328,
      "num_tokens": 32478118.0,
      "step": 14560
    },
    {
      "entropy": 5.4185503959655765,
      "epoch": 1.6358173751895322,
      "grad_norm": 1.125,
      "learning_rate": 0.0004741494735173689,
      "loss": 5.191,
      "mean_token_accuracy": 0.2056002840399742,
      "num_tokens": 32488255.0,
      "step": 14565
    },
    {
      "entropy": 5.507327938079834,
      "epoch": 1.6363789520974898,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004741307857670757,
      "loss": 5.2908,
      "mean_token_accuracy": 0.19692969173192978,
      "num_tokens": 32498685.0,
      "step": 14570
    },
    {
      "entropy": 5.497486019134522,
      "epoch": 1.6369405290054473,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047411209167638106,
      "loss": 5.3004,
      "mean_token_accuracy": 0.19256396889686583,
      "num_tokens": 32510457.0,
      "step": 14575
    },
    {
      "entropy": 5.448562812805176,
      "epoch": 1.6375021059134047,
      "grad_norm": 1.125,
      "learning_rate": 0.0004740933912458801,
      "loss": 5.2572,
      "mean_token_accuracy": 0.19564854800701142,
      "num_tokens": 32522495.0,
      "step": 14580
    },
    {
      "entropy": 5.607509517669678,
      "epoch": 1.6380636828213624,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047407468447616845,
      "loss": 5.3621,
      "mean_token_accuracy": 0.19127316027879715,
      "num_tokens": 32534891.0,
      "step": 14585
    },
    {
      "entropy": 5.461187124252319,
      "epoch": 1.63862525972932,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047405597136784166,
      "loss": 5.2229,
      "mean_token_accuracy": 0.19717327356338502,
      "num_tokens": 32545555.0,
      "step": 14590
    },
    {
      "entropy": 5.40444712638855,
      "epoch": 1.6391868366372775,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004740372519214955,
      "loss": 5.2846,
      "mean_token_accuracy": 0.19460931420326233,
      "num_tokens": 32556851.0,
      "step": 14595
    },
    {
      "entropy": 5.604703187942505,
      "epoch": 1.639748413545235,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047401852613772604,
      "loss": 5.4212,
      "mean_token_accuracy": 0.1845278963446617,
      "num_tokens": 32569028.0,
      "step": 14600
    },
    {
      "entropy": 5.670565891265869,
      "epoch": 1.6403099904531926,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004739997940171295,
      "loss": 5.3602,
      "mean_token_accuracy": 0.19195715039968492,
      "num_tokens": 32581510.0,
      "step": 14605
    },
    {
      "entropy": 5.455938816070557,
      "epoch": 1.6408715673611503,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047398105556030234,
      "loss": 5.2115,
      "mean_token_accuracy": 0.1932632029056549,
      "num_tokens": 32592914.0,
      "step": 14610
    },
    {
      "entropy": 5.373195362091065,
      "epoch": 1.6414331442691077,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047396231076784123,
      "loss": 5.1598,
      "mean_token_accuracy": 0.20752947479486467,
      "num_tokens": 32603931.0,
      "step": 14615
    },
    {
      "entropy": 5.543744945526123,
      "epoch": 1.6419947211770651,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004739435596403429,
      "loss": 5.3104,
      "mean_token_accuracy": 0.1974663883447647,
      "num_tokens": 32614508.0,
      "step": 14620
    },
    {
      "entropy": 5.410279273986816,
      "epoch": 1.6425562980850228,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004739248021784046,
      "loss": 5.164,
      "mean_token_accuracy": 0.20792541056871414,
      "num_tokens": 32625393.0,
      "step": 14625
    },
    {
      "entropy": 5.3330240726470945,
      "epoch": 1.6431178749929805,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004739060383826234,
      "loss": 5.1697,
      "mean_token_accuracy": 0.20206725895404815,
      "num_tokens": 32636645.0,
      "step": 14630
    },
    {
      "entropy": 5.584393167495728,
      "epoch": 1.6436794519009377,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047388726825359673,
      "loss": 5.4626,
      "mean_token_accuracy": 0.18707671165466308,
      "num_tokens": 32648792.0,
      "step": 14635
    },
    {
      "entropy": 5.577219247817993,
      "epoch": 1.6442410288088953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047386849179192235,
      "loss": 5.26,
      "mean_token_accuracy": 0.1996319115161896,
      "num_tokens": 32658825.0,
      "step": 14640
    },
    {
      "entropy": 5.532627058029175,
      "epoch": 1.644802605716853,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047384970899819813,
      "loss": 5.2861,
      "mean_token_accuracy": 0.19765844494104384,
      "num_tokens": 32670141.0,
      "step": 14645
    },
    {
      "entropy": 5.28882155418396,
      "epoch": 1.6453641826248104,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004738309198730219,
      "loss": 5.1366,
      "mean_token_accuracy": 0.20056047588586806,
      "num_tokens": 32681804.0,
      "step": 14650
    },
    {
      "entropy": 5.587597799301148,
      "epoch": 1.6459257595327679,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047381212441699217,
      "loss": 5.3872,
      "mean_token_accuracy": 0.1927493318915367,
      "num_tokens": 32692550.0,
      "step": 14655
    },
    {
      "entropy": 5.517039251327515,
      "epoch": 1.6464873364407255,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004737933226307072,
      "loss": 5.26,
      "mean_token_accuracy": 0.19721782952547073,
      "num_tokens": 32703833.0,
      "step": 14660
    },
    {
      "entropy": 5.41059627532959,
      "epoch": 1.6470489133486832,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047377451451476586,
      "loss": 5.3046,
      "mean_token_accuracy": 0.1915615662932396,
      "num_tokens": 32713928.0,
      "step": 14665
    },
    {
      "entropy": 5.507371091842652,
      "epoch": 1.6476104902566406,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004737557000697668,
      "loss": 5.313,
      "mean_token_accuracy": 0.19521367847919463,
      "num_tokens": 32725393.0,
      "step": 14670
    },
    {
      "entropy": 5.570739507675171,
      "epoch": 1.648172067164598,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004737368792963092,
      "loss": 5.2991,
      "mean_token_accuracy": 0.19338775128126146,
      "num_tokens": 32736047.0,
      "step": 14675
    },
    {
      "entropy": 5.498094844818115,
      "epoch": 1.6487336440725557,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047371805219499223,
      "loss": 5.3334,
      "mean_token_accuracy": 0.19092479348182678,
      "num_tokens": 32747960.0,
      "step": 14680
    },
    {
      "entropy": 5.436005353927612,
      "epoch": 1.6492952209805134,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047369921876641547,
      "loss": 5.2564,
      "mean_token_accuracy": 0.1973534882068634,
      "num_tokens": 32757851.0,
      "step": 14685
    },
    {
      "entropy": 5.494208478927613,
      "epoch": 1.6498567978884708,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047368037901117854,
      "loss": 5.2718,
      "mean_token_accuracy": 0.19450787007808684,
      "num_tokens": 32767317.0,
      "step": 14690
    },
    {
      "entropy": 5.565016841888427,
      "epoch": 1.6504183747964283,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004736615329298812,
      "loss": 5.2928,
      "mean_token_accuracy": 0.19438280314207076,
      "num_tokens": 32777729.0,
      "step": 14695
    },
    {
      "entropy": 5.480408334732056,
      "epoch": 1.650979951704386,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047364268052312367,
      "loss": 5.35,
      "mean_token_accuracy": 0.1826064944267273,
      "num_tokens": 32789547.0,
      "step": 14700
    },
    {
      "entropy": 5.497604322433472,
      "epoch": 1.6515415286123436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004736238217915061,
      "loss": 5.2449,
      "mean_token_accuracy": 0.19785625338554383,
      "num_tokens": 32801487.0,
      "step": 14705
    },
    {
      "entropy": 5.544666910171509,
      "epoch": 1.652103105520301,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047360495673562906,
      "loss": 5.315,
      "mean_token_accuracy": 0.19704079180955886,
      "num_tokens": 32813042.0,
      "step": 14710
    },
    {
      "entropy": 5.505541372299194,
      "epoch": 1.6526646824282585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004735860853560931,
      "loss": 5.3015,
      "mean_token_accuracy": 0.19589462578296662,
      "num_tokens": 32825467.0,
      "step": 14715
    },
    {
      "entropy": 5.481626319885254,
      "epoch": 1.6532262593362161,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047356720765349926,
      "loss": 5.2688,
      "mean_token_accuracy": 0.19094224274158478,
      "num_tokens": 32837113.0,
      "step": 14720
    },
    {
      "entropy": 5.55432825088501,
      "epoch": 1.6537878362441738,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004735483236284485,
      "loss": 5.3703,
      "mean_token_accuracy": 0.18702881634235383,
      "num_tokens": 32849779.0,
      "step": 14725
    },
    {
      "entropy": 5.445461559295654,
      "epoch": 1.654349413152131,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047352943328154203,
      "loss": 5.2744,
      "mean_token_accuracy": 0.19307452589273452,
      "num_tokens": 32861626.0,
      "step": 14730
    },
    {
      "entropy": 5.520581960678101,
      "epoch": 1.6549109900600887,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047351053661338145,
      "loss": 5.324,
      "mean_token_accuracy": 0.1982700139284134,
      "num_tokens": 32872875.0,
      "step": 14735
    },
    {
      "entropy": 5.46758804321289,
      "epoch": 1.6554725669680463,
      "grad_norm": 1.125,
      "learning_rate": 0.0004734916336245684,
      "loss": 5.1491,
      "mean_token_accuracy": 0.19871128648519515,
      "num_tokens": 32883899.0,
      "step": 14740
    },
    {
      "entropy": 5.366373872756958,
      "epoch": 1.6560341438760038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004734727243157047,
      "loss": 5.156,
      "mean_token_accuracy": 0.202083320915699,
      "num_tokens": 32894255.0,
      "step": 14745
    },
    {
      "entropy": 5.431836938858032,
      "epoch": 1.6565957207839612,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047345380868739253,
      "loss": 5.2549,
      "mean_token_accuracy": 0.1974031314253807,
      "num_tokens": 32903822.0,
      "step": 14750
    },
    {
      "entropy": 5.462358474731445,
      "epoch": 1.6571572976919189,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004734348867402341,
      "loss": 5.2419,
      "mean_token_accuracy": 0.19894727021455766,
      "num_tokens": 32915879.0,
      "step": 14755
    },
    {
      "entropy": 5.36098313331604,
      "epoch": 1.6577188745998765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004734159584748319,
      "loss": 5.272,
      "mean_token_accuracy": 0.19022632986307145,
      "num_tokens": 32928161.0,
      "step": 14760
    },
    {
      "entropy": 5.436191606521606,
      "epoch": 1.658280451507834,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047339702389178874,
      "loss": 5.2258,
      "mean_token_accuracy": 0.19999379813671112,
      "num_tokens": 32938399.0,
      "step": 14765
    },
    {
      "entropy": 5.447626829147339,
      "epoch": 1.6588420284157914,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047337808299170725,
      "loss": 5.1865,
      "mean_token_accuracy": 0.20350140631198882,
      "num_tokens": 32948221.0,
      "step": 14770
    },
    {
      "entropy": 5.4412473201751705,
      "epoch": 1.659403605323749,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047335913577519075,
      "loss": 5.2567,
      "mean_token_accuracy": 0.1995250850915909,
      "num_tokens": 32959802.0,
      "step": 14775
    },
    {
      "entropy": 5.488705015182495,
      "epoch": 1.6599651822317067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047334018224284244,
      "loss": 5.2878,
      "mean_token_accuracy": 0.19510946422815323,
      "num_tokens": 32970222.0,
      "step": 14780
    },
    {
      "entropy": 5.461812829971313,
      "epoch": 1.6605267591396642,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004733212223952657,
      "loss": 5.2491,
      "mean_token_accuracy": 0.20073420256376268,
      "num_tokens": 32982241.0,
      "step": 14785
    },
    {
      "entropy": 5.466669034957886,
      "epoch": 1.6610883360476216,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047330225623306443,
      "loss": 5.1951,
      "mean_token_accuracy": 0.20182580798864364,
      "num_tokens": 32992584.0,
      "step": 14790
    },
    {
      "entropy": 5.477460670471191,
      "epoch": 1.6616499129555793,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047328328375684237,
      "loss": 5.4,
      "mean_token_accuracy": 0.1908192053437233,
      "num_tokens": 33004080.0,
      "step": 14795
    },
    {
      "entropy": 5.446459341049194,
      "epoch": 1.662211489863537,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004732643049672036,
      "loss": 5.2129,
      "mean_token_accuracy": 0.197823166847229,
      "num_tokens": 33015290.0,
      "step": 14800
    },
    {
      "entropy": 5.467676401138306,
      "epoch": 1.6627730667714944,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004732453198647525,
      "loss": 5.1932,
      "mean_token_accuracy": 0.19475805908441543,
      "num_tokens": 33026470.0,
      "step": 14805
    },
    {
      "entropy": 5.406273221969604,
      "epoch": 1.6633346436794518,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004732263284500936,
      "loss": 5.1874,
      "mean_token_accuracy": 0.20261722505092622,
      "num_tokens": 33036766.0,
      "step": 14810
    },
    {
      "entropy": 5.3666017055511475,
      "epoch": 1.6638962205874095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004732073307238314,
      "loss": 5.2028,
      "mean_token_accuracy": 0.1886460840702057,
      "num_tokens": 33047675.0,
      "step": 14815
    },
    {
      "entropy": 5.49187159538269,
      "epoch": 1.6644577974953672,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000473188326686571,
      "loss": 5.2492,
      "mean_token_accuracy": 0.19694626778364183,
      "num_tokens": 33057613.0,
      "step": 14820
    },
    {
      "entropy": 5.5723340034484865,
      "epoch": 1.6650193744033244,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047316931633891736,
      "loss": 5.3639,
      "mean_token_accuracy": 0.18635020703077315,
      "num_tokens": 33070498.0,
      "step": 14825
    },
    {
      "entropy": 5.456169319152832,
      "epoch": 1.665580951311282,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004731502996814759,
      "loss": 5.2259,
      "mean_token_accuracy": 0.1980530396103859,
      "num_tokens": 33082026.0,
      "step": 14830
    },
    {
      "entropy": 5.432246971130371,
      "epoch": 1.6661425282192397,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047313127671485194,
      "loss": 5.2358,
      "mean_token_accuracy": 0.19229164123535156,
      "num_tokens": 33093841.0,
      "step": 14835
    },
    {
      "entropy": 5.505250883102417,
      "epoch": 1.6667041051271971,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004731122474396514,
      "loss": 5.2696,
      "mean_token_accuracy": 0.19567066580057144,
      "num_tokens": 33103817.0,
      "step": 14840
    },
    {
      "entropy": 5.55364875793457,
      "epoch": 1.6672656820351546,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047309321185648,
      "loss": 5.3255,
      "mean_token_accuracy": 0.19429995566606523,
      "num_tokens": 33116006.0,
      "step": 14845
    },
    {
      "entropy": 5.445116281509399,
      "epoch": 1.6678272589431122,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047307416996594386,
      "loss": 5.318,
      "mean_token_accuracy": 0.18888264745473862,
      "num_tokens": 33127649.0,
      "step": 14850
    },
    {
      "entropy": 5.393762254714966,
      "epoch": 1.66838883585107,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004730551217686495,
      "loss": 5.1534,
      "mean_token_accuracy": 0.202876478433609,
      "num_tokens": 33138748.0,
      "step": 14855
    },
    {
      "entropy": 5.522401428222656,
      "epoch": 1.6689504127590273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004730360672652031,
      "loss": 5.3216,
      "mean_token_accuracy": 0.1927672177553177,
      "num_tokens": 33150504.0,
      "step": 14860
    },
    {
      "entropy": 5.452508163452149,
      "epoch": 1.6695119896669848,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047301700645621154,
      "loss": 5.3353,
      "mean_token_accuracy": 0.18897348940372466,
      "num_tokens": 33160625.0,
      "step": 14865
    },
    {
      "entropy": 5.478543996810913,
      "epoch": 1.6700735665749424,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004729979393422817,
      "loss": 5.2344,
      "mean_token_accuracy": 0.19954254180192948,
      "num_tokens": 33172229.0,
      "step": 14870
    },
    {
      "entropy": 5.496869611740112,
      "epoch": 1.6706351434829,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004729788659240206,
      "loss": 5.272,
      "mean_token_accuracy": 0.19200197011232376,
      "num_tokens": 33183234.0,
      "step": 14875
    },
    {
      "entropy": 5.501498699188232,
      "epoch": 1.6711967203908575,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004729597862020358,
      "loss": 5.312,
      "mean_token_accuracy": 0.19559772163629532,
      "num_tokens": 33193976.0,
      "step": 14880
    },
    {
      "entropy": 5.437897109985352,
      "epoch": 1.671758297298815,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047294070017693453,
      "loss": 5.272,
      "mean_token_accuracy": 0.20058979392051696,
      "num_tokens": 33205959.0,
      "step": 14885
    },
    {
      "entropy": 5.434696531295776,
      "epoch": 1.6723198742067726,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047292160784932454,
      "loss": 5.1948,
      "mean_token_accuracy": 0.20296574831008912,
      "num_tokens": 33218405.0,
      "step": 14890
    },
    {
      "entropy": 5.426598930358887,
      "epoch": 1.6728814511147303,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047290250921981386,
      "loss": 5.249,
      "mean_token_accuracy": 0.194725601375103,
      "num_tokens": 33230778.0,
      "step": 14895
    },
    {
      "entropy": 5.415722751617432,
      "epoch": 1.6734430280226877,
      "grad_norm": 1.25,
      "learning_rate": 0.00047288340428901053,
      "loss": 5.2082,
      "mean_token_accuracy": 0.19918926805257797,
      "num_tokens": 33242069.0,
      "step": 14900
    },
    {
      "entropy": 5.482007837295532,
      "epoch": 1.6740046049306452,
      "grad_norm": 1.125,
      "learning_rate": 0.00047286429305752283,
      "loss": 5.2826,
      "mean_token_accuracy": 0.19658081084489823,
      "num_tokens": 33252942.0,
      "step": 14905
    },
    {
      "entropy": 5.578243350982666,
      "epoch": 1.6745661818386028,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047284517552595933,
      "loss": 5.3255,
      "mean_token_accuracy": 0.1905709609389305,
      "num_tokens": 33263892.0,
      "step": 14910
    },
    {
      "entropy": 5.404222297668457,
      "epoch": 1.6751277587465605,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004728260516949287,
      "loss": 5.2043,
      "mean_token_accuracy": 0.1931039199233055,
      "num_tokens": 33274383.0,
      "step": 14915
    },
    {
      "entropy": 5.502907037734985,
      "epoch": 1.6756893356545177,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004728069215650399,
      "loss": 5.2198,
      "mean_token_accuracy": 0.19201160222291946,
      "num_tokens": 33285897.0,
      "step": 14920
    },
    {
      "entropy": 5.535489702224732,
      "epoch": 1.6762509125624754,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004727877851369019,
      "loss": 5.3199,
      "mean_token_accuracy": 0.19218389838933944,
      "num_tokens": 33297192.0,
      "step": 14925
    },
    {
      "entropy": 5.4477410316467285,
      "epoch": 1.676812489470433,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047276864241112417,
      "loss": 5.2113,
      "mean_token_accuracy": 0.19755875170230866,
      "num_tokens": 33308250.0,
      "step": 14930
    },
    {
      "entropy": 5.432844352722168,
      "epoch": 1.6773740663783905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004727494933883161,
      "loss": 5.2215,
      "mean_token_accuracy": 0.1962057575583458,
      "num_tokens": 33318515.0,
      "step": 14935
    },
    {
      "entropy": 5.542074251174927,
      "epoch": 1.677935643286348,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004727303380690876,
      "loss": 5.359,
      "mean_token_accuracy": 0.19027905613183976,
      "num_tokens": 33329762.0,
      "step": 14940
    },
    {
      "entropy": 5.5367666721344,
      "epoch": 1.6784972201943056,
      "grad_norm": 1.125,
      "learning_rate": 0.0004727111764540484,
      "loss": 5.3492,
      "mean_token_accuracy": 0.1910214051604271,
      "num_tokens": 33341054.0,
      "step": 14945
    },
    {
      "entropy": 5.457042837142945,
      "epoch": 1.6790587971022632,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047269200854380854,
      "loss": 5.2013,
      "mean_token_accuracy": 0.19975604116916656,
      "num_tokens": 33351524.0,
      "step": 14950
    },
    {
      "entropy": 5.460369443893432,
      "epoch": 1.6796203740102207,
      "grad_norm": 1.125,
      "learning_rate": 0.00047267283433897854,
      "loss": 5.301,
      "mean_token_accuracy": 0.18565541356801987,
      "num_tokens": 33362953.0,
      "step": 14955
    },
    {
      "entropy": 5.369621610641479,
      "epoch": 1.6801819509181781,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047265365384016887,
      "loss": 5.1699,
      "mean_token_accuracy": 0.204598967730999,
      "num_tokens": 33373209.0,
      "step": 14960
    },
    {
      "entropy": 5.377118253707886,
      "epoch": 1.6807435278261358,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047263446704799017,
      "loss": 5.1781,
      "mean_token_accuracy": 0.19663191884756087,
      "num_tokens": 33384775.0,
      "step": 14965
    },
    {
      "entropy": 5.392876672744751,
      "epoch": 1.6813051047340934,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047261527396305343,
      "loss": 5.23,
      "mean_token_accuracy": 0.19281969368457794,
      "num_tokens": 33395457.0,
      "step": 14970
    },
    {
      "entropy": 5.58880844116211,
      "epoch": 1.6818666816420509,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047259607458596966,
      "loss": 5.3618,
      "mean_token_accuracy": 0.1950633004307747,
      "num_tokens": 33407506.0,
      "step": 14975
    },
    {
      "entropy": 5.549135446548462,
      "epoch": 1.6824282585500083,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047257686891735023,
      "loss": 5.2532,
      "mean_token_accuracy": 0.19640884548425674,
      "num_tokens": 33418154.0,
      "step": 14980
    },
    {
      "entropy": 5.431192016601562,
      "epoch": 1.682989835457966,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047255765695780665,
      "loss": 5.2759,
      "mean_token_accuracy": 0.19720015823841094,
      "num_tokens": 33428319.0,
      "step": 14985
    },
    {
      "entropy": 5.382882356643677,
      "epoch": 1.6835514123659236,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004725384387079507,
      "loss": 5.183,
      "mean_token_accuracy": 0.20026712864637375,
      "num_tokens": 33440429.0,
      "step": 14990
    },
    {
      "entropy": 5.638523530960083,
      "epoch": 1.684112989273881,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047251921416839425,
      "loss": 5.4046,
      "mean_token_accuracy": 0.18178529441356658,
      "num_tokens": 33451736.0,
      "step": 14995
    },
    {
      "entropy": 5.484934949874878,
      "epoch": 1.6846745661818385,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004724999833397494,
      "loss": 5.1837,
      "mean_token_accuracy": 0.19435687810182573,
      "num_tokens": 33461991.0,
      "step": 15000
    },
    {
      "epoch": 1.6846745661818385,
      "eval_entropy": 5.31126108629543,
      "eval_loss": 5.359016418457031,
      "eval_mean_token_accuracy": 0.1983783744200622,
      "eval_num_tokens": 33461991.0,
      "eval_runtime": 60.7711,
      "eval_samples_per_second": 510.325,
      "eval_steps_per_second": 63.797,
      "step": 15000
    },
    {
      "entropy": 5.467107629776001,
      "epoch": 1.6852361430897962,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004724807462226285,
      "loss": 5.3171,
      "mean_token_accuracy": 0.18784952014684678,
      "num_tokens": 33473646.0,
      "step": 15005
    },
    {
      "entropy": 5.4660145282745365,
      "epoch": 1.6857977199977539,
      "grad_norm": 1.109375,
      "learning_rate": 0.000472461502817644,
      "loss": 5.2711,
      "mean_token_accuracy": 0.19591651260852813,
      "num_tokens": 33485444.0,
      "step": 15010
    },
    {
      "entropy": 5.607774829864502,
      "epoch": 1.6863592969057113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004724422531254087,
      "loss": 5.4451,
      "mean_token_accuracy": 0.1862923875451088,
      "num_tokens": 33497199.0,
      "step": 15015
    },
    {
      "entropy": 5.530071163177491,
      "epoch": 1.6869208738136687,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004724229971465355,
      "loss": 5.3623,
      "mean_token_accuracy": 0.18910333067178725,
      "num_tokens": 33509045.0,
      "step": 15020
    },
    {
      "entropy": 5.4576112747192385,
      "epoch": 1.6874824507216264,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004724037348816375,
      "loss": 5.1493,
      "mean_token_accuracy": 0.20573700964450836,
      "num_tokens": 33519730.0,
      "step": 15025
    },
    {
      "entropy": 5.459954690933228,
      "epoch": 1.6880440276295838,
      "grad_norm": 1.28125,
      "learning_rate": 0.000472384466331328,
      "loss": 5.3464,
      "mean_token_accuracy": 0.18774333000183105,
      "num_tokens": 33530393.0,
      "step": 15030
    },
    {
      "entropy": 5.452465200424195,
      "epoch": 1.6886056045375413,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004723651914962205,
      "loss": 5.3035,
      "mean_token_accuracy": 0.19598446488380433,
      "num_tokens": 33543592.0,
      "step": 15035
    },
    {
      "entropy": 5.499931716918946,
      "epoch": 1.689167181445499,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047234591037692884,
      "loss": 5.2453,
      "mean_token_accuracy": 0.19345757067203523,
      "num_tokens": 33554736.0,
      "step": 15040
    },
    {
      "entropy": 5.4852948665618895,
      "epoch": 1.6897287583534566,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047232662297406683,
      "loss": 5.1697,
      "mean_token_accuracy": 0.20276740491390227,
      "num_tokens": 33565428.0,
      "step": 15045
    },
    {
      "entropy": 5.535889482498169,
      "epoch": 1.690290335261414,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004723073292882485,
      "loss": 5.4902,
      "mean_token_accuracy": 0.18611413836479188,
      "num_tokens": 33577102.0,
      "step": 15050
    },
    {
      "entropy": 5.5720357418060305,
      "epoch": 1.6908519121693715,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047228802932008836,
      "loss": 5.3081,
      "mean_token_accuracy": 0.19705154448747636,
      "num_tokens": 33587325.0,
      "step": 15055
    },
    {
      "entropy": 5.584289836883545,
      "epoch": 1.6914134890773291,
      "grad_norm": 1.125,
      "learning_rate": 0.0004722687230702009,
      "loss": 5.3471,
      "mean_token_accuracy": 0.18958540111780167,
      "num_tokens": 33599294.0,
      "step": 15060
    },
    {
      "entropy": 5.531365585327149,
      "epoch": 1.6919750659852868,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004722494105392007,
      "loss": 5.3953,
      "mean_token_accuracy": 0.18766674101352693,
      "num_tokens": 33609025.0,
      "step": 15065
    },
    {
      "entropy": 5.545634460449219,
      "epoch": 1.6925366428932442,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004722300917277029,
      "loss": 5.3226,
      "mean_token_accuracy": 0.1924744427204132,
      "num_tokens": 33620881.0,
      "step": 15070
    },
    {
      "entropy": 5.524413299560547,
      "epoch": 1.6930982198012017,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004722107666363224,
      "loss": 5.2707,
      "mean_token_accuracy": 0.19736731201410293,
      "num_tokens": 33631505.0,
      "step": 15075
    },
    {
      "entropy": 5.534477806091308,
      "epoch": 1.6936597967091593,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004721914352656746,
      "loss": 5.4297,
      "mean_token_accuracy": 0.18785444647073746,
      "num_tokens": 33642560.0,
      "step": 15080
    },
    {
      "entropy": 5.542061471939087,
      "epoch": 1.694221373617117,
      "grad_norm": 1.0625,
      "learning_rate": 0.000472172097616375,
      "loss": 5.2997,
      "mean_token_accuracy": 0.19611501097679138,
      "num_tokens": 33654706.0,
      "step": 15085
    },
    {
      "entropy": 5.549911642074585,
      "epoch": 1.6947829505250744,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004721527536890394,
      "loss": 5.3324,
      "mean_token_accuracy": 0.18690803200006484,
      "num_tokens": 33665392.0,
      "step": 15090
    },
    {
      "entropy": 5.366191291809082,
      "epoch": 1.6953445274330319,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004721334034842836,
      "loss": 5.1547,
      "mean_token_accuracy": 0.20463827401399612,
      "num_tokens": 33675504.0,
      "step": 15095
    },
    {
      "entropy": 5.481257772445678,
      "epoch": 1.6959061043409895,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004721140470027238,
      "loss": 5.3039,
      "mean_token_accuracy": 0.19443414509296417,
      "num_tokens": 33686505.0,
      "step": 15100
    },
    {
      "entropy": 5.546137285232544,
      "epoch": 1.6964676812489472,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047209468424497627,
      "loss": 5.3132,
      "mean_token_accuracy": 0.19328574240207672,
      "num_tokens": 33698752.0,
      "step": 15105
    },
    {
      "entropy": 5.582325458526611,
      "epoch": 1.6970292581569046,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004720753152116576,
      "loss": 5.3589,
      "mean_token_accuracy": 0.1860020011663437,
      "num_tokens": 33710323.0,
      "step": 15110
    },
    {
      "entropy": 5.550843000411987,
      "epoch": 1.697590835064862,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047205593990338443,
      "loss": 5.4268,
      "mean_token_accuracy": 0.17912996262311937,
      "num_tokens": 33721875.0,
      "step": 15115
    },
    {
      "entropy": 5.423304080963135,
      "epoch": 1.6981524119728197,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004720365583207737,
      "loss": 5.1258,
      "mean_token_accuracy": 0.20194465965032576,
      "num_tokens": 33731520.0,
      "step": 15120
    },
    {
      "entropy": 5.565987682342529,
      "epoch": 1.6987139888807772,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004720171704644425,
      "loss": 5.3699,
      "mean_token_accuracy": 0.19037993550300597,
      "num_tokens": 33743351.0,
      "step": 15125
    },
    {
      "entropy": 5.545731115341186,
      "epoch": 1.6992755657887346,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004719977763350083,
      "loss": 5.307,
      "mean_token_accuracy": 0.19767358154058456,
      "num_tokens": 33754879.0,
      "step": 15130
    },
    {
      "entropy": 5.4189637184143065,
      "epoch": 1.6998371426966923,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047197837593308833,
      "loss": 5.226,
      "mean_token_accuracy": 0.19588263034820558,
      "num_tokens": 33766671.0,
      "step": 15135
    },
    {
      "entropy": 5.50096173286438,
      "epoch": 1.70039871960465,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004719589692593006,
      "loss": 5.2811,
      "mean_token_accuracy": 0.1901957631111145,
      "num_tokens": 33778236.0,
      "step": 15140
    },
    {
      "entropy": 5.460691261291504,
      "epoch": 1.7009602965126074,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004719395563142629,
      "loss": 5.1802,
      "mean_token_accuracy": 0.19543107151985167,
      "num_tokens": 33788898.0,
      "step": 15145
    },
    {
      "entropy": 5.487517404556274,
      "epoch": 1.7015218734205648,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004719201370985933,
      "loss": 5.3439,
      "mean_token_accuracy": 0.1876659758388996,
      "num_tokens": 33800641.0,
      "step": 15150
    },
    {
      "entropy": 5.540626287460327,
      "epoch": 1.7020834503285225,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047190071161291007,
      "loss": 5.2475,
      "mean_token_accuracy": 0.19562939256429673,
      "num_tokens": 33812581.0,
      "step": 15155
    },
    {
      "entropy": 5.463412809371948,
      "epoch": 1.7026450272364801,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047188127985783195,
      "loss": 5.2045,
      "mean_token_accuracy": 0.20464757829904556,
      "num_tokens": 33823378.0,
      "step": 15160
    },
    {
      "entropy": 5.461296606063843,
      "epoch": 1.7032066041444376,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047186184183397754,
      "loss": 5.3165,
      "mean_token_accuracy": 0.1963017150759697,
      "num_tokens": 33834804.0,
      "step": 15165
    },
    {
      "entropy": 5.43390645980835,
      "epoch": 1.703768181052395,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047184239754196564,
      "loss": 5.2746,
      "mean_token_accuracy": 0.19094635248184205,
      "num_tokens": 33847044.0,
      "step": 15170
    },
    {
      "entropy": 5.508099842071533,
      "epoch": 1.7043297579603527,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004718229469824155,
      "loss": 5.2122,
      "mean_token_accuracy": 0.19231095761060715,
      "num_tokens": 33858077.0,
      "step": 15175
    },
    {
      "entropy": 5.490667390823364,
      "epoch": 1.7048913348683103,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047180349015594634,
      "loss": 5.2713,
      "mean_token_accuracy": 0.19268581122159958,
      "num_tokens": 33868765.0,
      "step": 15180
    },
    {
      "entropy": 5.427499103546142,
      "epoch": 1.7054529117762678,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004717840270631778,
      "loss": 5.2307,
      "mean_token_accuracy": 0.19691985994577407,
      "num_tokens": 33880415.0,
      "step": 15185
    },
    {
      "entropy": 5.442188405990601,
      "epoch": 1.7060144886842252,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047176455770472946,
      "loss": 5.2758,
      "mean_token_accuracy": 0.19425488114356995,
      "num_tokens": 33892081.0,
      "step": 15190
    },
    {
      "entropy": 5.5596880435943605,
      "epoch": 1.706576065592183,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047174508208122135,
      "loss": 5.3918,
      "mean_token_accuracy": 0.19346689730882644,
      "num_tokens": 33903433.0,
      "step": 15195
    },
    {
      "entropy": 5.5551369190216064,
      "epoch": 1.7071376425001406,
      "grad_norm": 1.0,
      "learning_rate": 0.0004717256001932735,
      "loss": 5.2714,
      "mean_token_accuracy": 0.19400326013565064,
      "num_tokens": 33914706.0,
      "step": 15200
    },
    {
      "entropy": 5.449495697021485,
      "epoch": 1.707699219408098,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047170611204150624,
      "loss": 5.3999,
      "mean_token_accuracy": 0.1895191490650177,
      "num_tokens": 33927195.0,
      "step": 15205
    },
    {
      "entropy": 5.500268507003784,
      "epoch": 1.7082607963160554,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004716866176265401,
      "loss": 5.2631,
      "mean_token_accuracy": 0.2011074662208557,
      "num_tokens": 33938422.0,
      "step": 15210
    },
    {
      "entropy": 5.5860333919525145,
      "epoch": 1.708822373224013,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047166711694899573,
      "loss": 5.307,
      "mean_token_accuracy": 0.1957274407148361,
      "num_tokens": 33949775.0,
      "step": 15215
    },
    {
      "entropy": 5.467221641540528,
      "epoch": 1.7093839501319705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047164761000949404,
      "loss": 5.2383,
      "mean_token_accuracy": 0.1993105500936508,
      "num_tokens": 33960371.0,
      "step": 15220
    },
    {
      "entropy": 5.385531806945801,
      "epoch": 1.709945527039928,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004716280968086563,
      "loss": 5.2032,
      "mean_token_accuracy": 0.19383062571287155,
      "num_tokens": 33971326.0,
      "step": 15225
    },
    {
      "entropy": 5.504047966003418,
      "epoch": 1.7105071039478856,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004716085773471036,
      "loss": 5.3208,
      "mean_token_accuracy": 0.18932489156723023,
      "num_tokens": 33982744.0,
      "step": 15230
    },
    {
      "entropy": 5.557673597335816,
      "epoch": 1.7110686808558433,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047158905162545756,
      "loss": 5.3597,
      "mean_token_accuracy": 0.1907482475042343,
      "num_tokens": 33993647.0,
      "step": 15235
    },
    {
      "entropy": 5.43685359954834,
      "epoch": 1.7116302577638007,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047156951964433984,
      "loss": 5.218,
      "mean_token_accuracy": 0.19216079264879227,
      "num_tokens": 34004026.0,
      "step": 15240
    },
    {
      "entropy": 5.456605243682861,
      "epoch": 1.7121918346717582,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047154998140437246,
      "loss": 5.2908,
      "mean_token_accuracy": 0.19980739653110505,
      "num_tokens": 34014882.0,
      "step": 15245
    },
    {
      "entropy": 5.462295484542847,
      "epoch": 1.7127534115797158,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047153043690617737,
      "loss": 5.3148,
      "mean_token_accuracy": 0.1958867073059082,
      "num_tokens": 34026548.0,
      "step": 15250
    },
    {
      "entropy": 5.478177642822265,
      "epoch": 1.7133149884876735,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000471510886150377,
      "loss": 5.1942,
      "mean_token_accuracy": 0.19848761707544327,
      "num_tokens": 34038393.0,
      "step": 15255
    },
    {
      "entropy": 5.488607406616211,
      "epoch": 1.713876565395631,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004714913291375938,
      "loss": 5.3194,
      "mean_token_accuracy": 0.18940568268299102,
      "num_tokens": 34050257.0,
      "step": 15260
    },
    {
      "entropy": 5.4760009765625,
      "epoch": 1.7144381423035884,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047147176586845036,
      "loss": 5.2623,
      "mean_token_accuracy": 0.19758107215166093,
      "num_tokens": 34061860.0,
      "step": 15265
    },
    {
      "entropy": 5.56711974143982,
      "epoch": 1.714999719211546,
      "grad_norm": 1.125,
      "learning_rate": 0.0004714521963435698,
      "loss": 5.3305,
      "mean_token_accuracy": 0.1937226578593254,
      "num_tokens": 34075247.0,
      "step": 15270
    },
    {
      "entropy": 5.47876615524292,
      "epoch": 1.7155612961195037,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047143262056357506,
      "loss": 5.1915,
      "mean_token_accuracy": 0.19317598789930343,
      "num_tokens": 34085801.0,
      "step": 15275
    },
    {
      "entropy": 5.505070304870605,
      "epoch": 1.7161228730274611,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047141303852908965,
      "loss": 5.2908,
      "mean_token_accuracy": 0.1918347880244255,
      "num_tokens": 34095686.0,
      "step": 15280
    },
    {
      "entropy": 5.481535577774048,
      "epoch": 1.7166844499354186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004713934502407367,
      "loss": 5.2306,
      "mean_token_accuracy": 0.20252699553966522,
      "num_tokens": 34106518.0,
      "step": 15285
    },
    {
      "entropy": 5.483408355712891,
      "epoch": 1.7172460268433762,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004713738556991402,
      "loss": 5.2838,
      "mean_token_accuracy": 0.20250847339630126,
      "num_tokens": 34117793.0,
      "step": 15290
    },
    {
      "entropy": 5.519010591506958,
      "epoch": 1.717807603751334,
      "grad_norm": 1.203125,
      "learning_rate": 0.000471354254904924,
      "loss": 5.3781,
      "mean_token_accuracy": 0.19711277782917022,
      "num_tokens": 34128748.0,
      "step": 15295
    },
    {
      "entropy": 5.456810760498047,
      "epoch": 1.7183691806592913,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047133464785871216,
      "loss": 5.2826,
      "mean_token_accuracy": 0.1997857466340065,
      "num_tokens": 34140597.0,
      "step": 15300
    },
    {
      "entropy": 5.529809427261353,
      "epoch": 1.7189307575672488,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000471315034561129,
      "loss": 5.2623,
      "mean_token_accuracy": 0.18573045134544372,
      "num_tokens": 34151823.0,
      "step": 15305
    },
    {
      "entropy": 5.455437755584716,
      "epoch": 1.7194923344752064,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047129541501279897,
      "loss": 5.2144,
      "mean_token_accuracy": 0.20137128233909607,
      "num_tokens": 34163697.0,
      "step": 15310
    },
    {
      "entropy": 5.5017376899719235,
      "epoch": 1.7200539113831639,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004712757892143468,
      "loss": 5.2105,
      "mean_token_accuracy": 0.2000478282570839,
      "num_tokens": 34175058.0,
      "step": 15315
    },
    {
      "entropy": 5.513131189346313,
      "epoch": 1.7206154882911213,
      "grad_norm": 1.125,
      "learning_rate": 0.00047125615716639744,
      "loss": 5.3766,
      "mean_token_accuracy": 0.19343230426311492,
      "num_tokens": 34186883.0,
      "step": 15320
    },
    {
      "entropy": 5.455159044265747,
      "epoch": 1.721177065199079,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004712365188695758,
      "loss": 5.2357,
      "mean_token_accuracy": 0.2001147523522377,
      "num_tokens": 34198103.0,
      "step": 15325
    },
    {
      "entropy": 5.435485792160034,
      "epoch": 1.7217386421070366,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004712168743245074,
      "loss": 5.2352,
      "mean_token_accuracy": 0.19808426797389983,
      "num_tokens": 34208747.0,
      "step": 15330
    },
    {
      "entropy": 5.4742655754089355,
      "epoch": 1.722300219014994,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047119722353181754,
      "loss": 5.344,
      "mean_token_accuracy": 0.19638944715261458,
      "num_tokens": 34220775.0,
      "step": 15335
    },
    {
      "entropy": 5.511271238327026,
      "epoch": 1.7228617959229515,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047117756649213204,
      "loss": 5.1973,
      "mean_token_accuracy": 0.20363293141126632,
      "num_tokens": 34231031.0,
      "step": 15340
    },
    {
      "entropy": 5.517343616485595,
      "epoch": 1.7234233728309092,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004711579032060767,
      "loss": 5.306,
      "mean_token_accuracy": 0.19267010390758516,
      "num_tokens": 34241850.0,
      "step": 15345
    },
    {
      "entropy": 5.462200450897217,
      "epoch": 1.7239849497388668,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004711382336742777,
      "loss": 5.2123,
      "mean_token_accuracy": 0.20199716836214066,
      "num_tokens": 34252732.0,
      "step": 15350
    },
    {
      "entropy": 5.44701771736145,
      "epoch": 1.7245465266468243,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004711185578973612,
      "loss": 5.2183,
      "mean_token_accuracy": 0.19148160070180892,
      "num_tokens": 34264066.0,
      "step": 15355
    },
    {
      "entropy": 5.579038953781128,
      "epoch": 1.7251081035547817,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004710988758759538,
      "loss": 5.3422,
      "mean_token_accuracy": 0.18472560942173005,
      "num_tokens": 34275239.0,
      "step": 15360
    },
    {
      "entropy": 5.4547121047973635,
      "epoch": 1.7256696804627394,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004710791876106821,
      "loss": 5.1846,
      "mean_token_accuracy": 0.19288887530565263,
      "num_tokens": 34285723.0,
      "step": 15365
    },
    {
      "entropy": 5.448443651199341,
      "epoch": 1.726231257370697,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047105949310217304,
      "loss": 5.2297,
      "mean_token_accuracy": 0.20415070056915283,
      "num_tokens": 34296292.0,
      "step": 15370
    },
    {
      "entropy": 5.4620811462402346,
      "epoch": 1.7267928342786545,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004710397923510537,
      "loss": 5.271,
      "mean_token_accuracy": 0.20204923003911973,
      "num_tokens": 34307232.0,
      "step": 15375
    },
    {
      "entropy": 5.456731128692627,
      "epoch": 1.727354411186612,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047102008535795134,
      "loss": 5.2846,
      "mean_token_accuracy": 0.20479981154203414,
      "num_tokens": 34318933.0,
      "step": 15380
    },
    {
      "entropy": 5.467982053756714,
      "epoch": 1.7279159880945696,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004710003721234934,
      "loss": 5.219,
      "mean_token_accuracy": 0.20281111299991608,
      "num_tokens": 34330431.0,
      "step": 15385
    },
    {
      "entropy": 5.538415193557739,
      "epoch": 1.7284775650025272,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047098065264830754,
      "loss": 5.3237,
      "mean_token_accuracy": 0.19124918133020402,
      "num_tokens": 34342006.0,
      "step": 15390
    },
    {
      "entropy": 5.49035267829895,
      "epoch": 1.7290391419104847,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047096092693302174,
      "loss": 5.2381,
      "mean_token_accuracy": 0.19346456080675126,
      "num_tokens": 34353390.0,
      "step": 15395
    },
    {
      "entropy": 5.392697095870972,
      "epoch": 1.7296007188184421,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047094119497826406,
      "loss": 5.2252,
      "mean_token_accuracy": 0.19382397532463075,
      "num_tokens": 34364211.0,
      "step": 15400
    },
    {
      "entropy": 5.492289972305298,
      "epoch": 1.7301622957263998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047092145678466273,
      "loss": 5.2501,
      "mean_token_accuracy": 0.19321486055850984,
      "num_tokens": 34375362.0,
      "step": 15405
    },
    {
      "entropy": 5.424299049377441,
      "epoch": 1.7307238726343572,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004709017123528462,
      "loss": 5.1217,
      "mean_token_accuracy": 0.2064367413520813,
      "num_tokens": 34385511.0,
      "step": 15410
    },
    {
      "entropy": 5.4561442852020265,
      "epoch": 1.7312854495423147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047088196168344315,
      "loss": 5.3609,
      "mean_token_accuracy": 0.1889231473207474,
      "num_tokens": 34397060.0,
      "step": 15415
    },
    {
      "entropy": 5.437074089050293,
      "epoch": 1.7318470264502723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047086220477708256,
      "loss": 5.26,
      "mean_token_accuracy": 0.1962616741657257,
      "num_tokens": 34407703.0,
      "step": 15420
    },
    {
      "entropy": 5.5071210861206055,
      "epoch": 1.73240860335823,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047084244163439335,
      "loss": 5.2773,
      "mean_token_accuracy": 0.198504975438118,
      "num_tokens": 34418929.0,
      "step": 15425
    },
    {
      "entropy": 5.429887056350708,
      "epoch": 1.7329701802661874,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004708226722560049,
      "loss": 5.1799,
      "mean_token_accuracy": 0.20248452574014664,
      "num_tokens": 34430693.0,
      "step": 15430
    },
    {
      "entropy": 5.418445587158203,
      "epoch": 1.7335317571741449,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004708028966425465,
      "loss": 5.2237,
      "mean_token_accuracy": 0.19408808797597885,
      "num_tokens": 34442299.0,
      "step": 15435
    },
    {
      "entropy": 5.463304042816162,
      "epoch": 1.7340933340821025,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047078311479464797,
      "loss": 5.2,
      "mean_token_accuracy": 0.1958053559064865,
      "num_tokens": 34453426.0,
      "step": 15440
    },
    {
      "entropy": 5.412197923660278,
      "epoch": 1.7346549109900602,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047076332671293917,
      "loss": 5.2327,
      "mean_token_accuracy": 0.19891296476125717,
      "num_tokens": 34463601.0,
      "step": 15445
    },
    {
      "entropy": 5.470721340179443,
      "epoch": 1.7352164878980176,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047074353239805005,
      "loss": 5.2399,
      "mean_token_accuracy": 0.19638391137123107,
      "num_tokens": 34474938.0,
      "step": 15450
    },
    {
      "entropy": 5.44405951499939,
      "epoch": 1.735778064805975,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004707237318506111,
      "loss": 5.1614,
      "mean_token_accuracy": 0.20331384241580963,
      "num_tokens": 34485244.0,
      "step": 15455
    },
    {
      "entropy": 5.372856950759887,
      "epoch": 1.7363396417139327,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047070392507125245,
      "loss": 5.1628,
      "mean_token_accuracy": 0.2047601506114006,
      "num_tokens": 34496912.0,
      "step": 15460
    },
    {
      "entropy": 5.367724132537842,
      "epoch": 1.7369012186218904,
      "grad_norm": 1.0,
      "learning_rate": 0.000470684112060605,
      "loss": 5.2424,
      "mean_token_accuracy": 0.20117949694395065,
      "num_tokens": 34508122.0,
      "step": 15465
    },
    {
      "entropy": 5.451254892349243,
      "epoch": 1.7374627955298478,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047066429281929944,
      "loss": 5.2435,
      "mean_token_accuracy": 0.19873499423265456,
      "num_tokens": 34519522.0,
      "step": 15470
    },
    {
      "entropy": 5.483699464797974,
      "epoch": 1.7380243724378053,
      "grad_norm": 1.03125,
      "learning_rate": 0.000470644467347967,
      "loss": 5.236,
      "mean_token_accuracy": 0.1985923245549202,
      "num_tokens": 34532832.0,
      "step": 15475
    },
    {
      "entropy": 5.485170793533325,
      "epoch": 1.738585949345763,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047062463564723884,
      "loss": 5.3564,
      "mean_token_accuracy": 0.19652123749256134,
      "num_tokens": 34544817.0,
      "step": 15480
    },
    {
      "entropy": 5.406769561767578,
      "epoch": 1.7391475262537206,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004706047977177464,
      "loss": 5.1687,
      "mean_token_accuracy": 0.19541578143835067,
      "num_tokens": 34555391.0,
      "step": 15485
    },
    {
      "entropy": 5.411785364151001,
      "epoch": 1.739709103161678,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047058495356012135,
      "loss": 5.2449,
      "mean_token_accuracy": 0.19859668761491775,
      "num_tokens": 34567531.0,
      "step": 15490
    },
    {
      "entropy": 5.455117177963257,
      "epoch": 1.7402706800696355,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047056510317499545,
      "loss": 5.2655,
      "mean_token_accuracy": 0.19563011229038238,
      "num_tokens": 34578212.0,
      "step": 15495
    },
    {
      "entropy": 5.396497392654419,
      "epoch": 1.7408322569775931,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047054524656300087,
      "loss": 5.1585,
      "mean_token_accuracy": 0.2035893142223358,
      "num_tokens": 34589331.0,
      "step": 15500
    },
    {
      "entropy": 5.37835464477539,
      "epoch": 1.7413938338855506,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047052538372476985,
      "loss": 5.2407,
      "mean_token_accuracy": 0.20102326571941376,
      "num_tokens": 34601467.0,
      "step": 15505
    },
    {
      "entropy": 5.386589241027832,
      "epoch": 1.741955410793508,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047050551466093466,
      "loss": 5.2042,
      "mean_token_accuracy": 0.20691325664520263,
      "num_tokens": 34611608.0,
      "step": 15510
    },
    {
      "entropy": 5.485888051986694,
      "epoch": 1.7425169877014657,
      "grad_norm": 1.125,
      "learning_rate": 0.0004704856393721281,
      "loss": 5.1735,
      "mean_token_accuracy": 0.20545875430107116,
      "num_tokens": 34622495.0,
      "step": 15515
    },
    {
      "entropy": 5.585084772109985,
      "epoch": 1.7430785646094233,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047046575785898297,
      "loss": 5.3814,
      "mean_token_accuracy": 0.19156555086374283,
      "num_tokens": 34633279.0,
      "step": 15520
    },
    {
      "entropy": 5.439335203170776,
      "epoch": 1.7436401415173808,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004704458701221323,
      "loss": 5.2352,
      "mean_token_accuracy": 0.19934522956609727,
      "num_tokens": 34644030.0,
      "step": 15525
    },
    {
      "entropy": 5.572230529785156,
      "epoch": 1.7442017184253382,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004704259761622094,
      "loss": 5.3478,
      "mean_token_accuracy": 0.18335260599851608,
      "num_tokens": 34655540.0,
      "step": 15530
    },
    {
      "entropy": 5.507998752593994,
      "epoch": 1.7447632953332959,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047040607597984745,
      "loss": 5.2581,
      "mean_token_accuracy": 0.19626591503620147,
      "num_tokens": 34666752.0,
      "step": 15535
    },
    {
      "entropy": 5.519788789749145,
      "epoch": 1.7453248722412535,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004703861695756804,
      "loss": 5.2682,
      "mean_token_accuracy": 0.18972883820533754,
      "num_tokens": 34678670.0,
      "step": 15540
    },
    {
      "entropy": 5.502021837234497,
      "epoch": 1.745886449149211,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004703662569503419,
      "loss": 5.2698,
      "mean_token_accuracy": 0.19230482280254363,
      "num_tokens": 34690848.0,
      "step": 15545
    },
    {
      "entropy": 5.496710634231567,
      "epoch": 1.7464480260571684,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047034633810446604,
      "loss": 5.307,
      "mean_token_accuracy": 0.1888074278831482,
      "num_tokens": 34703135.0,
      "step": 15550
    },
    {
      "entropy": 5.449506950378418,
      "epoch": 1.747009602965126,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004703264130386869,
      "loss": 5.2274,
      "mean_token_accuracy": 0.1977105662226677,
      "num_tokens": 34714329.0,
      "step": 15555
    },
    {
      "entropy": 5.447259521484375,
      "epoch": 1.7475711798730837,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047030648175363913,
      "loss": 5.2857,
      "mean_token_accuracy": 0.18809677958488463,
      "num_tokens": 34727032.0,
      "step": 15560
    },
    {
      "entropy": 5.492578363418579,
      "epoch": 1.7481327567810412,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004702865442499572,
      "loss": 5.2799,
      "mean_token_accuracy": 0.19404125064611435,
      "num_tokens": 34738600.0,
      "step": 15565
    },
    {
      "entropy": 5.538525152206421,
      "epoch": 1.7486943336889986,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004702666005282759,
      "loss": 5.3242,
      "mean_token_accuracy": 0.1916448712348938,
      "num_tokens": 34749810.0,
      "step": 15570
    },
    {
      "entropy": 5.664850044250488,
      "epoch": 1.7492559105969563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004702466505892304,
      "loss": 5.4812,
      "mean_token_accuracy": 0.18054275512695311,
      "num_tokens": 34762029.0,
      "step": 15575
    },
    {
      "entropy": 5.4572672843933105,
      "epoch": 1.749817487504914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047022669443345575,
      "loss": 5.3264,
      "mean_token_accuracy": 0.19237176775932313,
      "num_tokens": 34773160.0,
      "step": 15580
    },
    {
      "entropy": 5.430907821655273,
      "epoch": 1.7503790644128714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047020673206158743,
      "loss": 5.3018,
      "mean_token_accuracy": 0.1918123260140419,
      "num_tokens": 34784659.0,
      "step": 15585
    },
    {
      "entropy": 5.607436609268189,
      "epoch": 1.7509406413208288,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000470186763474261,
      "loss": 5.2669,
      "mean_token_accuracy": 0.19227909594774245,
      "num_tokens": 34796976.0,
      "step": 15590
    },
    {
      "entropy": 5.456044387817383,
      "epoch": 1.7515022182287865,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004701667886721125,
      "loss": 5.2123,
      "mean_token_accuracy": 0.19207327514886857,
      "num_tokens": 34809271.0,
      "step": 15595
    },
    {
      "entropy": 5.47929048538208,
      "epoch": 1.752063795136744,
      "grad_norm": 1.125,
      "learning_rate": 0.0004701468076557776,
      "loss": 5.2911,
      "mean_token_accuracy": 0.1998542532324791,
      "num_tokens": 34820291.0,
      "step": 15600
    },
    {
      "entropy": 5.388984823226929,
      "epoch": 1.7526253720447014,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047012682042589277,
      "loss": 5.2063,
      "mean_token_accuracy": 0.20087198317050933,
      "num_tokens": 34831224.0,
      "step": 15605
    },
    {
      "entropy": 5.501365089416504,
      "epoch": 1.753186948952659,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047010682698309423,
      "loss": 5.2765,
      "mean_token_accuracy": 0.19637789130210875,
      "num_tokens": 34842947.0,
      "step": 15610
    },
    {
      "entropy": 5.398352193832397,
      "epoch": 1.7537485258606167,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004700868273280186,
      "loss": 5.2275,
      "mean_token_accuracy": 0.2049104556441307,
      "num_tokens": 34853422.0,
      "step": 15615
    },
    {
      "entropy": 5.430916976928711,
      "epoch": 1.7543101027685741,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047006682146130273,
      "loss": 5.232,
      "mean_token_accuracy": 0.19215001314878463,
      "num_tokens": 34864082.0,
      "step": 15620
    },
    {
      "entropy": 5.470525598526001,
      "epoch": 1.7548716796765316,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004700468093835837,
      "loss": 5.2221,
      "mean_token_accuracy": 0.20426664054393767,
      "num_tokens": 34874094.0,
      "step": 15625
    },
    {
      "entropy": 5.4323986053466795,
      "epoch": 1.7554332565844892,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047002679109549864,
      "loss": 5.2472,
      "mean_token_accuracy": 0.20054281800985335,
      "num_tokens": 34884597.0,
      "step": 15630
    },
    {
      "entropy": 5.378446292877197,
      "epoch": 1.755994833492447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047000676659768477,
      "loss": 5.1238,
      "mean_token_accuracy": 0.20713614374399186,
      "num_tokens": 34894746.0,
      "step": 15635
    },
    {
      "entropy": 5.4310376167297365,
      "epoch": 1.7565564104004043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004699867358907799,
      "loss": 5.1406,
      "mean_token_accuracy": 0.2001330718398094,
      "num_tokens": 34906250.0,
      "step": 15640
    },
    {
      "entropy": 5.482785892486572,
      "epoch": 1.7571179873083618,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004699666989754217,
      "loss": 5.4081,
      "mean_token_accuracy": 0.1867247298359871,
      "num_tokens": 34917772.0,
      "step": 15645
    },
    {
      "entropy": 5.500496768951416,
      "epoch": 1.7576795642163194,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004699466558522482,
      "loss": 5.2443,
      "mean_token_accuracy": 0.2001166522502899,
      "num_tokens": 34930243.0,
      "step": 15650
    },
    {
      "entropy": 5.471747255325317,
      "epoch": 1.758241141124277,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046992660652189755,
      "loss": 5.2573,
      "mean_token_accuracy": 0.19575004726648332,
      "num_tokens": 34941357.0,
      "step": 15655
    },
    {
      "entropy": 5.59677939414978,
      "epoch": 1.7588027180322345,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046990655098500824,
      "loss": 5.3896,
      "mean_token_accuracy": 0.18695174157619476,
      "num_tokens": 34952116.0,
      "step": 15660
    },
    {
      "entropy": 5.5312329769134525,
      "epoch": 1.759364294940192,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046988648924221866,
      "loss": 5.2817,
      "mean_token_accuracy": 0.19602534025907517,
      "num_tokens": 34963707.0,
      "step": 15665
    },
    {
      "entropy": 5.5054035663604735,
      "epoch": 1.7599258718481496,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004698664212941677,
      "loss": 5.3786,
      "mean_token_accuracy": 0.19636891037225723,
      "num_tokens": 34974376.0,
      "step": 15670
    },
    {
      "entropy": 5.486743783950805,
      "epoch": 1.7604874487561073,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046984634714149437,
      "loss": 5.2709,
      "mean_token_accuracy": 0.19557513743638993,
      "num_tokens": 34986183.0,
      "step": 15675
    },
    {
      "entropy": 5.59564528465271,
      "epoch": 1.7610490256640647,
      "grad_norm": 1.125,
      "learning_rate": 0.0004698262667848377,
      "loss": 5.2901,
      "mean_token_accuracy": 0.19822811633348464,
      "num_tokens": 34997137.0,
      "step": 15680
    },
    {
      "entropy": 5.615262269973755,
      "epoch": 1.7616106025720222,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046980618022483714,
      "loss": 5.4418,
      "mean_token_accuracy": 0.18901116251945496,
      "num_tokens": 35008944.0,
      "step": 15685
    },
    {
      "entropy": 5.52647852897644,
      "epoch": 1.7621721794799798,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046978608746213224,
      "loss": 5.3271,
      "mean_token_accuracy": 0.18539503663778306,
      "num_tokens": 35019972.0,
      "step": 15690
    },
    {
      "entropy": 5.438716888427734,
      "epoch": 1.7627337563879373,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004697659884973628,
      "loss": 5.2,
      "mean_token_accuracy": 0.19455612301826478,
      "num_tokens": 35031285.0,
      "step": 15695
    },
    {
      "entropy": 5.41926679611206,
      "epoch": 1.7632953332958947,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004697458833311687,
      "loss": 5.1745,
      "mean_token_accuracy": 0.1998610109090805,
      "num_tokens": 35042553.0,
      "step": 15700
    },
    {
      "entropy": 5.589274215698242,
      "epoch": 1.7638569102038524,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004697257719641902,
      "loss": 5.4069,
      "mean_token_accuracy": 0.1867491528391838,
      "num_tokens": 35053637.0,
      "step": 15705
    },
    {
      "entropy": 5.545748233795166,
      "epoch": 1.76441848711181,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004697056543970676,
      "loss": 5.3255,
      "mean_token_accuracy": 0.19714967012405396,
      "num_tokens": 35064754.0,
      "step": 15710
    },
    {
      "entropy": 5.516838312149048,
      "epoch": 1.7649800640197675,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046968553063044146,
      "loss": 5.2512,
      "mean_token_accuracy": 0.19179097414016724,
      "num_tokens": 35075816.0,
      "step": 15715
    },
    {
      "entropy": 5.488126039505005,
      "epoch": 1.765541640927725,
      "grad_norm": 1.25,
      "learning_rate": 0.00046966540066495247,
      "loss": 5.3452,
      "mean_token_accuracy": 0.1930240049958229,
      "num_tokens": 35088647.0,
      "step": 15720
    },
    {
      "entropy": 5.480515480041504,
      "epoch": 1.7661032178356826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046964526450124164,
      "loss": 5.2294,
      "mean_token_accuracy": 0.19629358798265456,
      "num_tokens": 35099372.0,
      "step": 15725
    },
    {
      "entropy": 5.499585437774658,
      "epoch": 1.7666647947436402,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046962512213995,
      "loss": 5.3761,
      "mean_token_accuracy": 0.18661774545907975,
      "num_tokens": 35110663.0,
      "step": 15730
    },
    {
      "entropy": 5.648336124420166,
      "epoch": 1.7672263716515977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046960497358171913,
      "loss": 5.4534,
      "mean_token_accuracy": 0.18458335548639299,
      "num_tokens": 35122406.0,
      "step": 15735
    },
    {
      "entropy": 5.513351345062256,
      "epoch": 1.7677879485595551,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046958481882719035,
      "loss": 5.2351,
      "mean_token_accuracy": 0.19630270451307297,
      "num_tokens": 35133444.0,
      "step": 15740
    },
    {
      "entropy": 5.438093185424805,
      "epoch": 1.7683495254675128,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046956465787700554,
      "loss": 5.2382,
      "mean_token_accuracy": 0.19741030931472778,
      "num_tokens": 35143840.0,
      "step": 15745
    },
    {
      "entropy": 5.566193914413452,
      "epoch": 1.7689111023754704,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046954449073180644,
      "loss": 5.4177,
      "mean_token_accuracy": 0.1819252073764801,
      "num_tokens": 35155050.0,
      "step": 15750
    },
    {
      "entropy": 5.610992860794068,
      "epoch": 1.7694726792834279,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004695243173922354,
      "loss": 5.3381,
      "mean_token_accuracy": 0.19293665736913682,
      "num_tokens": 35167649.0,
      "step": 15755
    },
    {
      "entropy": 5.490816593170166,
      "epoch": 1.7700342561913853,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046950413785893465,
      "loss": 5.2591,
      "mean_token_accuracy": 0.19580239206552505,
      "num_tokens": 35179238.0,
      "step": 15760
    },
    {
      "entropy": 5.520217037200927,
      "epoch": 1.770595833099343,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046948395213254667,
      "loss": 5.309,
      "mean_token_accuracy": 0.19635985642671586,
      "num_tokens": 35189510.0,
      "step": 15765
    },
    {
      "entropy": 5.525603818893432,
      "epoch": 1.7711574100073006,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004694637602137143,
      "loss": 5.2999,
      "mean_token_accuracy": 0.19275762587785722,
      "num_tokens": 35199727.0,
      "step": 15770
    },
    {
      "entropy": 5.445923900604248,
      "epoch": 1.771718986915258,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004694435621030803,
      "loss": 5.2391,
      "mean_token_accuracy": 0.2015764221549034,
      "num_tokens": 35210311.0,
      "step": 15775
    },
    {
      "entropy": 5.498425102233886,
      "epoch": 1.7722805638232155,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046942335780128785,
      "loss": 5.2613,
      "mean_token_accuracy": 0.19993872940540314,
      "num_tokens": 35222755.0,
      "step": 15780
    },
    {
      "entropy": 5.504217624664307,
      "epoch": 1.7728421407311732,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046940314730898035,
      "loss": 5.2922,
      "mean_token_accuracy": 0.19344490617513657,
      "num_tokens": 35234337.0,
      "step": 15785
    },
    {
      "entropy": 5.456560945510864,
      "epoch": 1.7734037176391306,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004693829306268012,
      "loss": 5.2553,
      "mean_token_accuracy": 0.19621694087982178,
      "num_tokens": 35245736.0,
      "step": 15790
    },
    {
      "entropy": 5.473315620422364,
      "epoch": 1.773965294547088,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004693627077553942,
      "loss": 5.2117,
      "mean_token_accuracy": 0.20478270202875137,
      "num_tokens": 35256368.0,
      "step": 15795
    },
    {
      "entropy": 5.440543603897095,
      "epoch": 1.7745268714550457,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004693424786954031,
      "loss": 5.2918,
      "mean_token_accuracy": 0.19716603606939315,
      "num_tokens": 35267238.0,
      "step": 15800
    },
    {
      "entropy": 5.477315092086792,
      "epoch": 1.7750884483630034,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046932224344747224,
      "loss": 5.2303,
      "mean_token_accuracy": 0.1946992829442024,
      "num_tokens": 35277113.0,
      "step": 15805
    },
    {
      "entropy": 5.5034326076507565,
      "epoch": 1.7756500252709608,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046930200201224573,
      "loss": 5.1824,
      "mean_token_accuracy": 0.1909641832113266,
      "num_tokens": 35288504.0,
      "step": 15810
    },
    {
      "entropy": 5.443775177001953,
      "epoch": 1.7762116021789183,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004692817543903681,
      "loss": 5.2202,
      "mean_token_accuracy": 0.2007802963256836,
      "num_tokens": 35301215.0,
      "step": 15815
    },
    {
      "entropy": 5.40267972946167,
      "epoch": 1.776773179086876,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000469261500582484,
      "loss": 5.2132,
      "mean_token_accuracy": 0.19696372598409653,
      "num_tokens": 35311997.0,
      "step": 15820
    },
    {
      "entropy": 5.481546401977539,
      "epoch": 1.7773347559948336,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046924124058923843,
      "loss": 5.2891,
      "mean_token_accuracy": 0.1924848809838295,
      "num_tokens": 35323217.0,
      "step": 15825
    },
    {
      "entropy": 5.484462308883667,
      "epoch": 1.777896332902791,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004692209744112765,
      "loss": 5.3161,
      "mean_token_accuracy": 0.19927048087120056,
      "num_tokens": 35333102.0,
      "step": 15830
    },
    {
      "entropy": 5.587183046340942,
      "epoch": 1.7784579098107485,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004692007020492433,
      "loss": 5.3372,
      "mean_token_accuracy": 0.19091015607118605,
      "num_tokens": 35346394.0,
      "step": 15835
    },
    {
      "entropy": 5.3949127197265625,
      "epoch": 1.7790194867187061,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004691804235037845,
      "loss": 5.1393,
      "mean_token_accuracy": 0.203298519551754,
      "num_tokens": 35357827.0,
      "step": 15840
    },
    {
      "entropy": 5.410745096206665,
      "epoch": 1.7795810636266638,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004691601387755456,
      "loss": 5.2675,
      "mean_token_accuracy": 0.1943773955106735,
      "num_tokens": 35370517.0,
      "step": 15845
    },
    {
      "entropy": 5.503803491592407,
      "epoch": 1.7801426405346212,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046913984786517275,
      "loss": 5.2924,
      "mean_token_accuracy": 0.1901368245482445,
      "num_tokens": 35381830.0,
      "step": 15850
    },
    {
      "entropy": 5.606777811050415,
      "epoch": 1.7807042174425787,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046911955077331177,
      "loss": 5.3757,
      "mean_token_accuracy": 0.18714867383241654,
      "num_tokens": 35393161.0,
      "step": 15855
    },
    {
      "entropy": 5.52755184173584,
      "epoch": 1.7812657943505363,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046909924750060893,
      "loss": 5.2141,
      "mean_token_accuracy": 0.20724063068628312,
      "num_tokens": 35403338.0,
      "step": 15860
    },
    {
      "entropy": 5.49515929222107,
      "epoch": 1.781827371258494,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004690789380477109,
      "loss": 5.3072,
      "mean_token_accuracy": 0.19043599367141723,
      "num_tokens": 35415862.0,
      "step": 15865
    },
    {
      "entropy": 5.427636623382568,
      "epoch": 1.7823889481664514,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004690586224152641,
      "loss": 5.22,
      "mean_token_accuracy": 0.19370841532945632,
      "num_tokens": 35426696.0,
      "step": 15870
    },
    {
      "entropy": 5.492132997512817,
      "epoch": 1.7829505250744089,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004690383006039156,
      "loss": 5.2803,
      "mean_token_accuracy": 0.19535786509513856,
      "num_tokens": 35438624.0,
      "step": 15875
    },
    {
      "entropy": 5.583156633377075,
      "epoch": 1.7835121019823665,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004690179726143123,
      "loss": 5.3933,
      "mean_token_accuracy": 0.1896327555179596,
      "num_tokens": 35449841.0,
      "step": 15880
    },
    {
      "entropy": 5.679481792449951,
      "epoch": 1.784073678890324,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004689976384471014,
      "loss": 5.4925,
      "mean_token_accuracy": 0.18230774253606796,
      "num_tokens": 35462081.0,
      "step": 15885
    },
    {
      "entropy": 5.437917423248291,
      "epoch": 1.7846352557982814,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004689772981029306,
      "loss": 5.1847,
      "mean_token_accuracy": 0.20067851096391678,
      "num_tokens": 35472834.0,
      "step": 15890
    },
    {
      "entropy": 5.461607217788696,
      "epoch": 1.785196832706239,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004689569515824473,
      "loss": 5.1429,
      "mean_token_accuracy": 0.20348993986845015,
      "num_tokens": 35482975.0,
      "step": 15895
    },
    {
      "entropy": 5.392340517044067,
      "epoch": 1.7857584096141967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004689365988862995,
      "loss": 5.2522,
      "mean_token_accuracy": 0.19824526607990264,
      "num_tokens": 35494470.0,
      "step": 15900
    },
    {
      "entropy": 5.408385181427002,
      "epoch": 1.7863199865221542,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046891624001513515,
      "loss": 5.199,
      "mean_token_accuracy": 0.19572625756263734,
      "num_tokens": 35505464.0,
      "step": 15905
    },
    {
      "entropy": 5.428882598876953,
      "epoch": 1.7868815634301116,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046889587496960257,
      "loss": 5.2491,
      "mean_token_accuracy": 0.1960815206170082,
      "num_tokens": 35517609.0,
      "step": 15910
    },
    {
      "entropy": 5.457346963882446,
      "epoch": 1.7874431403380693,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046887550375034997,
      "loss": 5.1663,
      "mean_token_accuracy": 0.19810595065355302,
      "num_tokens": 35528994.0,
      "step": 15915
    },
    {
      "entropy": 5.498655128479004,
      "epoch": 1.788004717246027,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046885512635802627,
      "loss": 5.3601,
      "mean_token_accuracy": 0.18583411127328872,
      "num_tokens": 35541044.0,
      "step": 15920
    },
    {
      "entropy": 5.401097106933594,
      "epoch": 1.7885662941539844,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004688347427932801,
      "loss": 5.1783,
      "mean_token_accuracy": 0.2050304338335991,
      "num_tokens": 35552660.0,
      "step": 15925
    },
    {
      "entropy": 5.543985891342163,
      "epoch": 1.7891278710619418,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046881435305676055,
      "loss": 5.2892,
      "mean_token_accuracy": 0.18554833233356477,
      "num_tokens": 35565144.0,
      "step": 15930
    },
    {
      "entropy": 5.499610090255738,
      "epoch": 1.7896894479698995,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046879395714911686,
      "loss": 5.2466,
      "mean_token_accuracy": 0.19383473843336105,
      "num_tokens": 35577067.0,
      "step": 15935
    },
    {
      "entropy": 5.408252334594726,
      "epoch": 1.7902510248778571,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046877355507099846,
      "loss": 5.2292,
      "mean_token_accuracy": 0.19355790317058563,
      "num_tokens": 35588712.0,
      "step": 15940
    },
    {
      "entropy": 5.494270086288452,
      "epoch": 1.7908126017858146,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004687531468230549,
      "loss": 5.1873,
      "mean_token_accuracy": 0.19448510706424713,
      "num_tokens": 35599212.0,
      "step": 15945
    },
    {
      "entropy": 5.500825452804565,
      "epoch": 1.791374178693772,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004687327324059359,
      "loss": 5.24,
      "mean_token_accuracy": 0.20169495046138763,
      "num_tokens": 35609397.0,
      "step": 15950
    },
    {
      "entropy": 5.444353532791138,
      "epoch": 1.7919357556017297,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004687123118202916,
      "loss": 5.2399,
      "mean_token_accuracy": 0.19417669177055358,
      "num_tokens": 35620615.0,
      "step": 15955
    },
    {
      "entropy": 5.504662752151489,
      "epoch": 1.7924973325096873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004686918850667722,
      "loss": 5.2735,
      "mean_token_accuracy": 0.19662481546401978,
      "num_tokens": 35632684.0,
      "step": 15960
    },
    {
      "entropy": 5.451061630249024,
      "epoch": 1.7930589094176448,
      "grad_norm": 1.125,
      "learning_rate": 0.000468671452146028,
      "loss": 5.27,
      "mean_token_accuracy": 0.1996163472533226,
      "num_tokens": 35643896.0,
      "step": 15965
    },
    {
      "entropy": 5.5074351787567135,
      "epoch": 1.7936204863256022,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046865101305870965,
      "loss": 5.2564,
      "mean_token_accuracy": 0.19853272438049316,
      "num_tokens": 35655752.0,
      "step": 15970
    },
    {
      "entropy": 5.398624134063721,
      "epoch": 1.7941820632335599,
      "grad_norm": 1.09375,
      "learning_rate": 0.000468630567805468,
      "loss": 5.1545,
      "mean_token_accuracy": 0.19439476281404494,
      "num_tokens": 35665870.0,
      "step": 15975
    },
    {
      "entropy": 5.4185138702392575,
      "epoch": 1.7947436401415173,
      "grad_norm": 1.125,
      "learning_rate": 0.000468610116386954,
      "loss": 5.3174,
      "mean_token_accuracy": 0.19464239329099656,
      "num_tokens": 35677792.0,
      "step": 15980
    },
    {
      "entropy": 5.4135995388031,
      "epoch": 1.7953052170494748,
      "grad_norm": 1.125,
      "learning_rate": 0.0004685896588038187,
      "loss": 5.2182,
      "mean_token_accuracy": 0.19553089141845703,
      "num_tokens": 35689059.0,
      "step": 15985
    },
    {
      "entropy": 5.543794345855713,
      "epoch": 1.7958667939574324,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004685691950567136,
      "loss": 5.3074,
      "mean_token_accuracy": 0.18938867300748824,
      "num_tokens": 35699953.0,
      "step": 15990
    },
    {
      "entropy": 5.529967403411865,
      "epoch": 1.79642837086539,
      "grad_norm": 1.125,
      "learning_rate": 0.00046854872514629035,
      "loss": 5.2201,
      "mean_token_accuracy": 0.19743361026048661,
      "num_tokens": 35710894.0,
      "step": 15995
    },
    {
      "entropy": 5.441607904434204,
      "epoch": 1.7969899477733475,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004685282490732005,
      "loss": 5.2778,
      "mean_token_accuracy": 0.1943812847137451,
      "num_tokens": 35722440.0,
      "step": 16000
    },
    {
      "entropy": 5.451500701904297,
      "epoch": 1.797551524681305,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046850776683809623,
      "loss": 5.282,
      "mean_token_accuracy": 0.19846251755952835,
      "num_tokens": 35733768.0,
      "step": 16005
    },
    {
      "entropy": 5.510002946853637,
      "epoch": 1.7981131015892626,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004684872784416295,
      "loss": 5.2399,
      "mean_token_accuracy": 0.19835041761398314,
      "num_tokens": 35744211.0,
      "step": 16010
    },
    {
      "entropy": 5.4808416843414305,
      "epoch": 1.7986746784972203,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004684667838844529,
      "loss": 5.2539,
      "mean_token_accuracy": 0.19498499482870102,
      "num_tokens": 35755498.0,
      "step": 16015
    },
    {
      "entropy": 5.5295497417449955,
      "epoch": 1.7992362554051777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004684462831672188,
      "loss": 5.2788,
      "mean_token_accuracy": 0.19888902455568314,
      "num_tokens": 35767442.0,
      "step": 16020
    },
    {
      "entropy": 5.436308908462524,
      "epoch": 1.7997978323131352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046842577629058006,
      "loss": 5.2343,
      "mean_token_accuracy": 0.19678854644298555,
      "num_tokens": 35778040.0,
      "step": 16025
    },
    {
      "entropy": 5.4519213199615475,
      "epoch": 1.8003594092210928,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004684052632551896,
      "loss": 5.1788,
      "mean_token_accuracy": 0.20557676255702972,
      "num_tokens": 35788139.0,
      "step": 16030
    },
    {
      "entropy": 5.4884552478790285,
      "epoch": 1.8009209861290505,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046838474406170046,
      "loss": 5.2338,
      "mean_token_accuracy": 0.19548669308423997,
      "num_tokens": 35799077.0,
      "step": 16035
    },
    {
      "entropy": 5.480828952789307,
      "epoch": 1.801482563037008,
      "grad_norm": 1.125,
      "learning_rate": 0.00046836421871076616,
      "loss": 5.2878,
      "mean_token_accuracy": 0.19201088696718216,
      "num_tokens": 35809748.0,
      "step": 16040
    },
    {
      "entropy": 5.55096321105957,
      "epoch": 1.8020441399449654,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004683436872030401,
      "loss": 5.3038,
      "mean_token_accuracy": 0.19516561478376387,
      "num_tokens": 35821621.0,
      "step": 16045
    },
    {
      "entropy": 5.522806835174561,
      "epoch": 1.802605716852923,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046832314953917605,
      "loss": 5.2519,
      "mean_token_accuracy": 0.1982177034020424,
      "num_tokens": 35833497.0,
      "step": 16050
    },
    {
      "entropy": 5.539760637283325,
      "epoch": 1.8031672937608807,
      "grad_norm": 1.125,
      "learning_rate": 0.00046830260571982795,
      "loss": 5.3593,
      "mean_token_accuracy": 0.19536954164505005,
      "num_tokens": 35844266.0,
      "step": 16055
    },
    {
      "entropy": 5.587708520889282,
      "epoch": 1.8037288706688381,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004682820557456499,
      "loss": 5.3602,
      "mean_token_accuracy": 0.19403456151485443,
      "num_tokens": 35855260.0,
      "step": 16060
    },
    {
      "entropy": 5.535833501815796,
      "epoch": 1.8042904475767956,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046826149961729625,
      "loss": 5.2961,
      "mean_token_accuracy": 0.20049406588077545,
      "num_tokens": 35865950.0,
      "step": 16065
    },
    {
      "entropy": 5.503913021087646,
      "epoch": 1.8048520244847532,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046824093733542144,
      "loss": 5.445,
      "mean_token_accuracy": 0.18648439794778823,
      "num_tokens": 35877028.0,
      "step": 16070
    },
    {
      "entropy": 5.492651748657226,
      "epoch": 1.8054136013927107,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046822036890068033,
      "loss": 5.2894,
      "mean_token_accuracy": 0.19604576826095582,
      "num_tokens": 35887831.0,
      "step": 16075
    },
    {
      "entropy": 5.5385983943939205,
      "epoch": 1.8059751783006681,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046819979431372773,
      "loss": 5.2629,
      "mean_token_accuracy": 0.19517093896865845,
      "num_tokens": 35897946.0,
      "step": 16080
    },
    {
      "entropy": 5.603180360794068,
      "epoch": 1.8065367552086258,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046817921357521865,
      "loss": 5.4066,
      "mean_token_accuracy": 0.19121986776590347,
      "num_tokens": 35910973.0,
      "step": 16085
    },
    {
      "entropy": 5.517197322845459,
      "epoch": 1.8070983321165834,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046815862668580854,
      "loss": 5.3279,
      "mean_token_accuracy": 0.1891381949186325,
      "num_tokens": 35921478.0,
      "step": 16090
    },
    {
      "entropy": 5.503784275054931,
      "epoch": 1.8076599090245409,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004681380336461528,
      "loss": 5.2078,
      "mean_token_accuracy": 0.20318155437707902,
      "num_tokens": 35932211.0,
      "step": 16095
    },
    {
      "entropy": 5.4260437965393065,
      "epoch": 1.8082214859324983,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00046811743445690724,
      "loss": 5.3806,
      "mean_token_accuracy": 0.19121943563222885,
      "num_tokens": 35944350.0,
      "step": 16100
    },
    {
      "entropy": 5.521867036819458,
      "epoch": 1.808783062840456,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004680968291187276,
      "loss": 5.3439,
      "mean_token_accuracy": 0.18876016587018968,
      "num_tokens": 35955590.0,
      "step": 16105
    },
    {
      "entropy": 5.442447137832642,
      "epoch": 1.8093446397484136,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004680762176322701,
      "loss": 5.1196,
      "mean_token_accuracy": 0.21004122346639634,
      "num_tokens": 35965872.0,
      "step": 16110
    },
    {
      "entropy": 5.381133699417115,
      "epoch": 1.809906216656371,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046805559999819093,
      "loss": 5.2415,
      "mean_token_accuracy": 0.1935344398021698,
      "num_tokens": 35976851.0,
      "step": 16115
    },
    {
      "entropy": 5.4849670886993405,
      "epoch": 1.8104677935643285,
      "grad_norm": 1.125,
      "learning_rate": 0.0004680349762171465,
      "loss": 5.2041,
      "mean_token_accuracy": 0.19965680986642836,
      "num_tokens": 35986984.0,
      "step": 16120
    },
    {
      "entropy": 5.505698776245117,
      "epoch": 1.8110293704722862,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004680143462897936,
      "loss": 5.2773,
      "mean_token_accuracy": 0.19423348158597947,
      "num_tokens": 35999349.0,
      "step": 16125
    },
    {
      "entropy": 5.449534463882446,
      "epoch": 1.8115909473802438,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004679937102167891,
      "loss": 5.2896,
      "mean_token_accuracy": 0.20000939965248107,
      "num_tokens": 36011122.0,
      "step": 16130
    },
    {
      "entropy": 5.513085412979126,
      "epoch": 1.8121525242882013,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046797306799879,
      "loss": 5.2164,
      "mean_token_accuracy": 0.19834259301424026,
      "num_tokens": 36022729.0,
      "step": 16135
    },
    {
      "entropy": 5.4503247261047365,
      "epoch": 1.8127141011961587,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046795241963645356,
      "loss": 5.2182,
      "mean_token_accuracy": 0.19925459176301957,
      "num_tokens": 36034042.0,
      "step": 16140
    },
    {
      "entropy": 5.467282438278199,
      "epoch": 1.8132756781041164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004679317651304373,
      "loss": 5.2626,
      "mean_token_accuracy": 0.1887116625905037,
      "num_tokens": 36045357.0,
      "step": 16145
    },
    {
      "entropy": 5.350513744354248,
      "epoch": 1.813837255012074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004679111044813987,
      "loss": 5.0236,
      "mean_token_accuracy": 0.20837678462266923,
      "num_tokens": 36055180.0,
      "step": 16150
    },
    {
      "entropy": 5.3760003566741945,
      "epoch": 1.8143988319200315,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046789043768999573,
      "loss": 5.176,
      "mean_token_accuracy": 0.20194455981254578,
      "num_tokens": 36065574.0,
      "step": 16155
    },
    {
      "entropy": 5.428332757949829,
      "epoch": 1.814960408827989,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004678697647568865,
      "loss": 5.2658,
      "mean_token_accuracy": 0.1963086485862732,
      "num_tokens": 36077316.0,
      "step": 16160
    },
    {
      "entropy": 5.414357376098633,
      "epoch": 1.8155219857359466,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046784908568272903,
      "loss": 5.2145,
      "mean_token_accuracy": 0.19487461745738982,
      "num_tokens": 36088735.0,
      "step": 16165
    },
    {
      "entropy": 5.44441294670105,
      "epoch": 1.8160835626439042,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000467828400468182,
      "loss": 5.2645,
      "mean_token_accuracy": 0.19373282343149184,
      "num_tokens": 36098806.0,
      "step": 16170
    },
    {
      "entropy": 5.493496513366699,
      "epoch": 1.8166451395518615,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046780770911390387,
      "loss": 5.3036,
      "mean_token_accuracy": 0.19467793405056,
      "num_tokens": 36110333.0,
      "step": 16175
    },
    {
      "entropy": 5.5141997814178465,
      "epoch": 1.8172067164598191,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004677870116205534,
      "loss": 5.2599,
      "mean_token_accuracy": 0.19326572716236115,
      "num_tokens": 36121719.0,
      "step": 16180
    },
    {
      "entropy": 5.447924518585205,
      "epoch": 1.8177682933677768,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046776630798878985,
      "loss": 5.1539,
      "mean_token_accuracy": 0.2053838238120079,
      "num_tokens": 36132129.0,
      "step": 16185
    },
    {
      "entropy": 5.5286133766174315,
      "epoch": 1.8183298702757342,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004677455982192723,
      "loss": 5.4494,
      "mean_token_accuracy": 0.1912141978740692,
      "num_tokens": 36145189.0,
      "step": 16190
    },
    {
      "entropy": 5.450881576538086,
      "epoch": 1.8188914471836917,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046772488231266004,
      "loss": 5.1935,
      "mean_token_accuracy": 0.1981054201722145,
      "num_tokens": 36157896.0,
      "step": 16195
    },
    {
      "entropy": 5.522910785675049,
      "epoch": 1.8194530240916493,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046770416026961284,
      "loss": 5.3051,
      "mean_token_accuracy": 0.19353700876235963,
      "num_tokens": 36168842.0,
      "step": 16200
    },
    {
      "entropy": 5.356402206420898,
      "epoch": 1.820014600999607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004676834320907904,
      "loss": 5.1526,
      "mean_token_accuracy": 0.20072351694107055,
      "num_tokens": 36180075.0,
      "step": 16205
    },
    {
      "entropy": 5.407615089416504,
      "epoch": 1.8205761779075644,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004676626977768528,
      "loss": 5.196,
      "mean_token_accuracy": 0.2022756114602089,
      "num_tokens": 36190589.0,
      "step": 16210
    },
    {
      "entropy": 5.39691276550293,
      "epoch": 1.8211377548155219,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046764195732846013,
      "loss": 5.0724,
      "mean_token_accuracy": 0.20610568076372146,
      "num_tokens": 36201145.0,
      "step": 16215
    },
    {
      "entropy": 5.43857479095459,
      "epoch": 1.8216993317234795,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046762121074627285,
      "loss": 5.2228,
      "mean_token_accuracy": 0.19864938110113145,
      "num_tokens": 36213243.0,
      "step": 16220
    },
    {
      "entropy": 5.353356170654297,
      "epoch": 1.8222609086314372,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046760045803095153,
      "loss": 5.1938,
      "mean_token_accuracy": 0.2008663147687912,
      "num_tokens": 36224491.0,
      "step": 16225
    },
    {
      "entropy": 5.436663484573364,
      "epoch": 1.8228224855393946,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004675796991831569,
      "loss": 5.2452,
      "mean_token_accuracy": 0.19710069447755812,
      "num_tokens": 36236006.0,
      "step": 16230
    },
    {
      "entropy": 5.587708473205566,
      "epoch": 1.823384062447352,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046755893420355,
      "loss": 5.3377,
      "mean_token_accuracy": 0.1891933336853981,
      "num_tokens": 36246925.0,
      "step": 16235
    },
    {
      "entropy": 5.518993234634399,
      "epoch": 1.8239456393553097,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004675381630927919,
      "loss": 5.2537,
      "mean_token_accuracy": 0.19068574011325837,
      "num_tokens": 36258449.0,
      "step": 16240
    },
    {
      "entropy": 5.467200994491577,
      "epoch": 1.8245072162632674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046751738585154405,
      "loss": 5.2591,
      "mean_token_accuracy": 0.19682479798793792,
      "num_tokens": 36270820.0,
      "step": 16245
    },
    {
      "entropy": 5.498639965057373,
      "epoch": 1.8250687931712248,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046749660248046794,
      "loss": 5.3254,
      "mean_token_accuracy": 0.19491575211286544,
      "num_tokens": 36282346.0,
      "step": 16250
    },
    {
      "entropy": 5.545383739471435,
      "epoch": 1.8256303700791823,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046747581298022537,
      "loss": 5.3515,
      "mean_token_accuracy": 0.18515999615192413,
      "num_tokens": 36294111.0,
      "step": 16255
    },
    {
      "entropy": 5.48102822303772,
      "epoch": 1.82619194698714,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004674550173514782,
      "loss": 5.2159,
      "mean_token_accuracy": 0.1969439998269081,
      "num_tokens": 36304899.0,
      "step": 16260
    },
    {
      "entropy": 5.483670806884765,
      "epoch": 1.8267535238950976,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046743421559488875,
      "loss": 5.3663,
      "mean_token_accuracy": 0.18919069468975067,
      "num_tokens": 36316272.0,
      "step": 16265
    },
    {
      "entropy": 5.444512605667114,
      "epoch": 1.8273151008030548,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046741340771111917,
      "loss": 5.1498,
      "mean_token_accuracy": 0.20142329186201097,
      "num_tokens": 36326624.0,
      "step": 16270
    },
    {
      "entropy": 5.44782829284668,
      "epoch": 1.8278766777110125,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046739259370083205,
      "loss": 5.1643,
      "mean_token_accuracy": 0.20240344554185868,
      "num_tokens": 36338178.0,
      "step": 16275
    },
    {
      "entropy": 5.51089334487915,
      "epoch": 1.8284382546189701,
      "grad_norm": 1.125,
      "learning_rate": 0.0004673717735646902,
      "loss": 5.2629,
      "mean_token_accuracy": 0.1959345057606697,
      "num_tokens": 36349993.0,
      "step": 16280
    },
    {
      "entropy": 5.500933694839477,
      "epoch": 1.8289998315269276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004673509473033564,
      "loss": 5.327,
      "mean_token_accuracy": 0.19261912554502486,
      "num_tokens": 36361002.0,
      "step": 16285
    },
    {
      "entropy": 5.463829231262207,
      "epoch": 1.829561408434885,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046733011491749383,
      "loss": 5.2627,
      "mean_token_accuracy": 0.19682866632938384,
      "num_tokens": 36371688.0,
      "step": 16290
    },
    {
      "entropy": 5.455495691299438,
      "epoch": 1.8301229853428427,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004673092764077658,
      "loss": 5.1894,
      "mean_token_accuracy": 0.20718118399381638,
      "num_tokens": 36382683.0,
      "step": 16295
    },
    {
      "entropy": 5.526200532913208,
      "epoch": 1.8306845622508003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004672884317748359,
      "loss": 5.3446,
      "mean_token_accuracy": 0.18963032066822053,
      "num_tokens": 36394170.0,
      "step": 16300
    },
    {
      "entropy": 5.527466106414795,
      "epoch": 1.8312461391587578,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004672675810193677,
      "loss": 5.2913,
      "mean_token_accuracy": 0.19231263548135757,
      "num_tokens": 36405613.0,
      "step": 16305
    },
    {
      "entropy": 5.506656646728516,
      "epoch": 1.8318077160667152,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046724672414202514,
      "loss": 5.3249,
      "mean_token_accuracy": 0.1976760745048523,
      "num_tokens": 36416282.0,
      "step": 16310
    },
    {
      "entropy": 5.527156734466553,
      "epoch": 1.8323692929746729,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004672258611434724,
      "loss": 5.3443,
      "mean_token_accuracy": 0.18781709671020508,
      "num_tokens": 36427624.0,
      "step": 16315
    },
    {
      "entropy": 5.4442009925842285,
      "epoch": 1.8329308698826305,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046720499202437354,
      "loss": 5.1315,
      "mean_token_accuracy": 0.20778766721487046,
      "num_tokens": 36438230.0,
      "step": 16320
    },
    {
      "entropy": 5.486078929901123,
      "epoch": 1.833492446790588,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004671841167853933,
      "loss": 5.2705,
      "mean_token_accuracy": 0.19529445916414262,
      "num_tokens": 36448854.0,
      "step": 16325
    },
    {
      "entropy": 5.440536022186279,
      "epoch": 1.8340540236985454,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004671632354271962,
      "loss": 5.2764,
      "mean_token_accuracy": 0.1929825648665428,
      "num_tokens": 36460426.0,
      "step": 16330
    },
    {
      "entropy": 5.515605926513672,
      "epoch": 1.834615600606503,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046714234795044724,
      "loss": 5.2658,
      "mean_token_accuracy": 0.1934250235557556,
      "num_tokens": 36470914.0,
      "step": 16335
    },
    {
      "entropy": 5.434169721603394,
      "epoch": 1.8351771775144607,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004671214543558113,
      "loss": 5.1673,
      "mean_token_accuracy": 0.19939337074756622,
      "num_tokens": 36480753.0,
      "step": 16340
    },
    {
      "entropy": 5.380618000030518,
      "epoch": 1.8357387544224182,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004671005546439538,
      "loss": 5.1686,
      "mean_token_accuracy": 0.20086667835712432,
      "num_tokens": 36492858.0,
      "step": 16345
    },
    {
      "entropy": 5.355233383178711,
      "epoch": 1.8363003313303756,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046707964881554005,
      "loss": 5.1304,
      "mean_token_accuracy": 0.2012440785765648,
      "num_tokens": 36503912.0,
      "step": 16350
    },
    {
      "entropy": 5.408527183532715,
      "epoch": 1.8368619082383333,
      "grad_norm": 1.125,
      "learning_rate": 0.00046705873687123586,
      "loss": 5.2159,
      "mean_token_accuracy": 0.1951673373579979,
      "num_tokens": 36515410.0,
      "step": 16355
    },
    {
      "entropy": 5.460459232330322,
      "epoch": 1.837423485146291,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046703781881170703,
      "loss": 5.1603,
      "mean_token_accuracy": 0.20181371420621871,
      "num_tokens": 36526310.0,
      "step": 16360
    },
    {
      "entropy": 5.425773429870605,
      "epoch": 1.8379850620542482,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046701689463761947,
      "loss": 5.1158,
      "mean_token_accuracy": 0.2011399671435356,
      "num_tokens": 36537081.0,
      "step": 16365
    },
    {
      "entropy": 5.418057584762574,
      "epoch": 1.8385466389622058,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046699596434963953,
      "loss": 5.2127,
      "mean_token_accuracy": 0.19751242697238922,
      "num_tokens": 36549419.0,
      "step": 16370
    },
    {
      "entropy": 5.3919353008270265,
      "epoch": 1.8391082158701635,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004669750279484337,
      "loss": 5.2001,
      "mean_token_accuracy": 0.20440282970666884,
      "num_tokens": 36560842.0,
      "step": 16375
    },
    {
      "entropy": 5.434959363937378,
      "epoch": 1.839669792778121,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004669540854346685,
      "loss": 5.175,
      "mean_token_accuracy": 0.2019817352294922,
      "num_tokens": 36571717.0,
      "step": 16380
    },
    {
      "entropy": 5.455682373046875,
      "epoch": 1.8402313696860784,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046693313680901073,
      "loss": 5.209,
      "mean_token_accuracy": 0.19455764889717103,
      "num_tokens": 36581709.0,
      "step": 16385
    },
    {
      "entropy": 5.471096420288086,
      "epoch": 1.840792946594036,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046691218207212744,
      "loss": 5.2246,
      "mean_token_accuracy": 0.196385957300663,
      "num_tokens": 36593033.0,
      "step": 16390
    },
    {
      "entropy": 5.522565937042236,
      "epoch": 1.8413545235019937,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004668912212246859,
      "loss": 5.3249,
      "mean_token_accuracy": 0.193872332572937,
      "num_tokens": 36603605.0,
      "step": 16395
    },
    {
      "entropy": 5.505653142929077,
      "epoch": 1.8419161004099511,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046687025426735335,
      "loss": 5.3304,
      "mean_token_accuracy": 0.19150075018405915,
      "num_tokens": 36616219.0,
      "step": 16400
    },
    {
      "entropy": 5.552918195724487,
      "epoch": 1.8424776773179086,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004668492812007976,
      "loss": 5.2948,
      "mean_token_accuracy": 0.19474121928215027,
      "num_tokens": 36627777.0,
      "step": 16405
    },
    {
      "entropy": 5.484918260574341,
      "epoch": 1.8430392542258662,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004668283020256863,
      "loss": 5.2341,
      "mean_token_accuracy": 0.19648625403642656,
      "num_tokens": 36638792.0,
      "step": 16410
    },
    {
      "entropy": 5.479833507537842,
      "epoch": 1.843600831133824,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046680731674268744,
      "loss": 5.2655,
      "mean_token_accuracy": 0.19887581318616868,
      "num_tokens": 36650218.0,
      "step": 16415
    },
    {
      "entropy": 5.480641794204712,
      "epoch": 1.8441624080417813,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004667863253524692,
      "loss": 5.2285,
      "mean_token_accuracy": 0.1968456357717514,
      "num_tokens": 36660987.0,
      "step": 16420
    },
    {
      "entropy": 5.427464437484741,
      "epoch": 1.8447239849497388,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046676532785569995,
      "loss": 5.1976,
      "mean_token_accuracy": 0.19036560505628586,
      "num_tokens": 36671270.0,
      "step": 16425
    },
    {
      "entropy": 5.456794357299804,
      "epoch": 1.8452855618576964,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004667443242530484,
      "loss": 5.1999,
      "mean_token_accuracy": 0.2033836618065834,
      "num_tokens": 36682484.0,
      "step": 16430
    },
    {
      "entropy": 5.519985580444336,
      "epoch": 1.845847138765654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046672331454518314,
      "loss": 5.2992,
      "mean_token_accuracy": 0.19730817377567292,
      "num_tokens": 36693723.0,
      "step": 16435
    },
    {
      "entropy": 5.359445381164551,
      "epoch": 1.8464087156736115,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004667022987327732,
      "loss": 5.0972,
      "mean_token_accuracy": 0.2054307758808136,
      "num_tokens": 36704166.0,
      "step": 16440
    },
    {
      "entropy": 5.391143417358398,
      "epoch": 1.846970292581569,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004666812768164877,
      "loss": 5.1563,
      "mean_token_accuracy": 0.20609794855117797,
      "num_tokens": 36714139.0,
      "step": 16445
    },
    {
      "entropy": 5.44873685836792,
      "epoch": 1.8475318694895266,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046666024879699606,
      "loss": 5.1883,
      "mean_token_accuracy": 0.20303950160741807,
      "num_tokens": 36724273.0,
      "step": 16450
    },
    {
      "entropy": 5.462536144256592,
      "epoch": 1.8480934463974843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046663921467496766,
      "loss": 5.2619,
      "mean_token_accuracy": 0.1917540907859802,
      "num_tokens": 36735176.0,
      "step": 16455
    },
    {
      "entropy": 5.40107274055481,
      "epoch": 1.8486550233054415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004666181744510724,
      "loss": 5.2029,
      "mean_token_accuracy": 0.1995821252465248,
      "num_tokens": 36746170.0,
      "step": 16460
    },
    {
      "entropy": 5.551182985305786,
      "epoch": 1.8492166002133992,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046659712812598017,
      "loss": 5.3621,
      "mean_token_accuracy": 0.191077621281147,
      "num_tokens": 36756969.0,
      "step": 16465
    },
    {
      "entropy": 5.548001861572265,
      "epoch": 1.8497781771213568,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046657607570036103,
      "loss": 5.2211,
      "mean_token_accuracy": 0.20138726979494095,
      "num_tokens": 36767892.0,
      "step": 16470
    },
    {
      "entropy": 5.452150917053222,
      "epoch": 1.8503397540293143,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046655501717488533,
      "loss": 5.2244,
      "mean_token_accuracy": 0.19823097884655,
      "num_tokens": 36778967.0,
      "step": 16475
    },
    {
      "entropy": 5.337763929367066,
      "epoch": 1.8509013309372717,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004665339525502236,
      "loss": 5.0952,
      "mean_token_accuracy": 0.199875046312809,
      "num_tokens": 36789204.0,
      "step": 16480
    },
    {
      "entropy": 5.4093445301055905,
      "epoch": 1.8514629078452294,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046651288182704653,
      "loss": 5.1592,
      "mean_token_accuracy": 0.2065848022699356,
      "num_tokens": 36798990.0,
      "step": 16485
    },
    {
      "entropy": 5.511639404296875,
      "epoch": 1.852024484753187,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004664918050060251,
      "loss": 5.2691,
      "mean_token_accuracy": 0.19570061564445496,
      "num_tokens": 36809988.0,
      "step": 16490
    },
    {
      "entropy": 5.433776426315307,
      "epoch": 1.8525860616611445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004664707220878302,
      "loss": 5.1953,
      "mean_token_accuracy": 0.20575449764728546,
      "num_tokens": 36821579.0,
      "step": 16495
    },
    {
      "entropy": 5.534322500228882,
      "epoch": 1.853147638569102,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004664496330731334,
      "loss": 5.3331,
      "mean_token_accuracy": 0.19465134739875795,
      "num_tokens": 36833141.0,
      "step": 16500
    },
    {
      "entropy": 5.5263941287994385,
      "epoch": 1.8537092154770596,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004664285379626059,
      "loss": 5.3276,
      "mean_token_accuracy": 0.19896939396858215,
      "num_tokens": 36843347.0,
      "step": 16505
    },
    {
      "entropy": 5.539216756820679,
      "epoch": 1.8542707923850172,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046640743675691954,
      "loss": 5.3526,
      "mean_token_accuracy": 0.18484846949577333,
      "num_tokens": 36854459.0,
      "step": 16510
    },
    {
      "entropy": 5.400459146499633,
      "epoch": 1.8548323692929747,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004663863294567462,
      "loss": 5.0778,
      "mean_token_accuracy": 0.1987343817949295,
      "num_tokens": 36864761.0,
      "step": 16515
    },
    {
      "entropy": 5.580765295028686,
      "epoch": 1.8553939462009321,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004663652160627579,
      "loss": 5.3756,
      "mean_token_accuracy": 0.19152718931436538,
      "num_tokens": 36875413.0,
      "step": 16520
    },
    {
      "entropy": 5.457789564132691,
      "epoch": 1.8559555231088898,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046634409657562683,
      "loss": 5.2458,
      "mean_token_accuracy": 0.19583977311849593,
      "num_tokens": 36885350.0,
      "step": 16525
    },
    {
      "entropy": 5.453854990005493,
      "epoch": 1.8565171000168474,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004663229709960255,
      "loss": 5.1563,
      "mean_token_accuracy": 0.2017476513981819,
      "num_tokens": 36895355.0,
      "step": 16530
    },
    {
      "entropy": 5.464024496078491,
      "epoch": 1.8570786769248049,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046630183932462664,
      "loss": 5.242,
      "mean_token_accuracy": 0.20229275822639464,
      "num_tokens": 36907454.0,
      "step": 16535
    },
    {
      "entropy": 5.442661333084106,
      "epoch": 1.8576402538327623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004662807015621031,
      "loss": 5.2593,
      "mean_token_accuracy": 0.19481302052736282,
      "num_tokens": 36919199.0,
      "step": 16540
    },
    {
      "entropy": 5.563452243804932,
      "epoch": 1.85820183074072,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046625955770912764,
      "loss": 5.3332,
      "mean_token_accuracy": 0.19176400154829026,
      "num_tokens": 36930499.0,
      "step": 16545
    },
    {
      "entropy": 5.501461410522461,
      "epoch": 1.8587634076486776,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046623840776637384,
      "loss": 5.2615,
      "mean_token_accuracy": 0.19176605194807053,
      "num_tokens": 36943082.0,
      "step": 16550
    },
    {
      "entropy": 5.422438716888427,
      "epoch": 1.859324984556635,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004662172517345149,
      "loss": 5.1683,
      "mean_token_accuracy": 0.1977447509765625,
      "num_tokens": 36954591.0,
      "step": 16555
    },
    {
      "entropy": 5.397451639175415,
      "epoch": 1.8598865614645925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004661960896142245,
      "loss": 5.2676,
      "mean_token_accuracy": 0.19991008788347245,
      "num_tokens": 36967028.0,
      "step": 16560
    },
    {
      "entropy": 5.463958406448365,
      "epoch": 1.8604481383725502,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046617492140617654,
      "loss": 5.2945,
      "mean_token_accuracy": 0.19127349704504013,
      "num_tokens": 36978716.0,
      "step": 16565
    },
    {
      "entropy": 5.4492768287658695,
      "epoch": 1.8610097152805076,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004661537471110448,
      "loss": 5.1756,
      "mean_token_accuracy": 0.19747342467308043,
      "num_tokens": 36989416.0,
      "step": 16570
    },
    {
      "entropy": 5.459396409988403,
      "epoch": 1.861571292188465,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004661325667295037,
      "loss": 5.2306,
      "mean_token_accuracy": 0.19828828424215317,
      "num_tokens": 37002197.0,
      "step": 16575
    },
    {
      "entropy": 5.610602617263794,
      "epoch": 1.8621328690964227,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004661113802622275,
      "loss": 5.3399,
      "mean_token_accuracy": 0.18506837785243987,
      "num_tokens": 37012977.0,
      "step": 16580
    },
    {
      "entropy": 5.487373018264771,
      "epoch": 1.8626944460043804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004660901877098908,
      "loss": 5.2507,
      "mean_token_accuracy": 0.19659968614578247,
      "num_tokens": 37024141.0,
      "step": 16585
    },
    {
      "entropy": 5.402905750274658,
      "epoch": 1.8632560229123378,
      "grad_norm": 1.125,
      "learning_rate": 0.0004660689890731685,
      "loss": 5.1594,
      "mean_token_accuracy": 0.20519924610853196,
      "num_tokens": 37034471.0,
      "step": 16590
    },
    {
      "entropy": 5.420420789718628,
      "epoch": 1.8638175998202953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046604778435273546,
      "loss": 5.2434,
      "mean_token_accuracy": 0.19732611179351806,
      "num_tokens": 37046375.0,
      "step": 16595
    },
    {
      "entropy": 5.470990324020386,
      "epoch": 1.864379176728253,
      "grad_norm": 1.125,
      "learning_rate": 0.00046602657354926686,
      "loss": 5.2569,
      "mean_token_accuracy": 0.20421495139598847,
      "num_tokens": 37056617.0,
      "step": 16600
    },
    {
      "entropy": 5.387459993362427,
      "epoch": 1.8649407536362106,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000466005356663438,
      "loss": 5.088,
      "mean_token_accuracy": 0.20210430324077605,
      "num_tokens": 37066477.0,
      "step": 16605
    },
    {
      "entropy": 5.425457239151001,
      "epoch": 1.865502330544168,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046598413369592463,
      "loss": 5.2596,
      "mean_token_accuracy": 0.19388592392206191,
      "num_tokens": 37077150.0,
      "step": 16610
    },
    {
      "entropy": 5.525846862792969,
      "epoch": 1.8660639074521255,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004659629046474023,
      "loss": 5.3116,
      "mean_token_accuracy": 0.19255971312522888,
      "num_tokens": 37089961.0,
      "step": 16615
    },
    {
      "entropy": 5.5383423328399655,
      "epoch": 1.8666254843600831,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000465941669518547,
      "loss": 5.3092,
      "mean_token_accuracy": 0.1939225673675537,
      "num_tokens": 37100779.0,
      "step": 16620
    },
    {
      "entropy": 5.440025424957275,
      "epoch": 1.8671870612680408,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004659204283100349,
      "loss": 5.1586,
      "mean_token_accuracy": 0.2000011220574379,
      "num_tokens": 37111293.0,
      "step": 16625
    },
    {
      "entropy": 5.478985166549682,
      "epoch": 1.8677486381759982,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046589918102254237,
      "loss": 5.2768,
      "mean_token_accuracy": 0.1982982188463211,
      "num_tokens": 37123087.0,
      "step": 16630
    },
    {
      "entropy": 5.448623752593994,
      "epoch": 1.8683102150839557,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004658779276567458,
      "loss": 5.2451,
      "mean_token_accuracy": 0.19441010802984238,
      "num_tokens": 37135227.0,
      "step": 16635
    },
    {
      "entropy": 5.439109516143799,
      "epoch": 1.8688717919919133,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046585666821332203,
      "loss": 5.2171,
      "mean_token_accuracy": 0.19959462881088258,
      "num_tokens": 37146970.0,
      "step": 16640
    },
    {
      "entropy": 5.500585651397705,
      "epoch": 1.869433368899871,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046583540269294794,
      "loss": 5.2133,
      "mean_token_accuracy": 0.19257651418447494,
      "num_tokens": 37158144.0,
      "step": 16645
    },
    {
      "entropy": 5.508302307128906,
      "epoch": 1.8699949458078284,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004658141310963006,
      "loss": 5.3449,
      "mean_token_accuracy": 0.1924276128411293,
      "num_tokens": 37169265.0,
      "step": 16650
    },
    {
      "entropy": 5.515935468673706,
      "epoch": 1.8705565227157859,
      "grad_norm": 1.125,
      "learning_rate": 0.0004657928534240573,
      "loss": 5.2926,
      "mean_token_accuracy": 0.1949426531791687,
      "num_tokens": 37179186.0,
      "step": 16655
    },
    {
      "entropy": 5.492211532592774,
      "epoch": 1.8711180996237435,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004657715696768955,
      "loss": 5.222,
      "mean_token_accuracy": 0.20232529193162918,
      "num_tokens": 37188906.0,
      "step": 16660
    },
    {
      "entropy": 5.365943431854248,
      "epoch": 1.871679676531701,
      "grad_norm": 1.0,
      "learning_rate": 0.000465750279855493,
      "loss": 5.1551,
      "mean_token_accuracy": 0.20616481304168702,
      "num_tokens": 37200879.0,
      "step": 16665
    },
    {
      "entropy": 5.434335803985595,
      "epoch": 1.8722412534396584,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004657289839605276,
      "loss": 5.1712,
      "mean_token_accuracy": 0.2032892018556595,
      "num_tokens": 37211648.0,
      "step": 16670
    },
    {
      "entropy": 5.58739857673645,
      "epoch": 1.872802830347616,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004657076819926774,
      "loss": 5.3711,
      "mean_token_accuracy": 0.19582848697900773,
      "num_tokens": 37223164.0,
      "step": 16675
    },
    {
      "entropy": 5.509513425827026,
      "epoch": 1.8733644072555737,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004656863739526206,
      "loss": 5.2282,
      "mean_token_accuracy": 0.19520914256572724,
      "num_tokens": 37233860.0,
      "step": 16680
    },
    {
      "entropy": 5.524698543548584,
      "epoch": 1.8739259841635312,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046566505984103575,
      "loss": 5.1824,
      "mean_token_accuracy": 0.1982952684164047,
      "num_tokens": 37245431.0,
      "step": 16685
    },
    {
      "entropy": 5.493672466278076,
      "epoch": 1.8744875610714886,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004656437396586014,
      "loss": 5.3235,
      "mean_token_accuracy": 0.19340966492891312,
      "num_tokens": 37256394.0,
      "step": 16690
    },
    {
      "entropy": 5.460845470428467,
      "epoch": 1.8750491379794463,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004656224134059964,
      "loss": 5.2901,
      "mean_token_accuracy": 0.19576283246278764,
      "num_tokens": 37268367.0,
      "step": 16695
    },
    {
      "entropy": 5.4677431106567385,
      "epoch": 1.875610714887404,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046560108108389985,
      "loss": 5.263,
      "mean_token_accuracy": 0.1967165470123291,
      "num_tokens": 37279019.0,
      "step": 16700
    },
    {
      "entropy": 5.541481828689575,
      "epoch": 1.8761722917953614,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000465579742692991,
      "loss": 5.2057,
      "mean_token_accuracy": 0.19804907441139222,
      "num_tokens": 37289806.0,
      "step": 16705
    },
    {
      "entropy": 5.390263175964355,
      "epoch": 1.8767338687033188,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004655583982339492,
      "loss": 5.2392,
      "mean_token_accuracy": 0.19549369364976882,
      "num_tokens": 37300607.0,
      "step": 16710
    },
    {
      "entropy": 5.443153142929077,
      "epoch": 1.8772954456112765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004655370477074541,
      "loss": 5.2956,
      "mean_token_accuracy": 0.19506974667310714,
      "num_tokens": 37311966.0,
      "step": 16715
    },
    {
      "entropy": 5.396130084991455,
      "epoch": 1.8778570225192341,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004655156911141855,
      "loss": 5.2242,
      "mean_token_accuracy": 0.1985211431980133,
      "num_tokens": 37324459.0,
      "step": 16720
    },
    {
      "entropy": 5.469950866699219,
      "epoch": 1.8784185994271916,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004654943284548234,
      "loss": 5.2292,
      "mean_token_accuracy": 0.196366611123085,
      "num_tokens": 37335276.0,
      "step": 16725
    },
    {
      "entropy": 5.4683925151824955,
      "epoch": 1.878980176335149,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046547295973004807,
      "loss": 5.1467,
      "mean_token_accuracy": 0.20398791283369064,
      "num_tokens": 37347274.0,
      "step": 16730
    },
    {
      "entropy": 5.460577440261841,
      "epoch": 1.8795417532431067,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046545158494053966,
      "loss": 5.2755,
      "mean_token_accuracy": 0.1909830614924431,
      "num_tokens": 37359789.0,
      "step": 16735
    },
    {
      "entropy": 5.435467672348023,
      "epoch": 1.8801033301510643,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046543020408697905,
      "loss": 5.216,
      "mean_token_accuracy": 0.20039468854665757,
      "num_tokens": 37369851.0,
      "step": 16740
    },
    {
      "entropy": 5.545345497131348,
      "epoch": 1.8806649070590218,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046540881717004685,
      "loss": 5.3759,
      "mean_token_accuracy": 0.1927112430334091,
      "num_tokens": 37381008.0,
      "step": 16745
    },
    {
      "entropy": 5.466767311096191,
      "epoch": 1.8812264839669792,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004653874241904241,
      "loss": 5.2072,
      "mean_token_accuracy": 0.1926458492875099,
      "num_tokens": 37392110.0,
      "step": 16750
    },
    {
      "entropy": 5.517423677444458,
      "epoch": 1.8817880608749369,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004653660251487919,
      "loss": 5.2452,
      "mean_token_accuracy": 0.2020352378487587,
      "num_tokens": 37402995.0,
      "step": 16755
    },
    {
      "entropy": 5.487637567520141,
      "epoch": 1.8823496377828943,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046534462004583154,
      "loss": 5.2272,
      "mean_token_accuracy": 0.2001844674348831,
      "num_tokens": 37415072.0,
      "step": 16760
    },
    {
      "entropy": 5.511497974395752,
      "epoch": 1.8829112146908518,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004653232088822248,
      "loss": 5.3984,
      "mean_token_accuracy": 0.1877359688282013,
      "num_tokens": 37427375.0,
      "step": 16765
    },
    {
      "entropy": 5.407660722732544,
      "epoch": 1.8834727915988094,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004653017916586532,
      "loss": 5.1757,
      "mean_token_accuracy": 0.20452356338500977,
      "num_tokens": 37436829.0,
      "step": 16770
    },
    {
      "entropy": 5.523489952087402,
      "epoch": 1.884034368506767,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004652803683757987,
      "loss": 5.2843,
      "mean_token_accuracy": 0.18755460083484649,
      "num_tokens": 37447875.0,
      "step": 16775
    },
    {
      "entropy": 5.392043781280518,
      "epoch": 1.8845959454147245,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004652589390343436,
      "loss": 5.1482,
      "mean_token_accuracy": 0.19770222753286362,
      "num_tokens": 37458408.0,
      "step": 16780
    },
    {
      "entropy": 5.402614879608154,
      "epoch": 1.885157522322682,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046523750363497,
      "loss": 5.2147,
      "mean_token_accuracy": 0.19525437653064728,
      "num_tokens": 37468437.0,
      "step": 16785
    },
    {
      "entropy": 5.562240648269653,
      "epoch": 1.8857190992306396,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004652160621783605,
      "loss": 5.3083,
      "mean_token_accuracy": 0.19734027981758118,
      "num_tokens": 37479359.0,
      "step": 16790
    },
    {
      "entropy": 5.472897243499756,
      "epoch": 1.8862806761385973,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046519461466519785,
      "loss": 5.2345,
      "mean_token_accuracy": 0.20217179656028747,
      "num_tokens": 37491321.0,
      "step": 16795
    },
    {
      "entropy": 5.463063144683838,
      "epoch": 1.8868422530465547,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004651731610961649,
      "loss": 5.2484,
      "mean_token_accuracy": 0.19329507052898406,
      "num_tokens": 37501828.0,
      "step": 16800
    },
    {
      "entropy": 5.570747756958008,
      "epoch": 1.8874038299545122,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004651517014719446,
      "loss": 5.386,
      "mean_token_accuracy": 0.18740827441215516,
      "num_tokens": 37514525.0,
      "step": 16805
    },
    {
      "entropy": 5.503907585144043,
      "epoch": 1.8879654068624698,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004651302357932206,
      "loss": 5.1532,
      "mean_token_accuracy": 0.19461924135684966,
      "num_tokens": 37525588.0,
      "step": 16810
    },
    {
      "entropy": 5.43377423286438,
      "epoch": 1.8885269837704275,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000465108764060676,
      "loss": 5.2276,
      "mean_token_accuracy": 0.1948937341570854,
      "num_tokens": 37537941.0,
      "step": 16815
    },
    {
      "entropy": 5.450218152999878,
      "epoch": 1.889088560678385,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046508728627499464,
      "loss": 5.2439,
      "mean_token_accuracy": 0.19229685813188552,
      "num_tokens": 37549663.0,
      "step": 16820
    },
    {
      "entropy": 5.524704837799073,
      "epoch": 1.8896501375863424,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004650658024368604,
      "loss": 5.3005,
      "mean_token_accuracy": 0.19765808284282685,
      "num_tokens": 37559793.0,
      "step": 16825
    },
    {
      "entropy": 5.431095838546753,
      "epoch": 1.8902117144943,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004650443125469573,
      "loss": 5.1753,
      "mean_token_accuracy": 0.1964648738503456,
      "num_tokens": 37570203.0,
      "step": 16830
    },
    {
      "entropy": 5.414950895309448,
      "epoch": 1.8907732914022577,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004650228166059695,
      "loss": 5.2703,
      "mean_token_accuracy": 0.19715822339057923,
      "num_tokens": 37581858.0,
      "step": 16835
    },
    {
      "entropy": 5.457394742965699,
      "epoch": 1.8913348683102151,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004650013146145816,
      "loss": 5.2163,
      "mean_token_accuracy": 0.19874331653118132,
      "num_tokens": 37592311.0,
      "step": 16840
    },
    {
      "entropy": 5.516688299179077,
      "epoch": 1.8918964452181726,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004649798065734781,
      "loss": 5.3581,
      "mean_token_accuracy": 0.18968896567821503,
      "num_tokens": 37604721.0,
      "step": 16845
    },
    {
      "entropy": 5.580503273010254,
      "epoch": 1.8924580221261302,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046495829248334393,
      "loss": 5.4533,
      "mean_token_accuracy": 0.1888548105955124,
      "num_tokens": 37616044.0,
      "step": 16850
    },
    {
      "entropy": 5.508128023147583,
      "epoch": 1.8930195990340877,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004649367723448641,
      "loss": 5.3023,
      "mean_token_accuracy": 0.19902894347906114,
      "num_tokens": 37626501.0,
      "step": 16855
    },
    {
      "entropy": 5.539801597595215,
      "epoch": 1.8935811759420451,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046491524615872366,
      "loss": 5.2702,
      "mean_token_accuracy": 0.1941537395119667,
      "num_tokens": 37637861.0,
      "step": 16860
    },
    {
      "entropy": 5.413544797897339,
      "epoch": 1.8941427528500028,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046489371392560816,
      "loss": 5.1192,
      "mean_token_accuracy": 0.20064548254013062,
      "num_tokens": 37649288.0,
      "step": 16865
    },
    {
      "entropy": 5.380209016799927,
      "epoch": 1.8947043297579604,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046487217564620317,
      "loss": 5.1395,
      "mean_token_accuracy": 0.20221962481737138,
      "num_tokens": 37659700.0,
      "step": 16870
    },
    {
      "entropy": 5.473927736282349,
      "epoch": 1.8952659066659179,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004648506313211945,
      "loss": 5.2907,
      "mean_token_accuracy": 0.1938436284661293,
      "num_tokens": 37671739.0,
      "step": 16875
    },
    {
      "entropy": 5.474382638931274,
      "epoch": 1.8958274835738753,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046482908095126803,
      "loss": 5.1796,
      "mean_token_accuracy": 0.20720221996307372,
      "num_tokens": 37682250.0,
      "step": 16880
    },
    {
      "entropy": 5.454179382324218,
      "epoch": 1.896389060481833,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046480752453711,
      "loss": 5.2504,
      "mean_token_accuracy": 0.20065705180168153,
      "num_tokens": 37694052.0,
      "step": 16885
    },
    {
      "entropy": 5.539553689956665,
      "epoch": 1.8969506373897906,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004647859620794068,
      "loss": 5.3349,
      "mean_token_accuracy": 0.1960662394762039,
      "num_tokens": 37704818.0,
      "step": 16890
    },
    {
      "entropy": 5.598157691955566,
      "epoch": 1.897512214297748,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046476439357884497,
      "loss": 5.3799,
      "mean_token_accuracy": 0.18657071888446808,
      "num_tokens": 37717679.0,
      "step": 16895
    },
    {
      "entropy": 5.439261960983276,
      "epoch": 1.8980737912057055,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046474281903611126,
      "loss": 5.2011,
      "mean_token_accuracy": 0.19970800578594208,
      "num_tokens": 37728711.0,
      "step": 16900
    },
    {
      "entropy": 5.453038501739502,
      "epoch": 1.8986353681136632,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046472123845189253,
      "loss": 5.214,
      "mean_token_accuracy": 0.19857898205518723,
      "num_tokens": 37740082.0,
      "step": 16905
    },
    {
      "entropy": 5.524498796463012,
      "epoch": 1.8991969450216208,
      "grad_norm": 1.171875,
      "learning_rate": 0.000464699651826876,
      "loss": 5.3461,
      "mean_token_accuracy": 0.18871774524450302,
      "num_tokens": 37750715.0,
      "step": 16910
    },
    {
      "entropy": 5.52142186164856,
      "epoch": 1.8997585219295783,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046467805916174907,
      "loss": 5.3137,
      "mean_token_accuracy": 0.1888110488653183,
      "num_tokens": 37763230.0,
      "step": 16915
    },
    {
      "entropy": 5.4834840297698975,
      "epoch": 1.9003200988375357,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046465646045719915,
      "loss": 5.2002,
      "mean_token_accuracy": 0.19617471992969512,
      "num_tokens": 37773621.0,
      "step": 16920
    },
    {
      "entropy": 5.484920692443848,
      "epoch": 1.9008816757454934,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004646348557139139,
      "loss": 5.2721,
      "mean_token_accuracy": 0.19421189576387404,
      "num_tokens": 37785068.0,
      "step": 16925
    },
    {
      "entropy": 5.435989189147949,
      "epoch": 1.901443252653451,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046461324493258133,
      "loss": 5.2612,
      "mean_token_accuracy": 0.2018831804394722,
      "num_tokens": 37796925.0,
      "step": 16930
    },
    {
      "entropy": 5.4933570384979244,
      "epoch": 1.9020048295614085,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046459162811388955,
      "loss": 5.2847,
      "mean_token_accuracy": 0.1948854923248291,
      "num_tokens": 37808291.0,
      "step": 16935
    },
    {
      "entropy": 5.4373047828674315,
      "epoch": 1.902566406469366,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004645700052585267,
      "loss": 5.1193,
      "mean_token_accuracy": 0.21289151310920715,
      "num_tokens": 37819337.0,
      "step": 16940
    },
    {
      "entropy": 5.52640380859375,
      "epoch": 1.9031279833773236,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004645483763671815,
      "loss": 5.3188,
      "mean_token_accuracy": 0.19600376933813096,
      "num_tokens": 37831832.0,
      "step": 16945
    },
    {
      "entropy": 5.4921496391296385,
      "epoch": 1.903689560285281,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046452674144054236,
      "loss": 5.2322,
      "mean_token_accuracy": 0.19636641293764115,
      "num_tokens": 37842586.0,
      "step": 16950
    },
    {
      "entropy": 5.435227298736573,
      "epoch": 1.9042511371932385,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046450510047929826,
      "loss": 5.1579,
      "mean_token_accuracy": 0.2063504546880722,
      "num_tokens": 37852198.0,
      "step": 16955
    },
    {
      "entropy": 5.446521186828614,
      "epoch": 1.9048127141011961,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046448345348413833,
      "loss": 5.2065,
      "mean_token_accuracy": 0.1997394159436226,
      "num_tokens": 37862967.0,
      "step": 16960
    },
    {
      "entropy": 5.446404647827149,
      "epoch": 1.9053742910091538,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004644618004557517,
      "loss": 5.3262,
      "mean_token_accuracy": 0.187962144613266,
      "num_tokens": 37874501.0,
      "step": 16965
    },
    {
      "entropy": 5.452405071258545,
      "epoch": 1.9059358679171112,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004644401413948279,
      "loss": 5.0901,
      "mean_token_accuracy": 0.20705420672893524,
      "num_tokens": 37886404.0,
      "step": 16970
    },
    {
      "entropy": 5.553581809997558,
      "epoch": 1.9064974448250687,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004644184763020565,
      "loss": 5.286,
      "mean_token_accuracy": 0.195398685336113,
      "num_tokens": 37897344.0,
      "step": 16975
    },
    {
      "entropy": 5.379161643981933,
      "epoch": 1.9070590217330263,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004643968051781273,
      "loss": 5.152,
      "mean_token_accuracy": 0.20995626151561736,
      "num_tokens": 37909226.0,
      "step": 16980
    },
    {
      "entropy": 5.484614324569702,
      "epoch": 1.907620598640984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004643751280237305,
      "loss": 5.2976,
      "mean_token_accuracy": 0.19569606482982635,
      "num_tokens": 37920578.0,
      "step": 16985
    },
    {
      "entropy": 5.516542768478393,
      "epoch": 1.9081821755489414,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004643534448395561,
      "loss": 5.1747,
      "mean_token_accuracy": 0.1942470669746399,
      "num_tokens": 37931814.0,
      "step": 16990
    },
    {
      "entropy": 5.460321283340454,
      "epoch": 1.9087437524568989,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004643317556262946,
      "loss": 5.2201,
      "mean_token_accuracy": 0.19769106805324554,
      "num_tokens": 37942345.0,
      "step": 16995
    },
    {
      "entropy": 5.3960329532623295,
      "epoch": 1.9093053293648565,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046431006038463653,
      "loss": 5.1294,
      "mean_token_accuracy": 0.20145775228738785,
      "num_tokens": 37953027.0,
      "step": 17000
    },
    {
      "entropy": 5.474249935150146,
      "epoch": 1.9098669062728142,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004642883591152727,
      "loss": 5.2799,
      "mean_token_accuracy": 0.19447172284126282,
      "num_tokens": 37963290.0,
      "step": 17005
    },
    {
      "entropy": 5.499614906311035,
      "epoch": 1.9104284831807716,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004642666518188942,
      "loss": 5.3712,
      "mean_token_accuracy": 0.18781432658433914,
      "num_tokens": 37974424.0,
      "step": 17010
    },
    {
      "entropy": 5.598958158493042,
      "epoch": 1.910990060088729,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046424493849619205,
      "loss": 5.365,
      "mean_token_accuracy": 0.19186062961816788,
      "num_tokens": 37987993.0,
      "step": 17015
    },
    {
      "entropy": 5.462058448791504,
      "epoch": 1.9115516369966867,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046422321914785757,
      "loss": 5.1461,
      "mean_token_accuracy": 0.2007363885641098,
      "num_tokens": 37999340.0,
      "step": 17020
    },
    {
      "entropy": 5.416390752792358,
      "epoch": 1.9121132139046444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046420149377458255,
      "loss": 5.212,
      "mean_token_accuracy": 0.1935009092092514,
      "num_tokens": 38009737.0,
      "step": 17025
    },
    {
      "entropy": 5.463556098937988,
      "epoch": 1.9126747908126018,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046417976237705846,
      "loss": 5.209,
      "mean_token_accuracy": 0.19952737987041474,
      "num_tokens": 38021283.0,
      "step": 17030
    },
    {
      "entropy": 5.575936651229858,
      "epoch": 1.9132363677205593,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046415802495597744,
      "loss": 5.3492,
      "mean_token_accuracy": 0.19020410180091857,
      "num_tokens": 38034044.0,
      "step": 17035
    },
    {
      "entropy": 5.480965566635132,
      "epoch": 1.913797944628517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046413628151203154,
      "loss": 5.1929,
      "mean_token_accuracy": 0.20085353553295135,
      "num_tokens": 38044585.0,
      "step": 17040
    },
    {
      "entropy": 5.405561113357544,
      "epoch": 1.9143595215364744,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046411453204591303,
      "loss": 5.1526,
      "mean_token_accuracy": 0.1980004131793976,
      "num_tokens": 38055771.0,
      "step": 17045
    },
    {
      "entropy": 5.435069799423218,
      "epoch": 1.9149210984444318,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046409277655831455,
      "loss": 5.2282,
      "mean_token_accuracy": 0.1987907186150551,
      "num_tokens": 38066344.0,
      "step": 17050
    },
    {
      "entropy": 5.508805561065674,
      "epoch": 1.9154826753523895,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004640710150499287,
      "loss": 5.1705,
      "mean_token_accuracy": 0.20338296592235566,
      "num_tokens": 38077785.0,
      "step": 17055
    },
    {
      "entropy": 5.435651159286499,
      "epoch": 1.9160442522603471,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046404924752144834,
      "loss": 5.2865,
      "mean_token_accuracy": 0.19283691346645354,
      "num_tokens": 38089023.0,
      "step": 17060
    },
    {
      "entropy": 5.468222332000733,
      "epoch": 1.9166058291683046,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046402747397356666,
      "loss": 5.1455,
      "mean_token_accuracy": 0.2024591252207756,
      "num_tokens": 38099428.0,
      "step": 17065
    },
    {
      "entropy": 5.433711338043213,
      "epoch": 1.917167406076262,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004640056944069768,
      "loss": 5.1586,
      "mean_token_accuracy": 0.19864461421966553,
      "num_tokens": 38110757.0,
      "step": 17070
    },
    {
      "entropy": 5.478124761581421,
      "epoch": 1.9177289829842197,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004639839088223725,
      "loss": 5.2399,
      "mean_token_accuracy": 0.19660273492336272,
      "num_tokens": 38121442.0,
      "step": 17075
    },
    {
      "entropy": 5.4433839321136475,
      "epoch": 1.9182905598921773,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046396211722044716,
      "loss": 5.2251,
      "mean_token_accuracy": 0.20031900107860565,
      "num_tokens": 38132917.0,
      "step": 17080
    },
    {
      "entropy": 5.401716613769532,
      "epoch": 1.9188521368001348,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046394031960189473,
      "loss": 5.1489,
      "mean_token_accuracy": 0.19604706019163132,
      "num_tokens": 38143188.0,
      "step": 17085
    },
    {
      "entropy": 5.565263891220093,
      "epoch": 1.9194137137080922,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004639185159674093,
      "loss": 5.3361,
      "mean_token_accuracy": 0.19754895865917205,
      "num_tokens": 38154300.0,
      "step": 17090
    },
    {
      "entropy": 5.53109393119812,
      "epoch": 1.9199752906160499,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046389670631768494,
      "loss": 5.3315,
      "mean_token_accuracy": 0.19322279542684556,
      "num_tokens": 38165697.0,
      "step": 17095
    },
    {
      "entropy": 5.479623174667358,
      "epoch": 1.9205368675240075,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004638748906534162,
      "loss": 5.2268,
      "mean_token_accuracy": 0.1981937736272812,
      "num_tokens": 38176601.0,
      "step": 17100
    },
    {
      "entropy": 5.494045972824097,
      "epoch": 1.921098444431965,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004638530689752977,
      "loss": 5.3215,
      "mean_token_accuracy": 0.19902921169996263,
      "num_tokens": 38188334.0,
      "step": 17105
    },
    {
      "entropy": 5.507667446136475,
      "epoch": 1.9216600213399224,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046383124128402433,
      "loss": 5.2507,
      "mean_token_accuracy": 0.19804709404706955,
      "num_tokens": 38199483.0,
      "step": 17110
    },
    {
      "entropy": 5.524251270294189,
      "epoch": 1.92222159824788,
      "grad_norm": 1.125,
      "learning_rate": 0.00046380940758029095,
      "loss": 5.3363,
      "mean_token_accuracy": 0.19040213227272035,
      "num_tokens": 38211391.0,
      "step": 17115
    },
    {
      "entropy": 5.4890039443969725,
      "epoch": 1.9227831751558377,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046378756786479274,
      "loss": 5.2241,
      "mean_token_accuracy": 0.1998496562242508,
      "num_tokens": 38223415.0,
      "step": 17120
    },
    {
      "entropy": 5.4596755027771,
      "epoch": 1.9233447520637952,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004637657221382252,
      "loss": 5.1803,
      "mean_token_accuracy": 0.20113856196403504,
      "num_tokens": 38235093.0,
      "step": 17125
    },
    {
      "entropy": 5.463783931732178,
      "epoch": 1.9239063289717526,
      "grad_norm": 1.125,
      "learning_rate": 0.00046374387040128384,
      "loss": 5.2996,
      "mean_token_accuracy": 0.187471242249012,
      "num_tokens": 38247192.0,
      "step": 17130
    },
    {
      "entropy": 5.4878153800964355,
      "epoch": 1.9244679058797103,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046372201265466444,
      "loss": 5.282,
      "mean_token_accuracy": 0.19151819348335267,
      "num_tokens": 38257965.0,
      "step": 17135
    },
    {
      "entropy": 5.580102920532227,
      "epoch": 1.9250294827876677,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000463700148899063,
      "loss": 5.3683,
      "mean_token_accuracy": 0.18687620460987092,
      "num_tokens": 38268529.0,
      "step": 17140
    },
    {
      "entropy": 5.463663864135742,
      "epoch": 1.9255910596956252,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004636782791351756,
      "loss": 5.2356,
      "mean_token_accuracy": 0.1919669434428215,
      "num_tokens": 38279480.0,
      "step": 17145
    },
    {
      "entropy": 5.488632726669311,
      "epoch": 1.9261526366035828,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004636564033636987,
      "loss": 5.3035,
      "mean_token_accuracy": 0.1978561446070671,
      "num_tokens": 38291670.0,
      "step": 17150
    },
    {
      "entropy": 5.405059671401977,
      "epoch": 1.9267142135115405,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046363452158532857,
      "loss": 5.0849,
      "mean_token_accuracy": 0.21451079249382018,
      "num_tokens": 38303073.0,
      "step": 17155
    },
    {
      "entropy": 5.4871845722198485,
      "epoch": 1.927275790419498,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004636126338007622,
      "loss": 5.2504,
      "mean_token_accuracy": 0.19328262209892272,
      "num_tokens": 38313915.0,
      "step": 17160
    },
    {
      "entropy": 5.464030742645264,
      "epoch": 1.9278373673274554,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004635907400106965,
      "loss": 5.2125,
      "mean_token_accuracy": 0.1906324401497841,
      "num_tokens": 38324845.0,
      "step": 17165
    },
    {
      "entropy": 5.467322254180909,
      "epoch": 1.928398944235413,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004635688402158285,
      "loss": 5.249,
      "mean_token_accuracy": 0.1999796837568283,
      "num_tokens": 38335096.0,
      "step": 17170
    },
    {
      "entropy": 5.500621318817139,
      "epoch": 1.9289605211433707,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046354693441685537,
      "loss": 5.2263,
      "mean_token_accuracy": 0.19897073209285737,
      "num_tokens": 38345559.0,
      "step": 17175
    },
    {
      "entropy": 5.422845697402954,
      "epoch": 1.9295220980513281,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046352502261447475,
      "loss": 5.1452,
      "mean_token_accuracy": 0.21408585608005523,
      "num_tokens": 38356318.0,
      "step": 17180
    },
    {
      "entropy": 5.420642328262329,
      "epoch": 1.9300836749592856,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004635031048093843,
      "loss": 5.2127,
      "mean_token_accuracy": 0.1991479814052582,
      "num_tokens": 38367246.0,
      "step": 17185
    },
    {
      "entropy": 5.432358312606811,
      "epoch": 1.9306452518672432,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004634811810022819,
      "loss": 5.2324,
      "mean_token_accuracy": 0.19524916559457778,
      "num_tokens": 38378767.0,
      "step": 17190
    },
    {
      "entropy": 5.447893142700195,
      "epoch": 1.931206828775201,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046345925119386556,
      "loss": 5.1825,
      "mean_token_accuracy": 0.1989078029990196,
      "num_tokens": 38390414.0,
      "step": 17195
    },
    {
      "entropy": 5.471879339218139,
      "epoch": 1.9317684056831583,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004634373153848336,
      "loss": 5.1853,
      "mean_token_accuracy": 0.20094675570726395,
      "num_tokens": 38401569.0,
      "step": 17200
    },
    {
      "entropy": 5.577297782897949,
      "epoch": 1.9323299825911158,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004634153735758845,
      "loss": 5.3271,
      "mean_token_accuracy": 0.19419706463813782,
      "num_tokens": 38413561.0,
      "step": 17205
    },
    {
      "entropy": 5.494162225723267,
      "epoch": 1.9328915594990734,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004633934257677167,
      "loss": 5.2541,
      "mean_token_accuracy": 0.20286345183849336,
      "num_tokens": 38423965.0,
      "step": 17210
    },
    {
      "entropy": 5.408056259155273,
      "epoch": 1.933453136407031,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004633714719610292,
      "loss": 5.1789,
      "mean_token_accuracy": 0.19720540642738343,
      "num_tokens": 38434158.0,
      "step": 17215
    },
    {
      "entropy": 5.422565126419068,
      "epoch": 1.9340147133149885,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046334951215652097,
      "loss": 5.2198,
      "mean_token_accuracy": 0.19582920521497726,
      "num_tokens": 38445449.0,
      "step": 17220
    },
    {
      "entropy": 5.520533037185669,
      "epoch": 1.934576290222946,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046332754635489126,
      "loss": 5.3184,
      "mean_token_accuracy": 0.19835469573736192,
      "num_tokens": 38457601.0,
      "step": 17225
    },
    {
      "entropy": 5.488399171829224,
      "epoch": 1.9351378671309036,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004633055745568394,
      "loss": 5.1959,
      "mean_token_accuracy": 0.20097000598907472,
      "num_tokens": 38468631.0,
      "step": 17230
    },
    {
      "entropy": 5.405942296981811,
      "epoch": 1.935699444038861,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046328359676306506,
      "loss": 5.2488,
      "mean_token_accuracy": 0.19931226074695588,
      "num_tokens": 38479567.0,
      "step": 17235
    },
    {
      "entropy": 5.3931702136993405,
      "epoch": 1.9362610209468185,
      "grad_norm": 1.1875,
      "learning_rate": 0.000463261612974268,
      "loss": 5.2171,
      "mean_token_accuracy": 0.2005762130022049,
      "num_tokens": 38489934.0,
      "step": 17240
    },
    {
      "entropy": 5.586167812347412,
      "epoch": 1.9368225978547762,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004632396231911481,
      "loss": 5.3992,
      "mean_token_accuracy": 0.18913425058126448,
      "num_tokens": 38502214.0,
      "step": 17245
    },
    {
      "entropy": 5.560330724716186,
      "epoch": 1.9373841747627338,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004632176274144056,
      "loss": 5.227,
      "mean_token_accuracy": 0.19509009569883345,
      "num_tokens": 38513915.0,
      "step": 17250
    },
    {
      "entropy": 5.468756866455078,
      "epoch": 1.9379457516706913,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046319562564474083,
      "loss": 5.1914,
      "mean_token_accuracy": 0.2018933430314064,
      "num_tokens": 38525039.0,
      "step": 17255
    },
    {
      "entropy": 5.36847620010376,
      "epoch": 1.9385073285786487,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046317361788285436,
      "loss": 5.1809,
      "mean_token_accuracy": 0.2076302394270897,
      "num_tokens": 38537264.0,
      "step": 17260
    },
    {
      "entropy": 5.481237077713013,
      "epoch": 1.9390689054866064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004631516041294469,
      "loss": 5.2738,
      "mean_token_accuracy": 0.1979181796312332,
      "num_tokens": 38547926.0,
      "step": 17265
    },
    {
      "entropy": 5.478799676895141,
      "epoch": 1.939630482394564,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004631295843852194,
      "loss": 5.2485,
      "mean_token_accuracy": 0.19916255921125411,
      "num_tokens": 38558745.0,
      "step": 17270
    },
    {
      "entropy": 5.439381742477417,
      "epoch": 1.9401920593025215,
      "grad_norm": 1.15625,
      "learning_rate": 0.000463107558650873,
      "loss": 5.1869,
      "mean_token_accuracy": 0.2005079433321953,
      "num_tokens": 38568190.0,
      "step": 17275
    },
    {
      "entropy": 5.512634944915772,
      "epoch": 1.940753636210479,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046308552692710895,
      "loss": 5.2943,
      "mean_token_accuracy": 0.19765460789203643,
      "num_tokens": 38578932.0,
      "step": 17280
    },
    {
      "entropy": 5.4692127227783205,
      "epoch": 1.9413152131184366,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046306348921462876,
      "loss": 5.249,
      "mean_token_accuracy": 0.19438978284597397,
      "num_tokens": 38590501.0,
      "step": 17285
    },
    {
      "entropy": 5.441666126251221,
      "epoch": 1.9418767900263942,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004630414455141342,
      "loss": 5.1529,
      "mean_token_accuracy": 0.20501628518104553,
      "num_tokens": 38600498.0,
      "step": 17290
    },
    {
      "entropy": 5.403835201263428,
      "epoch": 1.9424383669343517,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000463019395826327,
      "loss": 5.1674,
      "mean_token_accuracy": 0.19831580221652984,
      "num_tokens": 38611991.0,
      "step": 17295
    },
    {
      "entropy": 5.408307123184204,
      "epoch": 1.9429999438423091,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046299734015190934,
      "loss": 5.1835,
      "mean_token_accuracy": 0.2026299938559532,
      "num_tokens": 38622284.0,
      "step": 17300
    },
    {
      "entropy": 5.498871850967407,
      "epoch": 1.9435615207502668,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046297527849158344,
      "loss": 5.2866,
      "mean_token_accuracy": 0.19303154647350312,
      "num_tokens": 38633777.0,
      "step": 17305
    },
    {
      "entropy": 5.523743343353272,
      "epoch": 1.9441230976582244,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004629532108460519,
      "loss": 5.2462,
      "mean_token_accuracy": 0.1925231248140335,
      "num_tokens": 38644360.0,
      "step": 17310
    },
    {
      "entropy": 5.442286348342895,
      "epoch": 1.9446846745661819,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046293113721601706,
      "loss": 5.246,
      "mean_token_accuracy": 0.19642017930746078,
      "num_tokens": 38655242.0,
      "step": 17315
    },
    {
      "entropy": 5.419105339050293,
      "epoch": 1.9452462514741393,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046290905760218193,
      "loss": 5.1301,
      "mean_token_accuracy": 0.2068805143237114,
      "num_tokens": 38667792.0,
      "step": 17320
    },
    {
      "entropy": 5.368481588363648,
      "epoch": 1.945807828382097,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046288697200524964,
      "loss": 5.1216,
      "mean_token_accuracy": 0.2080197423696518,
      "num_tokens": 38677735.0,
      "step": 17325
    },
    {
      "entropy": 5.484059238433838,
      "epoch": 1.9463694052900544,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046286488042592325,
      "loss": 5.2307,
      "mean_token_accuracy": 0.19770711064338684,
      "num_tokens": 38689556.0,
      "step": 17330
    },
    {
      "entropy": 5.452941513061523,
      "epoch": 1.9469309821980119,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004628427828649062,
      "loss": 5.2127,
      "mean_token_accuracy": 0.20585621297359466,
      "num_tokens": 38700556.0,
      "step": 17335
    },
    {
      "entropy": 5.426540517807007,
      "epoch": 1.9474925591059695,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046282067932290204,
      "loss": 5.1891,
      "mean_token_accuracy": 0.19898008704185485,
      "num_tokens": 38711019.0,
      "step": 17340
    },
    {
      "entropy": 5.406993341445923,
      "epoch": 1.9480541360139272,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004627985698006146,
      "loss": 5.214,
      "mean_token_accuracy": 0.198685884475708,
      "num_tokens": 38722228.0,
      "step": 17345
    },
    {
      "entropy": 5.408995151519775,
      "epoch": 1.9486157129218846,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000462776454298748,
      "loss": 5.1257,
      "mean_token_accuracy": 0.21163479685783387,
      "num_tokens": 38732183.0,
      "step": 17350
    },
    {
      "entropy": 5.396948766708374,
      "epoch": 1.949177289829842,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046275433281800613,
      "loss": 5.1217,
      "mean_token_accuracy": 0.20836301594972612,
      "num_tokens": 38743659.0,
      "step": 17355
    },
    {
      "entropy": 5.488188123703003,
      "epoch": 1.9497388667377997,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004627322053590935,
      "loss": 5.2781,
      "mean_token_accuracy": 0.19503349363803862,
      "num_tokens": 38755032.0,
      "step": 17360
    },
    {
      "entropy": 5.464019870758056,
      "epoch": 1.9503004436457574,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004627100719227147,
      "loss": 5.1879,
      "mean_token_accuracy": 0.20382087528705597,
      "num_tokens": 38766126.0,
      "step": 17365
    },
    {
      "entropy": 5.500922536849975,
      "epoch": 1.9508620205537148,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046268793250957437,
      "loss": 5.3234,
      "mean_token_accuracy": 0.19810323566198348,
      "num_tokens": 38776921.0,
      "step": 17370
    },
    {
      "entropy": 5.442402410507202,
      "epoch": 1.9514235974616723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046266578712037747,
      "loss": 5.185,
      "mean_token_accuracy": 0.1986000120639801,
      "num_tokens": 38788884.0,
      "step": 17375
    },
    {
      "entropy": 5.562372922897339,
      "epoch": 1.95198517436963,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046264363575582916,
      "loss": 5.3627,
      "mean_token_accuracy": 0.1904572382569313,
      "num_tokens": 38800505.0,
      "step": 17380
    },
    {
      "entropy": 5.472084617614746,
      "epoch": 1.9525467512775876,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004626214784166347,
      "loss": 5.1463,
      "mean_token_accuracy": 0.19802062064409257,
      "num_tokens": 38811132.0,
      "step": 17385
    },
    {
      "entropy": 5.521532678604126,
      "epoch": 1.953108328185545,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004625993151034995,
      "loss": 5.2385,
      "mean_token_accuracy": 0.20619104504585267,
      "num_tokens": 38821349.0,
      "step": 17390
    },
    {
      "entropy": 5.486487293243409,
      "epoch": 1.9536699050935025,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004625771458171294,
      "loss": 5.2641,
      "mean_token_accuracy": 0.19598303586244584,
      "num_tokens": 38832970.0,
      "step": 17395
    },
    {
      "entropy": 5.427934741973877,
      "epoch": 1.9542314820014601,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046255497055823025,
      "loss": 5.2211,
      "mean_token_accuracy": 0.20310211777687073,
      "num_tokens": 38844220.0,
      "step": 17400
    },
    {
      "entropy": 5.4370650291442875,
      "epoch": 1.9547930589094178,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004625327893275082,
      "loss": 5.207,
      "mean_token_accuracy": 0.19337136745452882,
      "num_tokens": 38854848.0,
      "step": 17405
    },
    {
      "entropy": 5.51825213432312,
      "epoch": 1.9553546358173752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046251060212566927,
      "loss": 5.3288,
      "mean_token_accuracy": 0.19359671771526338,
      "num_tokens": 38864827.0,
      "step": 17410
    },
    {
      "entropy": 5.483699893951416,
      "epoch": 1.9559162127253327,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004624884089534201,
      "loss": 5.1656,
      "mean_token_accuracy": 0.20050750374794007,
      "num_tokens": 38875914.0,
      "step": 17415
    },
    {
      "entropy": 5.3950018882751465,
      "epoch": 1.9564777896332903,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004624662098114672,
      "loss": 5.1512,
      "mean_token_accuracy": 0.2009650155901909,
      "num_tokens": 38886565.0,
      "step": 17420
    },
    {
      "entropy": 5.5279758930206295,
      "epoch": 1.9570393665412478,
      "grad_norm": 1.125,
      "learning_rate": 0.00046244400470051754,
      "loss": 5.423,
      "mean_token_accuracy": 0.18694986402988434,
      "num_tokens": 38897319.0,
      "step": 17425
    },
    {
      "entropy": 5.459879922866821,
      "epoch": 1.9576009434492052,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000462421793621278,
      "loss": 5.1626,
      "mean_token_accuracy": 0.2020094558596611,
      "num_tokens": 38908824.0,
      "step": 17430
    },
    {
      "entropy": 5.4830629348754885,
      "epoch": 1.9581625203571629,
      "grad_norm": 1.09375,
      "learning_rate": 0.000462399576574456,
      "loss": 5.2752,
      "mean_token_accuracy": 0.19632184207439424,
      "num_tokens": 38921800.0,
      "step": 17435
    },
    {
      "entropy": 5.473407459259033,
      "epoch": 1.9587240972651205,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004623773535607587,
      "loss": 5.2244,
      "mean_token_accuracy": 0.19653329402208328,
      "num_tokens": 38932284.0,
      "step": 17440
    },
    {
      "entropy": 5.449905920028686,
      "epoch": 1.959285674173078,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004623551245808938,
      "loss": 5.233,
      "mean_token_accuracy": 0.20368038564920427,
      "num_tokens": 38943632.0,
      "step": 17445
    },
    {
      "entropy": 5.49415397644043,
      "epoch": 1.9598472510810354,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004623328896355691,
      "loss": 5.1703,
      "mean_token_accuracy": 0.20023696422576903,
      "num_tokens": 38954701.0,
      "step": 17450
    },
    {
      "entropy": 5.592452764511108,
      "epoch": 1.960408827988993,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046231064872549244,
      "loss": 5.2602,
      "mean_token_accuracy": 0.19546638876199723,
      "num_tokens": 38965625.0,
      "step": 17455
    },
    {
      "entropy": 5.485908651351929,
      "epoch": 1.9609704048969507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046228840185137215,
      "loss": 5.2262,
      "mean_token_accuracy": 0.19822994321584703,
      "num_tokens": 38976989.0,
      "step": 17460
    },
    {
      "entropy": 5.394016981124878,
      "epoch": 1.9615319818049082,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004622661490139165,
      "loss": 5.255,
      "mean_token_accuracy": 0.20444424897432328,
      "num_tokens": 38987661.0,
      "step": 17465
    },
    {
      "entropy": 5.405404329299927,
      "epoch": 1.9620935587128656,
      "grad_norm": 1.125,
      "learning_rate": 0.0004622438902138341,
      "loss": 5.2201,
      "mean_token_accuracy": 0.20042105317115783,
      "num_tokens": 38998014.0,
      "step": 17470
    },
    {
      "entropy": 5.433750772476197,
      "epoch": 1.9626551356208233,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046222162545183357,
      "loss": 5.1313,
      "mean_token_accuracy": 0.20771558284759523,
      "num_tokens": 39008843.0,
      "step": 17475
    },
    {
      "entropy": 5.498402547836304,
      "epoch": 1.963216712528781,
      "grad_norm": 1.125,
      "learning_rate": 0.00046219935472862386,
      "loss": 5.2869,
      "mean_token_accuracy": 0.19607864320278168,
      "num_tokens": 39020173.0,
      "step": 17480
    },
    {
      "entropy": 5.47962384223938,
      "epoch": 1.9637782894367384,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046217707804491404,
      "loss": 5.3027,
      "mean_token_accuracy": 0.18916794508695603,
      "num_tokens": 39033087.0,
      "step": 17485
    },
    {
      "entropy": 5.548334455490112,
      "epoch": 1.9643398663446958,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004621547954014135,
      "loss": 5.2535,
      "mean_token_accuracy": 0.19969939440488815,
      "num_tokens": 39045252.0,
      "step": 17490
    },
    {
      "entropy": 5.534248733520508,
      "epoch": 1.9649014432526535,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004621325067988317,
      "loss": 5.3803,
      "mean_token_accuracy": 0.19075103104114532,
      "num_tokens": 39056755.0,
      "step": 17495
    },
    {
      "entropy": 5.449520206451416,
      "epoch": 1.9654630201606111,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046211021223787827,
      "loss": 5.1537,
      "mean_token_accuracy": 0.20327087342739106,
      "num_tokens": 39066993.0,
      "step": 17500
    },
    {
      "entropy": 5.457622623443603,
      "epoch": 1.9660245970685686,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046208791171926315,
      "loss": 5.2252,
      "mean_token_accuracy": 0.20211263000965118,
      "num_tokens": 39078671.0,
      "step": 17505
    },
    {
      "entropy": 5.496820974349975,
      "epoch": 1.966586173976526,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046206560524369636,
      "loss": 5.2195,
      "mean_token_accuracy": 0.1970143586397171,
      "num_tokens": 39089563.0,
      "step": 17510
    },
    {
      "entropy": 5.543226623535157,
      "epoch": 1.9671477508844837,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046204329281188816,
      "loss": 5.2716,
      "mean_token_accuracy": 0.1950434923171997,
      "num_tokens": 39099837.0,
      "step": 17515
    },
    {
      "entropy": 5.532700252532959,
      "epoch": 1.9677093277924411,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004620209744245489,
      "loss": 5.2697,
      "mean_token_accuracy": 0.19125046133995055,
      "num_tokens": 39112061.0,
      "step": 17520
    },
    {
      "entropy": 5.4774268627166744,
      "epoch": 1.9682709047003986,
      "grad_norm": 1.125,
      "learning_rate": 0.00046199865008238937,
      "loss": 5.2143,
      "mean_token_accuracy": 0.2042147174477577,
      "num_tokens": 39122777.0,
      "step": 17525
    },
    {
      "entropy": 5.415557765960694,
      "epoch": 1.9688324816083562,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004619763197861203,
      "loss": 5.1719,
      "mean_token_accuracy": 0.20116323083639145,
      "num_tokens": 39134827.0,
      "step": 17530
    },
    {
      "entropy": 5.439550447463989,
      "epoch": 1.9693940585163139,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046195398353645256,
      "loss": 5.2262,
      "mean_token_accuracy": 0.19726840108633042,
      "num_tokens": 39144352.0,
      "step": 17535
    },
    {
      "entropy": 5.430553293228149,
      "epoch": 1.9699556354242713,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004619316413340976,
      "loss": 5.1568,
      "mean_token_accuracy": 0.2000689134001732,
      "num_tokens": 39155610.0,
      "step": 17540
    },
    {
      "entropy": 5.403349590301514,
      "epoch": 1.9705172123322288,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046190929317976664,
      "loss": 5.1961,
      "mean_token_accuracy": 0.20000596195459366,
      "num_tokens": 39167058.0,
      "step": 17545
    },
    {
      "entropy": 5.4965376377105715,
      "epoch": 1.9710787892401864,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004618869390741713,
      "loss": 5.3163,
      "mean_token_accuracy": 0.19153826385736467,
      "num_tokens": 39178797.0,
      "step": 17550
    },
    {
      "entropy": 5.462180423736572,
      "epoch": 1.971640366148144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004618645790180233,
      "loss": 5.1562,
      "mean_token_accuracy": 0.2074221134185791,
      "num_tokens": 39188980.0,
      "step": 17555
    },
    {
      "entropy": 5.437407207489014,
      "epoch": 1.9722019430561015,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004618422130120347,
      "loss": 5.1737,
      "mean_token_accuracy": 0.1956959381699562,
      "num_tokens": 39200058.0,
      "step": 17560
    },
    {
      "entropy": 5.433461666107178,
      "epoch": 1.972763519964059,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004618198410569175,
      "loss": 5.2446,
      "mean_token_accuracy": 0.19840811789035798,
      "num_tokens": 39211369.0,
      "step": 17565
    },
    {
      "entropy": 5.456657552719117,
      "epoch": 1.9733250968720166,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046179746315338414,
      "loss": 5.281,
      "mean_token_accuracy": 0.19754408895969391,
      "num_tokens": 39222438.0,
      "step": 17570
    },
    {
      "entropy": 5.495665073394775,
      "epoch": 1.9738866737799743,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046177507930214714,
      "loss": 5.2572,
      "mean_token_accuracy": 0.19137495756149292,
      "num_tokens": 39233047.0,
      "step": 17575
    },
    {
      "entropy": 5.478396368026734,
      "epoch": 1.9744482506879317,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004617526895039191,
      "loss": 5.1886,
      "mean_token_accuracy": 0.19990155398845671,
      "num_tokens": 39244547.0,
      "step": 17580
    },
    {
      "entropy": 5.468042230606079,
      "epoch": 1.9750098275958892,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046173029375941305,
      "loss": 5.1669,
      "mean_token_accuracy": 0.1995212808251381,
      "num_tokens": 39255549.0,
      "step": 17585
    },
    {
      "entropy": 5.383805847167968,
      "epoch": 1.9755714045038468,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000461707892069342,
      "loss": 5.1491,
      "mean_token_accuracy": 0.201088647544384,
      "num_tokens": 39266527.0,
      "step": 17590
    },
    {
      "entropy": 5.43171238899231,
      "epoch": 1.9761329814118045,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046168548443441926,
      "loss": 5.2509,
      "mean_token_accuracy": 0.19938703626394272,
      "num_tokens": 39278013.0,
      "step": 17595
    },
    {
      "entropy": 5.475345325469971,
      "epoch": 1.976694558319762,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004616630708553583,
      "loss": 5.2134,
      "mean_token_accuracy": 0.20027145445346833,
      "num_tokens": 39289349.0,
      "step": 17600
    },
    {
      "entropy": 5.415403127670288,
      "epoch": 1.9772561352277194,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046164065133287276,
      "loss": 5.1983,
      "mean_token_accuracy": 0.2033061057329178,
      "num_tokens": 39301213.0,
      "step": 17605
    },
    {
      "entropy": 5.380880069732666,
      "epoch": 1.977817712135677,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004616182258676765,
      "loss": 5.0803,
      "mean_token_accuracy": 0.20803125947713852,
      "num_tokens": 39310960.0,
      "step": 17610
    },
    {
      "entropy": 5.478777551651001,
      "epoch": 1.9783792890436345,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004615957944604835,
      "loss": 5.2758,
      "mean_token_accuracy": 0.19571359753608703,
      "num_tokens": 39322603.0,
      "step": 17615
    },
    {
      "entropy": 5.492423868179321,
      "epoch": 1.978940865951592,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004615733571120081,
      "loss": 5.2037,
      "mean_token_accuracy": 0.2020836740732193,
      "num_tokens": 39333811.0,
      "step": 17620
    },
    {
      "entropy": 5.536708736419678,
      "epoch": 1.9795024428595496,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004615509138229646,
      "loss": 5.2468,
      "mean_token_accuracy": 0.196746826171875,
      "num_tokens": 39345218.0,
      "step": 17625
    },
    {
      "entropy": 5.39954400062561,
      "epoch": 1.9800640197675072,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046152846459406765,
      "loss": 5.0786,
      "mean_token_accuracy": 0.20189063251018524,
      "num_tokens": 39356777.0,
      "step": 17630
    },
    {
      "entropy": 5.374990081787109,
      "epoch": 1.9806255966754647,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004615060094260321,
      "loss": 5.1884,
      "mean_token_accuracy": 0.2019709512591362,
      "num_tokens": 39368288.0,
      "step": 17635
    },
    {
      "entropy": 5.507363176345825,
      "epoch": 1.9811871735834221,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046148354831957285,
      "loss": 5.2031,
      "mean_token_accuracy": 0.2046121507883072,
      "num_tokens": 39379398.0,
      "step": 17640
    },
    {
      "entropy": 5.41421799659729,
      "epoch": 1.9817487504913798,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046146108127540506,
      "loss": 5.206,
      "mean_token_accuracy": 0.19975281357765198,
      "num_tokens": 39391129.0,
      "step": 17645
    },
    {
      "entropy": 5.501135921478271,
      "epoch": 1.9823103273993374,
      "grad_norm": 1.125,
      "learning_rate": 0.0004614386082942442,
      "loss": 5.3117,
      "mean_token_accuracy": 0.1867342248558998,
      "num_tokens": 39402442.0,
      "step": 17650
    },
    {
      "entropy": 5.589143705368042,
      "epoch": 1.9828719043072949,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004614161293768057,
      "loss": 5.2526,
      "mean_token_accuracy": 0.19941292554140091,
      "num_tokens": 39413044.0,
      "step": 17655
    },
    {
      "entropy": 5.486483812332153,
      "epoch": 1.9834334812152523,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046139364452380536,
      "loss": 5.2048,
      "mean_token_accuracy": 0.2027856007218361,
      "num_tokens": 39423739.0,
      "step": 17660
    },
    {
      "entropy": 5.464673566818237,
      "epoch": 1.98399505812321,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000461371153735959,
      "loss": 5.3457,
      "mean_token_accuracy": 0.19283718168735503,
      "num_tokens": 39436199.0,
      "step": 17665
    },
    {
      "entropy": 5.506896734237671,
      "epoch": 1.9845566350311676,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046134865701398287,
      "loss": 5.2234,
      "mean_token_accuracy": 0.202064910531044,
      "num_tokens": 39449302.0,
      "step": 17670
    },
    {
      "entropy": 5.503790330886841,
      "epoch": 1.985118211939125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046132615435859326,
      "loss": 5.2806,
      "mean_token_accuracy": 0.1987873300909996,
      "num_tokens": 39460723.0,
      "step": 17675
    },
    {
      "entropy": 5.412898874282837,
      "epoch": 1.9856797888470825,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046130364577050655,
      "loss": 5.1852,
      "mean_token_accuracy": 0.20379892736673355,
      "num_tokens": 39471251.0,
      "step": 17680
    },
    {
      "entropy": 5.48502836227417,
      "epoch": 1.9862413657550402,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046128113125043957,
      "loss": 5.1934,
      "mean_token_accuracy": 0.20433451384305953,
      "num_tokens": 39482613.0,
      "step": 17685
    },
    {
      "entropy": 5.436874055862427,
      "epoch": 1.9868029426629978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046125861079910913,
      "loss": 5.2146,
      "mean_token_accuracy": 0.20829022228717803,
      "num_tokens": 39494626.0,
      "step": 17690
    },
    {
      "entropy": 5.380635070800781,
      "epoch": 1.9873645195709553,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004612360844172322,
      "loss": 5.1522,
      "mean_token_accuracy": 0.1982518494129181,
      "num_tokens": 39505302.0,
      "step": 17695
    },
    {
      "entropy": 5.458251237869263,
      "epoch": 1.9879260964789127,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046121355210552626,
      "loss": 5.1284,
      "mean_token_accuracy": 0.19946794509887694,
      "num_tokens": 39515854.0,
      "step": 17700
    },
    {
      "entropy": 5.505066871643066,
      "epoch": 1.9884876733868704,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004611910138647085,
      "loss": 5.2991,
      "mean_token_accuracy": 0.19136724919080733,
      "num_tokens": 39527504.0,
      "step": 17705
    },
    {
      "entropy": 5.495616340637207,
      "epoch": 1.989049250294828,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046116846969549666,
      "loss": 5.2007,
      "mean_token_accuracy": 0.19955624788999557,
      "num_tokens": 39538885.0,
      "step": 17710
    },
    {
      "entropy": 5.456606817245484,
      "epoch": 1.9896108272027853,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004611459195986085,
      "loss": 5.1857,
      "mean_token_accuracy": 0.206529837846756,
      "num_tokens": 39549401.0,
      "step": 17715
    },
    {
      "entropy": 5.42933669090271,
      "epoch": 1.990172404110743,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046112336357476217,
      "loss": 5.2423,
      "mean_token_accuracy": 0.19985499680042268,
      "num_tokens": 39560814.0,
      "step": 17720
    },
    {
      "entropy": 5.435828971862793,
      "epoch": 1.9907339810187006,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004611008016246757,
      "loss": 5.2092,
      "mean_token_accuracy": 0.1969259038567543,
      "num_tokens": 39571490.0,
      "step": 17725
    },
    {
      "entropy": 5.377770185470581,
      "epoch": 1.991295557926658,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046107823374906754,
      "loss": 5.1284,
      "mean_token_accuracy": 0.20340136736631392,
      "num_tokens": 39582815.0,
      "step": 17730
    },
    {
      "entropy": 5.419006204605102,
      "epoch": 1.9918571348346155,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046105565994865614,
      "loss": 5.2253,
      "mean_token_accuracy": 0.200182768702507,
      "num_tokens": 39593712.0,
      "step": 17735
    },
    {
      "entropy": 5.492513036727905,
      "epoch": 1.9924187117425731,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004610330802241605,
      "loss": 5.2059,
      "mean_token_accuracy": 0.20106678158044816,
      "num_tokens": 39604504.0,
      "step": 17740
    },
    {
      "entropy": 5.4835577487945555,
      "epoch": 1.9929802886505308,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004610104945762994,
      "loss": 5.2127,
      "mean_token_accuracy": 0.20139239579439164,
      "num_tokens": 39614971.0,
      "step": 17745
    },
    {
      "entropy": 5.443051433563232,
      "epoch": 1.9935418655584882,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000460987903005792,
      "loss": 5.1625,
      "mean_token_accuracy": 0.20598647743463516,
      "num_tokens": 39626647.0,
      "step": 17750
    },
    {
      "entropy": 5.48618369102478,
      "epoch": 1.9941034424664457,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004609653055133576,
      "loss": 5.2892,
      "mean_token_accuracy": 0.19547720700502397,
      "num_tokens": 39636432.0,
      "step": 17755
    },
    {
      "entropy": 5.509604740142822,
      "epoch": 1.9946650193744033,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046094270209971576,
      "loss": 5.242,
      "mean_token_accuracy": 0.19769977629184723,
      "num_tokens": 39647055.0,
      "step": 17760
    },
    {
      "entropy": 5.4818892002105715,
      "epoch": 1.995226596282361,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004609200927655862,
      "loss": 5.2119,
      "mean_token_accuracy": 0.20348577201366425,
      "num_tokens": 39657891.0,
      "step": 17765
    },
    {
      "entropy": 5.336728525161743,
      "epoch": 1.9957881731903184,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004608974775116888,
      "loss": 5.0962,
      "mean_token_accuracy": 0.2095554456114769,
      "num_tokens": 39668023.0,
      "step": 17770
    },
    {
      "entropy": 5.423732805252075,
      "epoch": 1.9963497500982759,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004608748563387436,
      "loss": 5.213,
      "mean_token_accuracy": 0.20259709507226945,
      "num_tokens": 39678398.0,
      "step": 17775
    },
    {
      "entropy": 5.429116821289062,
      "epoch": 1.9969113270062335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004608522292474708,
      "loss": 5.2194,
      "mean_token_accuracy": 0.19839489758014678,
      "num_tokens": 39690619.0,
      "step": 17780
    },
    {
      "entropy": 5.409654188156128,
      "epoch": 1.9974729039141912,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004608295962385911,
      "loss": 5.1314,
      "mean_token_accuracy": 0.20759865939617156,
      "num_tokens": 39701690.0,
      "step": 17785
    },
    {
      "entropy": 5.460873937606811,
      "epoch": 1.9980344808221486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046080695731282484,
      "loss": 5.288,
      "mean_token_accuracy": 0.1898223027586937,
      "num_tokens": 39713878.0,
      "step": 17790
    },
    {
      "entropy": 5.498700046539307,
      "epoch": 1.998596057730106,
      "grad_norm": 1.046875,
      "learning_rate": 0.000460784312470893,
      "loss": 5.2149,
      "mean_token_accuracy": 0.20086885690689088,
      "num_tokens": 39725164.0,
      "step": 17795
    },
    {
      "entropy": 5.446836137771607,
      "epoch": 1.9991576346380637,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046076166171351673,
      "loss": 5.1486,
      "mean_token_accuracy": 0.197618767619133,
      "num_tokens": 39735354.0,
      "step": 17800
    },
    {
      "entropy": 5.5237993717193605,
      "epoch": 1.9997192115460214,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000460739005041417,
      "loss": 5.2979,
      "mean_token_accuracy": 0.19139618277549744,
      "num_tokens": 39746618.0,
      "step": 17805
    },
    {
      "entropy": 5.478542115953234,
      "epoch": 2.000224630763183,
      "grad_norm": 1.125,
      "learning_rate": 0.0004607163424553153,
      "loss": 5.2336,
      "mean_token_accuracy": 0.1988481299744712,
      "num_tokens": 39755697.0,
      "step": 17810
    },
    {
      "entropy": 5.375217008590698,
      "epoch": 2.0007862076711405,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046069367395593325,
      "loss": 5.1448,
      "mean_token_accuracy": 0.19753584563732146,
      "num_tokens": 39766311.0,
      "step": 17815
    },
    {
      "entropy": 5.493737888336182,
      "epoch": 2.001347784579098,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004606709995439926,
      "loss": 5.2503,
      "mean_token_accuracy": 0.19518130272626877,
      "num_tokens": 39777504.0,
      "step": 17820
    },
    {
      "entropy": 5.506360960006714,
      "epoch": 2.001909361487056,
      "grad_norm": 1.125,
      "learning_rate": 0.00046064831922021534,
      "loss": 5.1152,
      "mean_token_accuracy": 0.1954638808965683,
      "num_tokens": 39789190.0,
      "step": 17825
    },
    {
      "entropy": 5.456617164611816,
      "epoch": 2.002470938395013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004606256329853235,
      "loss": 5.1164,
      "mean_token_accuracy": 0.19629822969436644,
      "num_tokens": 39801534.0,
      "step": 17830
    },
    {
      "entropy": 5.480358552932739,
      "epoch": 2.0030325153029707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004606029408400396,
      "loss": 5.1745,
      "mean_token_accuracy": 0.19565867334604264,
      "num_tokens": 39812353.0,
      "step": 17835
    },
    {
      "entropy": 5.422595739364624,
      "epoch": 2.0035940922109283,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004605802427850861,
      "loss": 5.0751,
      "mean_token_accuracy": 0.2053135707974434,
      "num_tokens": 39823313.0,
      "step": 17840
    },
    {
      "entropy": 5.358608102798462,
      "epoch": 2.004155669118886,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046055753882118564,
      "loss": 5.0494,
      "mean_token_accuracy": 0.21111321151256562,
      "num_tokens": 39833966.0,
      "step": 17845
    },
    {
      "entropy": 5.429851007461548,
      "epoch": 2.004717246026843,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004605348289490611,
      "loss": 5.1694,
      "mean_token_accuracy": 0.20227143168449402,
      "num_tokens": 39843963.0,
      "step": 17850
    },
    {
      "entropy": 5.4828473091125485,
      "epoch": 2.005278822934801,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004605121131694358,
      "loss": 5.1418,
      "mean_token_accuracy": 0.20275248140096663,
      "num_tokens": 39855237.0,
      "step": 17855
    },
    {
      "entropy": 5.473034334182739,
      "epoch": 2.0058403998427585,
      "grad_norm": 1.0,
      "learning_rate": 0.0004604893914830328,
      "loss": 5.2772,
      "mean_token_accuracy": 0.1957951307296753,
      "num_tokens": 39867270.0,
      "step": 17860
    },
    {
      "entropy": 5.504860591888428,
      "epoch": 2.006401976750716,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046046666389057563,
      "loss": 5.1878,
      "mean_token_accuracy": 0.20437454134225846,
      "num_tokens": 39878103.0,
      "step": 17865
    },
    {
      "entropy": 5.427053928375244,
      "epoch": 2.0069635536586734,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000460443930392788,
      "loss": 5.1095,
      "mean_token_accuracy": 0.20045692771673201,
      "num_tokens": 39888945.0,
      "step": 17870
    },
    {
      "entropy": 5.5109058856964115,
      "epoch": 2.007525130566631,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046042119099039367,
      "loss": 5.3086,
      "mean_token_accuracy": 0.19451324194669722,
      "num_tokens": 39899848.0,
      "step": 17875
    },
    {
      "entropy": 5.465785312652588,
      "epoch": 2.0080867074745887,
      "grad_norm": 1.125,
      "learning_rate": 0.00046039844568411667,
      "loss": 5.1672,
      "mean_token_accuracy": 0.19846513122320175,
      "num_tokens": 39911824.0,
      "step": 17880
    },
    {
      "entropy": 5.437666034698486,
      "epoch": 2.0086482843825464,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004603756944746813,
      "loss": 5.1157,
      "mean_token_accuracy": 0.20339842289686202,
      "num_tokens": 39923257.0,
      "step": 17885
    },
    {
      "entropy": 5.376050567626953,
      "epoch": 2.0092098612905036,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046035293736281185,
      "loss": 5.0813,
      "mean_token_accuracy": 0.20711201876401902,
      "num_tokens": 39935312.0,
      "step": 17890
    },
    {
      "entropy": 5.380534362792969,
      "epoch": 2.0097714381984613,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004603301743492331,
      "loss": 5.1118,
      "mean_token_accuracy": 0.20278966277837754,
      "num_tokens": 39946468.0,
      "step": 17895
    },
    {
      "entropy": 5.292778491973877,
      "epoch": 2.010333015106419,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004603074054346696,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21465659737586976,
      "num_tokens": 39957517.0,
      "step": 17900
    },
    {
      "entropy": 5.366285181045532,
      "epoch": 2.0108945920143766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004602846306198465,
      "loss": 5.1095,
      "mean_token_accuracy": 0.2004966050386429,
      "num_tokens": 39968385.0,
      "step": 17905
    },
    {
      "entropy": 5.3870128154754635,
      "epoch": 2.011456168922334,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004602618499054889,
      "loss": 5.0448,
      "mean_token_accuracy": 0.2043771043419838,
      "num_tokens": 39979393.0,
      "step": 17910
    },
    {
      "entropy": 5.452572631835937,
      "epoch": 2.0120177458302915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004602390632923221,
      "loss": 5.1317,
      "mean_token_accuracy": 0.20247626453638076,
      "num_tokens": 39989709.0,
      "step": 17915
    },
    {
      "entropy": 5.444457292556763,
      "epoch": 2.012579322738249,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046021627078107177,
      "loss": 5.0769,
      "mean_token_accuracy": 0.2018706366419792,
      "num_tokens": 40000472.0,
      "step": 17920
    },
    {
      "entropy": 5.294516372680664,
      "epoch": 2.0131408996462063,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046019347237246343,
      "loss": 4.9651,
      "mean_token_accuracy": 0.2092535302042961,
      "num_tokens": 40011062.0,
      "step": 17925
    },
    {
      "entropy": 5.4835052490234375,
      "epoch": 2.013702476554164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004601706680672232,
      "loss": 5.2223,
      "mean_token_accuracy": 0.1911832183599472,
      "num_tokens": 40022430.0,
      "step": 17930
    },
    {
      "entropy": 5.378408432006836,
      "epoch": 2.0142640534621217,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004601478578660771,
      "loss": 5.1182,
      "mean_token_accuracy": 0.20309155732393264,
      "num_tokens": 40033077.0,
      "step": 17935
    },
    {
      "entropy": 5.419387531280518,
      "epoch": 2.0148256303700793,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004601250417697513,
      "loss": 5.1314,
      "mean_token_accuracy": 0.20527995228767396,
      "num_tokens": 40044258.0,
      "step": 17940
    },
    {
      "entropy": 5.4610981941223145,
      "epoch": 2.0153872072780366,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004601022197789724,
      "loss": 5.1488,
      "mean_token_accuracy": 0.19742362797260285,
      "num_tokens": 40055876.0,
      "step": 17945
    },
    {
      "entropy": 5.3891064643859865,
      "epoch": 2.015948784185994,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004600793918944671,
      "loss": 5.0561,
      "mean_token_accuracy": 0.21038345545530318,
      "num_tokens": 40066691.0,
      "step": 17950
    },
    {
      "entropy": 5.377702474594116,
      "epoch": 2.016510361093952,
      "grad_norm": 1.25,
      "learning_rate": 0.0004600565581169621,
      "loss": 5.0414,
      "mean_token_accuracy": 0.2023324579000473,
      "num_tokens": 40077335.0,
      "step": 17955
    },
    {
      "entropy": 5.377427053451538,
      "epoch": 2.0170719380019095,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046003371844718455,
      "loss": 5.1843,
      "mean_token_accuracy": 0.19583686143159867,
      "num_tokens": 40089053.0,
      "step": 17960
    },
    {
      "entropy": 5.410895347595215,
      "epoch": 2.0176335149098668,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046001087288586164,
      "loss": 5.064,
      "mean_token_accuracy": 0.21343138813972473,
      "num_tokens": 40099371.0,
      "step": 17965
    },
    {
      "entropy": 5.44838547706604,
      "epoch": 2.0181950918178244,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004599880214337208,
      "loss": 5.1837,
      "mean_token_accuracy": 0.19507868885993956,
      "num_tokens": 40110274.0,
      "step": 17970
    },
    {
      "entropy": 5.447851467132568,
      "epoch": 2.018756668725782,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004599651640914896,
      "loss": 5.2016,
      "mean_token_accuracy": 0.19923406094312668,
      "num_tokens": 40120364.0,
      "step": 17975
    },
    {
      "entropy": 5.411220073699951,
      "epoch": 2.0193182456337397,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004599423008598957,
      "loss": 5.1498,
      "mean_token_accuracy": 0.2032131403684616,
      "num_tokens": 40132400.0,
      "step": 17980
    },
    {
      "entropy": 5.461653137207032,
      "epoch": 2.019879822541697,
      "grad_norm": 1.25,
      "learning_rate": 0.00045991943173966744,
      "loss": 5.1848,
      "mean_token_accuracy": 0.19805140495300294,
      "num_tokens": 40142820.0,
      "step": 17985
    },
    {
      "entropy": 5.452293825149536,
      "epoch": 2.0204413994496546,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004598965567315326,
      "loss": 5.1996,
      "mean_token_accuracy": 0.19969666600227357,
      "num_tokens": 40153620.0,
      "step": 17990
    },
    {
      "entropy": 5.413561582565308,
      "epoch": 2.0210029763576123,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004598736758362198,
      "loss": 5.1238,
      "mean_token_accuracy": 0.20340935736894608,
      "num_tokens": 40164250.0,
      "step": 17995
    },
    {
      "entropy": 5.411157274246216,
      "epoch": 2.02156455326557,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045985078905445734,
      "loss": 5.0615,
      "mean_token_accuracy": 0.20011104941368102,
      "num_tokens": 40175809.0,
      "step": 18000
    },
    {
      "epoch": 2.02156455326557,
      "eval_entropy": 5.321115857459603,
      "eval_loss": 5.305892467498779,
      "eval_mean_token_accuracy": 0.20272016978606694,
      "eval_num_tokens": 40175809.0,
      "eval_runtime": 59.3033,
      "eval_samples_per_second": 522.956,
      "eval_steps_per_second": 65.376,
      "step": 18000
    },
    {
      "entropy": 5.342676639556885,
      "epoch": 2.022126130173527,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045982789638697414,
      "loss": 5.0159,
      "mean_token_accuracy": 0.2111600697040558,
      "num_tokens": 40187855.0,
      "step": 18005
    },
    {
      "entropy": 5.4333912372589115,
      "epoch": 2.022687707081485,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045980499783449897,
      "loss": 5.0945,
      "mean_token_accuracy": 0.20565036237239837,
      "num_tokens": 40198802.0,
      "step": 18010
    },
    {
      "entropy": 5.5392069816589355,
      "epoch": 2.0232492839894425,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045978209339776104,
      "loss": 5.2913,
      "mean_token_accuracy": 0.19172475039958953,
      "num_tokens": 40210636.0,
      "step": 18015
    },
    {
      "entropy": 5.48834981918335,
      "epoch": 2.0238108608973997,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045975918307748965,
      "loss": 5.177,
      "mean_token_accuracy": 0.20160965770483016,
      "num_tokens": 40221216.0,
      "step": 18020
    },
    {
      "entropy": 5.357290506362915,
      "epoch": 2.0243724378053574,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004597362668744141,
      "loss": 5.0046,
      "mean_token_accuracy": 0.21414878219366074,
      "num_tokens": 40231988.0,
      "step": 18025
    },
    {
      "entropy": 5.464912080764771,
      "epoch": 2.024934014713315,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004597133447892642,
      "loss": 5.3246,
      "mean_token_accuracy": 0.1910130575299263,
      "num_tokens": 40243789.0,
      "step": 18030
    },
    {
      "entropy": 5.443649530410767,
      "epoch": 2.0254955916212727,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004596904168227698,
      "loss": 5.1388,
      "mean_token_accuracy": 0.20498851090669631,
      "num_tokens": 40254771.0,
      "step": 18035
    },
    {
      "entropy": 5.445785856246948,
      "epoch": 2.02605716852923,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045966748297566085,
      "loss": 5.1119,
      "mean_token_accuracy": 0.20620031654834747,
      "num_tokens": 40265444.0,
      "step": 18040
    },
    {
      "entropy": 5.415645790100098,
      "epoch": 2.0266187454371876,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045964454324866767,
      "loss": 5.132,
      "mean_token_accuracy": 0.20005106925964355,
      "num_tokens": 40277146.0,
      "step": 18045
    },
    {
      "entropy": 5.396540737152099,
      "epoch": 2.0271803223451452,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045962159764252055,
      "loss": 5.183,
      "mean_token_accuracy": 0.206864395737648,
      "num_tokens": 40286493.0,
      "step": 18050
    },
    {
      "entropy": 5.450122547149658,
      "epoch": 2.027741899253103,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004595986461579502,
      "loss": 5.1386,
      "mean_token_accuracy": 0.20539827197790145,
      "num_tokens": 40297209.0,
      "step": 18055
    },
    {
      "entropy": 5.372835731506347,
      "epoch": 2.02830347616106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045957568879568736,
      "loss": 5.1048,
      "mean_token_accuracy": 0.20194728523492814,
      "num_tokens": 40308666.0,
      "step": 18060
    },
    {
      "entropy": 5.4655797481536865,
      "epoch": 2.0288650530690178,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045955272555646306,
      "loss": 5.2179,
      "mean_token_accuracy": 0.19341700226068498,
      "num_tokens": 40321145.0,
      "step": 18065
    },
    {
      "entropy": 5.511399030685425,
      "epoch": 2.0294266299769754,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004595297564410083,
      "loss": 5.14,
      "mean_token_accuracy": 0.20271724164485933,
      "num_tokens": 40332418.0,
      "step": 18070
    },
    {
      "entropy": 5.383347415924073,
      "epoch": 2.029988206884933,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045950678145005456,
      "loss": 5.0994,
      "mean_token_accuracy": 0.2037181079387665,
      "num_tokens": 40342838.0,
      "step": 18075
    },
    {
      "entropy": 5.344696950912476,
      "epoch": 2.0305497837928903,
      "grad_norm": 1.125,
      "learning_rate": 0.0004594838005843333,
      "loss": 5.0671,
      "mean_token_accuracy": 0.20574889779090882,
      "num_tokens": 40352940.0,
      "step": 18080
    },
    {
      "entropy": 5.551378393173218,
      "epoch": 2.031111360700848,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045946081384457633,
      "loss": 5.1909,
      "mean_token_accuracy": 0.19839725643396378,
      "num_tokens": 40364273.0,
      "step": 18085
    },
    {
      "entropy": 5.464349460601807,
      "epoch": 2.0316729376088056,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045943782123151556,
      "loss": 5.0678,
      "mean_token_accuracy": 0.2119649678468704,
      "num_tokens": 40374617.0,
      "step": 18090
    },
    {
      "entropy": 5.406306219100952,
      "epoch": 2.0322345145167633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004594148227458829,
      "loss": 5.1752,
      "mean_token_accuracy": 0.19823758900165558,
      "num_tokens": 40385398.0,
      "step": 18095
    },
    {
      "entropy": 5.546410942077637,
      "epoch": 2.0327960914247205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004593918183884108,
      "loss": 5.263,
      "mean_token_accuracy": 0.19752512574195863,
      "num_tokens": 40397030.0,
      "step": 18100
    },
    {
      "entropy": 5.546636247634888,
      "epoch": 2.033357668332678,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004593688081598318,
      "loss": 5.2165,
      "mean_token_accuracy": 0.19881228655576705,
      "num_tokens": 40408027.0,
      "step": 18105
    },
    {
      "entropy": 5.5013751029968265,
      "epoch": 2.033919245240636,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045934579206087837,
      "loss": 5.1683,
      "mean_token_accuracy": 0.1980292335152626,
      "num_tokens": 40418755.0,
      "step": 18110
    },
    {
      "entropy": 5.4994605541229244,
      "epoch": 2.034480822148593,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045932277009228336,
      "loss": 5.1846,
      "mean_token_accuracy": 0.1996479034423828,
      "num_tokens": 40430694.0,
      "step": 18115
    },
    {
      "entropy": 5.474099493026733,
      "epoch": 2.0350423990565507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045929974225477986,
      "loss": 5.1406,
      "mean_token_accuracy": 0.20339277535676956,
      "num_tokens": 40442019.0,
      "step": 18120
    },
    {
      "entropy": 5.468339729309082,
      "epoch": 2.0356039759645084,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045927670854910117,
      "loss": 5.1561,
      "mean_token_accuracy": 0.1907990410923958,
      "num_tokens": 40454120.0,
      "step": 18125
    },
    {
      "entropy": 5.468143033981323,
      "epoch": 2.036165552872466,
      "grad_norm": 1.0,
      "learning_rate": 0.0004592536689759806,
      "loss": 5.2878,
      "mean_token_accuracy": 0.1962733805179596,
      "num_tokens": 40467445.0,
      "step": 18130
    },
    {
      "entropy": 5.396000623703003,
      "epoch": 2.0367271297804233,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045923062353615167,
      "loss": 5.0807,
      "mean_token_accuracy": 0.19919672310352327,
      "num_tokens": 40477512.0,
      "step": 18135
    },
    {
      "entropy": 5.377988529205322,
      "epoch": 2.037288706688381,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045920757223034833,
      "loss": 5.0848,
      "mean_token_accuracy": 0.2078073188662529,
      "num_tokens": 40488209.0,
      "step": 18140
    },
    {
      "entropy": 5.503016757965088,
      "epoch": 2.0378502835963386,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004591845150593044,
      "loss": 5.1584,
      "mean_token_accuracy": 0.20099363029003142,
      "num_tokens": 40499793.0,
      "step": 18145
    },
    {
      "entropy": 5.492732429504395,
      "epoch": 2.0384118605042962,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045916145202375407,
      "loss": 5.2591,
      "mean_token_accuracy": 0.18891579508781434,
      "num_tokens": 40510598.0,
      "step": 18150
    },
    {
      "entropy": 5.550028085708618,
      "epoch": 2.0389734374122535,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004591383831244317,
      "loss": 5.1737,
      "mean_token_accuracy": 0.20601840019226075,
      "num_tokens": 40521289.0,
      "step": 18155
    },
    {
      "entropy": 5.514279079437256,
      "epoch": 2.039535014320211,
      "grad_norm": 2.09375,
      "learning_rate": 0.00045911530836207173,
      "loss": 5.2287,
      "mean_token_accuracy": 0.20284309089183808,
      "num_tokens": 40532048.0,
      "step": 18160
    },
    {
      "entropy": 5.421380043029785,
      "epoch": 2.0400965912281688,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045909222773740904,
      "loss": 5.219,
      "mean_token_accuracy": 0.202123162150383,
      "num_tokens": 40543699.0,
      "step": 18165
    },
    {
      "entropy": 5.511491966247559,
      "epoch": 2.0406581681361264,
      "grad_norm": 1.125,
      "learning_rate": 0.00045906914125117833,
      "loss": 5.2368,
      "mean_token_accuracy": 0.20196765661239624,
      "num_tokens": 40554626.0,
      "step": 18170
    },
    {
      "entropy": 5.497847557067871,
      "epoch": 2.0412197450440837,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045904604890411476,
      "loss": 5.1873,
      "mean_token_accuracy": 0.19653425216674805,
      "num_tokens": 40567046.0,
      "step": 18175
    },
    {
      "entropy": 5.581865406036377,
      "epoch": 2.0417813219520413,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045902295069695373,
      "loss": 5.2394,
      "mean_token_accuracy": 0.19376737773418426,
      "num_tokens": 40578472.0,
      "step": 18180
    },
    {
      "entropy": 5.43798828125,
      "epoch": 2.042342898859999,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004589998466304305,
      "loss": 5.1616,
      "mean_token_accuracy": 0.20614684820175172,
      "num_tokens": 40589012.0,
      "step": 18185
    },
    {
      "entropy": 5.367080783843994,
      "epoch": 2.0429044757679566,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045897673670528086,
      "loss": 5.1082,
      "mean_token_accuracy": 0.20400851219892502,
      "num_tokens": 40600807.0,
      "step": 18190
    },
    {
      "entropy": 5.541237449645996,
      "epoch": 2.043466052675914,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004589536209222404,
      "loss": 5.2549,
      "mean_token_accuracy": 0.19673992544412613,
      "num_tokens": 40612578.0,
      "step": 18195
    },
    {
      "entropy": 5.422832822799682,
      "epoch": 2.0440276295838715,
      "grad_norm": 1.125,
      "learning_rate": 0.0004589304992820454,
      "loss": 5.0054,
      "mean_token_accuracy": 0.21189106851816178,
      "num_tokens": 40623504.0,
      "step": 18200
    },
    {
      "entropy": 5.4539172649383545,
      "epoch": 2.044589206491829,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000458907371785432,
      "loss": 5.1834,
      "mean_token_accuracy": 0.20232882499694824,
      "num_tokens": 40633676.0,
      "step": 18205
    },
    {
      "entropy": 5.46436653137207,
      "epoch": 2.0451507833997864,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004588842384331366,
      "loss": 5.2307,
      "mean_token_accuracy": 0.19831877201795578,
      "num_tokens": 40645991.0,
      "step": 18210
    },
    {
      "entropy": 5.406685876846313,
      "epoch": 2.045712360307744,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045886109922589564,
      "loss": 5.0581,
      "mean_token_accuracy": 0.20341494381427766,
      "num_tokens": 40656617.0,
      "step": 18215
    },
    {
      "entropy": 5.4644303798675535,
      "epoch": 2.0462739372157017,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045883795416444594,
      "loss": 5.1888,
      "mean_token_accuracy": 0.2039027139544487,
      "num_tokens": 40669708.0,
      "step": 18220
    },
    {
      "entropy": 5.429202508926392,
      "epoch": 2.0468355141236594,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045881480324952455,
      "loss": 5.1642,
      "mean_token_accuracy": 0.19802865236997605,
      "num_tokens": 40680248.0,
      "step": 18225
    },
    {
      "entropy": 5.529597806930542,
      "epoch": 2.0473970910316166,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045879164648186853,
      "loss": 5.2533,
      "mean_token_accuracy": 0.1969241663813591,
      "num_tokens": 40692704.0,
      "step": 18230
    },
    {
      "entropy": 5.526028156280518,
      "epoch": 2.0479586679395743,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045876848386221513,
      "loss": 5.1539,
      "mean_token_accuracy": 0.2051002413034439,
      "num_tokens": 40703156.0,
      "step": 18235
    },
    {
      "entropy": 5.4386255741119385,
      "epoch": 2.048520244847532,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045874531539130204,
      "loss": 5.117,
      "mean_token_accuracy": 0.2031269609928131,
      "num_tokens": 40713996.0,
      "step": 18240
    },
    {
      "entropy": 5.371656608581543,
      "epoch": 2.0490818217554896,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045872214106986684,
      "loss": 5.0362,
      "mean_token_accuracy": 0.2048846647143364,
      "num_tokens": 40724532.0,
      "step": 18245
    },
    {
      "entropy": 5.483946418762207,
      "epoch": 2.049643398663447,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045869896089864733,
      "loss": 5.2416,
      "mean_token_accuracy": 0.19358222484588622,
      "num_tokens": 40736045.0,
      "step": 18250
    },
    {
      "entropy": 5.337462568283081,
      "epoch": 2.0502049755714045,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004586757748783817,
      "loss": 5.0272,
      "mean_token_accuracy": 0.20495191663503648,
      "num_tokens": 40747398.0,
      "step": 18255
    },
    {
      "entropy": 5.323112201690674,
      "epoch": 2.050766552479362,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045865258300980817,
      "loss": 4.9858,
      "mean_token_accuracy": 0.21282731890678405,
      "num_tokens": 40758219.0,
      "step": 18260
    },
    {
      "entropy": 5.454240703582764,
      "epoch": 2.05132812938732,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045862938529366514,
      "loss": 5.09,
      "mean_token_accuracy": 0.20576892495155336,
      "num_tokens": 40770002.0,
      "step": 18265
    },
    {
      "entropy": 5.378674936294556,
      "epoch": 2.051889706295277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004586061817306913,
      "loss": 5.065,
      "mean_token_accuracy": 0.2059107542037964,
      "num_tokens": 40780087.0,
      "step": 18270
    },
    {
      "entropy": 5.4030139446258545,
      "epoch": 2.0524512832032347,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045858297232162536,
      "loss": 5.2182,
      "mean_token_accuracy": 0.19829004555940627,
      "num_tokens": 40792225.0,
      "step": 18275
    },
    {
      "entropy": 5.499179172515869,
      "epoch": 2.0530128601111923,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045855975706720644,
      "loss": 5.2646,
      "mean_token_accuracy": 0.1952887460589409,
      "num_tokens": 40804437.0,
      "step": 18280
    },
    {
      "entropy": 5.409182548522949,
      "epoch": 2.05357443701915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004585365359681736,
      "loss": 5.0205,
      "mean_token_accuracy": 0.21883408427238465,
      "num_tokens": 40814625.0,
      "step": 18285
    },
    {
      "entropy": 5.3476338386535645,
      "epoch": 2.054136013927107,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045851330902526633,
      "loss": 5.0638,
      "mean_token_accuracy": 0.20834892839193345,
      "num_tokens": 40825843.0,
      "step": 18290
    },
    {
      "entropy": 5.419632816314698,
      "epoch": 2.054697590835065,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045849007623922406,
      "loss": 5.241,
      "mean_token_accuracy": 0.19405852556228637,
      "num_tokens": 40837146.0,
      "step": 18295
    },
    {
      "entropy": 5.560157108306885,
      "epoch": 2.0552591677430225,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004584668376107867,
      "loss": 5.317,
      "mean_token_accuracy": 0.1893691748380661,
      "num_tokens": 40849109.0,
      "step": 18300
    },
    {
      "entropy": 5.447722721099853,
      "epoch": 2.0558207446509797,
      "grad_norm": 1.109375,
      "learning_rate": 0.000458443593140694,
      "loss": 5.0744,
      "mean_token_accuracy": 0.20514724403619766,
      "num_tokens": 40861168.0,
      "step": 18305
    },
    {
      "entropy": 5.456511831283569,
      "epoch": 2.0563823215589374,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045842034282968615,
      "loss": 5.1933,
      "mean_token_accuracy": 0.20332808643579484,
      "num_tokens": 40873403.0,
      "step": 18310
    },
    {
      "entropy": 5.419351530075073,
      "epoch": 2.056943898466895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004583970866785035,
      "loss": 5.1011,
      "mean_token_accuracy": 0.20136093050241471,
      "num_tokens": 40883434.0,
      "step": 18315
    },
    {
      "entropy": 5.448317337036133,
      "epoch": 2.0575054753748527,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045837382468788643,
      "loss": 5.1265,
      "mean_token_accuracy": 0.20032256245613098,
      "num_tokens": 40894742.0,
      "step": 18320
    },
    {
      "entropy": 5.359233283996582,
      "epoch": 2.05806705228281,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004583505568585757,
      "loss": 5.1075,
      "mean_token_accuracy": 0.20630411952733993,
      "num_tokens": 40905233.0,
      "step": 18325
    },
    {
      "entropy": 5.384528303146363,
      "epoch": 2.0586286291907676,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045832728319131205,
      "loss": 5.0912,
      "mean_token_accuracy": 0.2106534093618393,
      "num_tokens": 40915897.0,
      "step": 18330
    },
    {
      "entropy": 5.392107009887695,
      "epoch": 2.0591902060987253,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004583040036868367,
      "loss": 5.1009,
      "mean_token_accuracy": 0.1980993553996086,
      "num_tokens": 40928178.0,
      "step": 18335
    },
    {
      "entropy": 5.419974660873413,
      "epoch": 2.059751783006683,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004582807183458907,
      "loss": 5.197,
      "mean_token_accuracy": 0.19679423570632934,
      "num_tokens": 40938988.0,
      "step": 18340
    },
    {
      "entropy": 5.402058029174805,
      "epoch": 2.06031335991464,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045825742716921553,
      "loss": 5.0603,
      "mean_token_accuracy": 0.2110185816884041,
      "num_tokens": 40949671.0,
      "step": 18345
    },
    {
      "entropy": 5.333623218536377,
      "epoch": 2.060874936822598,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004582341301575529,
      "loss": 5.024,
      "mean_token_accuracy": 0.20600511878728867,
      "num_tokens": 40960818.0,
      "step": 18350
    },
    {
      "entropy": 5.516381692886353,
      "epoch": 2.0614365137305555,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045821082731164445,
      "loss": 5.2698,
      "mean_token_accuracy": 0.19057075828313827,
      "num_tokens": 40973866.0,
      "step": 18355
    },
    {
      "entropy": 5.427972030639649,
      "epoch": 2.061998090638513,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004581875186322322,
      "loss": 5.1414,
      "mean_token_accuracy": 0.19878353029489518,
      "num_tokens": 40983355.0,
      "step": 18360
    },
    {
      "entropy": 5.40568733215332,
      "epoch": 2.0625596675464704,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045816420412005826,
      "loss": 5.0394,
      "mean_token_accuracy": 0.20313866138458253,
      "num_tokens": 40994045.0,
      "step": 18365
    },
    {
      "entropy": 5.344874572753906,
      "epoch": 2.063121244454428,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045814088377586505,
      "loss": 5.0601,
      "mean_token_accuracy": 0.21148548722267152,
      "num_tokens": 41003974.0,
      "step": 18370
    },
    {
      "entropy": 5.418494415283203,
      "epoch": 2.0636828213623857,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045811755760039513,
      "loss": 5.2131,
      "mean_token_accuracy": 0.20303902626037598,
      "num_tokens": 41015262.0,
      "step": 18375
    },
    {
      "entropy": 5.526286554336548,
      "epoch": 2.0642443982703433,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045809422559439113,
      "loss": 5.2442,
      "mean_token_accuracy": 0.19860045462846757,
      "num_tokens": 41027103.0,
      "step": 18380
    },
    {
      "entropy": 5.481543779373169,
      "epoch": 2.0648059751783006,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045807088775859595,
      "loss": 5.1236,
      "mean_token_accuracy": 0.20718812793493271,
      "num_tokens": 41037240.0,
      "step": 18385
    },
    {
      "entropy": 5.492831611633301,
      "epoch": 2.065367552086258,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004580475440937527,
      "loss": 5.1288,
      "mean_token_accuracy": 0.2053864374756813,
      "num_tokens": 41048258.0,
      "step": 18390
    },
    {
      "entropy": 5.467458629608155,
      "epoch": 2.065929128994216,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045802419460060464,
      "loss": 5.2155,
      "mean_token_accuracy": 0.19467367082834244,
      "num_tokens": 41058204.0,
      "step": 18395
    },
    {
      "entropy": 5.5256345748901365,
      "epoch": 2.066490705902173,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004580008392798953,
      "loss": 5.2129,
      "mean_token_accuracy": 0.19870574325323104,
      "num_tokens": 41070019.0,
      "step": 18400
    },
    {
      "entropy": 5.524305391311645,
      "epoch": 2.0670522828101308,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045797747813236815,
      "loss": 5.1885,
      "mean_token_accuracy": 0.20117858648300171,
      "num_tokens": 41081503.0,
      "step": 18405
    },
    {
      "entropy": 5.471236801147461,
      "epoch": 2.0676138597180884,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045795411115876725,
      "loss": 5.1433,
      "mean_token_accuracy": 0.19625366777181624,
      "num_tokens": 41091959.0,
      "step": 18410
    },
    {
      "entropy": 5.44686918258667,
      "epoch": 2.068175436626046,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045793073835983644,
      "loss": 5.1891,
      "mean_token_accuracy": 0.19635324776172638,
      "num_tokens": 41103193.0,
      "step": 18415
    },
    {
      "entropy": 5.514858627319336,
      "epoch": 2.0687370135340033,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045790735973631996,
      "loss": 5.2382,
      "mean_token_accuracy": 0.2046363964676857,
      "num_tokens": 41115886.0,
      "step": 18420
    },
    {
      "entropy": 5.456508922576904,
      "epoch": 2.069298590441961,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045788397528896224,
      "loss": 5.1028,
      "mean_token_accuracy": 0.20412856042385102,
      "num_tokens": 41126868.0,
      "step": 18425
    },
    {
      "entropy": 5.498434734344483,
      "epoch": 2.0698601673499186,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045786058501850774,
      "loss": 5.1354,
      "mean_token_accuracy": 0.20957268327474593,
      "num_tokens": 41136417.0,
      "step": 18430
    },
    {
      "entropy": 5.462555646896362,
      "epoch": 2.0704217442578763,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045783718892570136,
      "loss": 5.2289,
      "mean_token_accuracy": 0.19695803970098497,
      "num_tokens": 41149515.0,
      "step": 18435
    },
    {
      "entropy": 5.480551862716675,
      "epoch": 2.0709833211658335,
      "grad_norm": 1.109375,
      "learning_rate": 0.000457813787011288,
      "loss": 5.2395,
      "mean_token_accuracy": 0.20340424925088882,
      "num_tokens": 41160023.0,
      "step": 18440
    },
    {
      "entropy": 5.3311539649963375,
      "epoch": 2.071544898073791,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004577903792760127,
      "loss": 5.015,
      "mean_token_accuracy": 0.2142372488975525,
      "num_tokens": 41170909.0,
      "step": 18445
    },
    {
      "entropy": 5.474853134155273,
      "epoch": 2.072106474981749,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004577669657206209,
      "loss": 5.2234,
      "mean_token_accuracy": 0.19887248575687408,
      "num_tokens": 41181960.0,
      "step": 18450
    },
    {
      "entropy": 5.537634372711182,
      "epoch": 2.0726680518897065,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004577435463458579,
      "loss": 5.2323,
      "mean_token_accuracy": 0.1956140100955963,
      "num_tokens": 41192292.0,
      "step": 18455
    },
    {
      "entropy": 5.509148359298706,
      "epoch": 2.0732296287976637,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004577201211524696,
      "loss": 5.1618,
      "mean_token_accuracy": 0.20363277643918992,
      "num_tokens": 41203826.0,
      "step": 18460
    },
    {
      "entropy": 5.395159101486206,
      "epoch": 2.0737912057056214,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004576966901412018,
      "loss": 5.1844,
      "mean_token_accuracy": 0.203907510638237,
      "num_tokens": 41215095.0,
      "step": 18465
    },
    {
      "entropy": 5.459643125534058,
      "epoch": 2.074352782613579,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045767325331280044,
      "loss": 5.0925,
      "mean_token_accuracy": 0.20108969360589982,
      "num_tokens": 41224953.0,
      "step": 18470
    },
    {
      "entropy": 5.464155530929565,
      "epoch": 2.0749143595215367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004576498106680119,
      "loss": 5.1939,
      "mean_token_accuracy": 0.19282402992248535,
      "num_tokens": 41235479.0,
      "step": 18475
    },
    {
      "entropy": 5.477154350280761,
      "epoch": 2.075475936429494,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004576263622075826,
      "loss": 5.1999,
      "mean_token_accuracy": 0.20027188062667847,
      "num_tokens": 41248520.0,
      "step": 18480
    },
    {
      "entropy": 5.565487384796143,
      "epoch": 2.0760375133374516,
      "grad_norm": 1.0625,
      "learning_rate": 0.000457602907932259,
      "loss": 5.2896,
      "mean_token_accuracy": 0.19452620595693587,
      "num_tokens": 41259112.0,
      "step": 18485
    },
    {
      "entropy": 5.431046628952027,
      "epoch": 2.0765990902454092,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045757944784278803,
      "loss": 5.1179,
      "mean_token_accuracy": 0.20431412160396575,
      "num_tokens": 41269554.0,
      "step": 18490
    },
    {
      "entropy": 5.419406127929688,
      "epoch": 2.0771606671533664,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004575559819399166,
      "loss": 5.1737,
      "mean_token_accuracy": 0.2000341922044754,
      "num_tokens": 41281339.0,
      "step": 18495
    },
    {
      "entropy": 5.378335285186767,
      "epoch": 2.077722244061324,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045753251022439195,
      "loss": 5.0698,
      "mean_token_accuracy": 0.21373364478349685,
      "num_tokens": 41291629.0,
      "step": 18500
    },
    {
      "entropy": 5.557219362258911,
      "epoch": 2.0782838209692818,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004575090326969614,
      "loss": 5.1964,
      "mean_token_accuracy": 0.20161436796188353,
      "num_tokens": 41302162.0,
      "step": 18505
    },
    {
      "entropy": 5.40969409942627,
      "epoch": 2.0788453978772394,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004574855493583723,
      "loss": 5.1543,
      "mean_token_accuracy": 0.20534624606370927,
      "num_tokens": 41313667.0,
      "step": 18510
    },
    {
      "entropy": 5.391064834594727,
      "epoch": 2.0794069747851966,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045746206020937276,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19790396988391876,
      "num_tokens": 41325063.0,
      "step": 18515
    },
    {
      "entropy": 5.434782218933106,
      "epoch": 2.0799685516931543,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045743856525071034,
      "loss": 5.1745,
      "mean_token_accuracy": 0.20320819318294525,
      "num_tokens": 41335974.0,
      "step": 18520
    },
    {
      "entropy": 5.448341941833496,
      "epoch": 2.080530128601112,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045741506448313327,
      "loss": 5.1315,
      "mean_token_accuracy": 0.20061599016189574,
      "num_tokens": 41347469.0,
      "step": 18525
    },
    {
      "entropy": 5.528663539886475,
      "epoch": 2.0810917055090696,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045739155790738975,
      "loss": 5.2057,
      "mean_token_accuracy": 0.20094305127859116,
      "num_tokens": 41358418.0,
      "step": 18530
    },
    {
      "entropy": 5.439347124099731,
      "epoch": 2.081653282417027,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045736804552422836,
      "loss": 5.1837,
      "mean_token_accuracy": 0.20304896384477616,
      "num_tokens": 41368859.0,
      "step": 18535
    },
    {
      "entropy": 5.453055143356323,
      "epoch": 2.0822148593249845,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004573445273343976,
      "loss": 5.2435,
      "mean_token_accuracy": 0.19476781487464906,
      "num_tokens": 41380929.0,
      "step": 18540
    },
    {
      "entropy": 5.4635162353515625,
      "epoch": 2.082776436232942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004573210033386464,
      "loss": 5.0651,
      "mean_token_accuracy": 0.2080380767583847,
      "num_tokens": 41392190.0,
      "step": 18545
    },
    {
      "entropy": 5.517323398590088,
      "epoch": 2.0833380131409,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004572974735377238,
      "loss": 5.1152,
      "mean_token_accuracy": 0.20934794694185258,
      "num_tokens": 41402646.0,
      "step": 18550
    },
    {
      "entropy": 5.43314905166626,
      "epoch": 2.083899590048857,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004572739379323788,
      "loss": 5.1509,
      "mean_token_accuracy": 0.2019234076142311,
      "num_tokens": 41415179.0,
      "step": 18555
    },
    {
      "entropy": 5.417903470993042,
      "epoch": 2.0844611669568147,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045725039652336105,
      "loss": 5.2,
      "mean_token_accuracy": 0.19754811972379685,
      "num_tokens": 41426695.0,
      "step": 18560
    },
    {
      "entropy": 5.39660415649414,
      "epoch": 2.0850227438647724,
      "grad_norm": 1.125,
      "learning_rate": 0.0004572268493114199,
      "loss": 5.1422,
      "mean_token_accuracy": 0.19991028308868408,
      "num_tokens": 41438078.0,
      "step": 18565
    },
    {
      "entropy": 5.49943060874939,
      "epoch": 2.08558432077273,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004572032962973052,
      "loss": 5.2063,
      "mean_token_accuracy": 0.1975611314177513,
      "num_tokens": 41448862.0,
      "step": 18570
    },
    {
      "entropy": 5.394349050521851,
      "epoch": 2.0861458976806873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045717973748176695,
      "loss": 4.9884,
      "mean_token_accuracy": 0.21039617210626602,
      "num_tokens": 41458528.0,
      "step": 18575
    },
    {
      "entropy": 5.361552572250366,
      "epoch": 2.086707474588645,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045715617286555516,
      "loss": 5.0953,
      "mean_token_accuracy": 0.20048311054706575,
      "num_tokens": 41469553.0,
      "step": 18580
    },
    {
      "entropy": 5.4895062923431395,
      "epoch": 2.0872690514966026,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045713260244942017,
      "loss": 5.2666,
      "mean_token_accuracy": 0.1961825355887413,
      "num_tokens": 41482327.0,
      "step": 18585
    },
    {
      "entropy": 5.423326110839843,
      "epoch": 2.08783062840456,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004571090262341125,
      "loss": 5.0215,
      "mean_token_accuracy": 0.21568716764450074,
      "num_tokens": 41493126.0,
      "step": 18590
    },
    {
      "entropy": 5.462135696411133,
      "epoch": 2.0883922053125175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045708544422038284,
      "loss": 5.2227,
      "mean_token_accuracy": 0.19605877697467805,
      "num_tokens": 41505597.0,
      "step": 18595
    },
    {
      "entropy": 5.472991895675659,
      "epoch": 2.088953782220475,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004570618564089819,
      "loss": 5.2494,
      "mean_token_accuracy": 0.1958896115422249,
      "num_tokens": 41517296.0,
      "step": 18600
    },
    {
      "entropy": 5.4867640972137455,
      "epoch": 2.089515359128433,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045703826280066095,
      "loss": 5.1345,
      "mean_token_accuracy": 0.20461816638708114,
      "num_tokens": 41529900.0,
      "step": 18605
    },
    {
      "entropy": 5.509304237365723,
      "epoch": 2.09007693603639,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004570146633961711,
      "loss": 5.1882,
      "mean_token_accuracy": 0.20411134660243987,
      "num_tokens": 41540641.0,
      "step": 18610
    },
    {
      "entropy": 5.430690717697144,
      "epoch": 2.0906385129443477,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045699105819626374,
      "loss": 5.1592,
      "mean_token_accuracy": 0.1952848806977272,
      "num_tokens": 41551335.0,
      "step": 18615
    },
    {
      "entropy": 5.386656427383423,
      "epoch": 2.0912000898523053,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004569674472016904,
      "loss": 5.1221,
      "mean_token_accuracy": 0.20962560176849365,
      "num_tokens": 41562265.0,
      "step": 18620
    },
    {
      "entropy": 5.4246316909790036,
      "epoch": 2.091761666760263,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004569438304132031,
      "loss": 5.047,
      "mean_token_accuracy": 0.20731533616781234,
      "num_tokens": 41574758.0,
      "step": 18625
    },
    {
      "entropy": 5.3818775653839115,
      "epoch": 2.09232324366822,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045692020783155376,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20591089278459548,
      "num_tokens": 41585007.0,
      "step": 18630
    },
    {
      "entropy": 5.353652048110962,
      "epoch": 2.092884820576178,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004568965794574943,
      "loss": 5.0601,
      "mean_token_accuracy": 0.20557635724544526,
      "num_tokens": 41596026.0,
      "step": 18635
    },
    {
      "entropy": 5.466334581375122,
      "epoch": 2.0934463974841355,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004568729452917772,
      "loss": 5.2112,
      "mean_token_accuracy": 0.20319215953350067,
      "num_tokens": 41608440.0,
      "step": 18640
    },
    {
      "entropy": 5.475106048583984,
      "epoch": 2.094007974392093,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045684930533515506,
      "loss": 5.0996,
      "mean_token_accuracy": 0.20157624930143356,
      "num_tokens": 41620467.0,
      "step": 18645
    },
    {
      "entropy": 5.37600622177124,
      "epoch": 2.0945695513000504,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045682565958838045,
      "loss": 5.1108,
      "mean_token_accuracy": 0.20145183801651,
      "num_tokens": 41630918.0,
      "step": 18650
    },
    {
      "entropy": 5.3887773036956785,
      "epoch": 2.095131128208008,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004568020080522063,
      "loss": 5.1062,
      "mean_token_accuracy": 0.19668225198984146,
      "num_tokens": 41641148.0,
      "step": 18655
    },
    {
      "entropy": 5.469299411773681,
      "epoch": 2.0956927051159657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004567783507273858,
      "loss": 5.1611,
      "mean_token_accuracy": 0.2015403240919113,
      "num_tokens": 41652148.0,
      "step": 18660
    },
    {
      "entropy": 5.399859428405762,
      "epoch": 2.0962542820239234,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045675468761467204,
      "loss": 5.1285,
      "mean_token_accuracy": 0.20415159165859223,
      "num_tokens": 41662275.0,
      "step": 18665
    },
    {
      "entropy": 5.472311973571777,
      "epoch": 2.0968158589318806,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045673101871481853,
      "loss": 5.1838,
      "mean_token_accuracy": 0.19498195499181747,
      "num_tokens": 41673141.0,
      "step": 18670
    },
    {
      "entropy": 5.562878036499024,
      "epoch": 2.0973774358398383,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004567073440285789,
      "loss": 5.2819,
      "mean_token_accuracy": 0.19360179752111434,
      "num_tokens": 41683473.0,
      "step": 18675
    },
    {
      "entropy": 5.323286437988282,
      "epoch": 2.097939012747796,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045668366355670697,
      "loss": 4.9456,
      "mean_token_accuracy": 0.20927121192216874,
      "num_tokens": 41693847.0,
      "step": 18680
    },
    {
      "entropy": 5.407052087783813,
      "epoch": 2.098500589655753,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004566599772999567,
      "loss": 5.1971,
      "mean_token_accuracy": 0.20385240912437438,
      "num_tokens": 41705810.0,
      "step": 18685
    },
    {
      "entropy": 5.428302955627442,
      "epoch": 2.099062166563711,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045663628525908237,
      "loss": 5.0532,
      "mean_token_accuracy": 0.20395218580961227,
      "num_tokens": 41717160.0,
      "step": 18690
    },
    {
      "entropy": 5.4835858821868895,
      "epoch": 2.0996237434716685,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004566125874348382,
      "loss": 5.2142,
      "mean_token_accuracy": 0.200162111222744,
      "num_tokens": 41728631.0,
      "step": 18695
    },
    {
      "entropy": 5.515122652053833,
      "epoch": 2.100185320379626,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045658888382797886,
      "loss": 5.2544,
      "mean_token_accuracy": 0.1960071235895157,
      "num_tokens": 41738933.0,
      "step": 18700
    },
    {
      "entropy": 5.539283609390258,
      "epoch": 2.1007468972875833,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045656517443925897,
      "loss": 5.2546,
      "mean_token_accuracy": 0.20130504220724105,
      "num_tokens": 41750386.0,
      "step": 18705
    },
    {
      "entropy": 5.420370292663574,
      "epoch": 2.101308474195541,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004565414592694336,
      "loss": 5.1691,
      "mean_token_accuracy": 0.20445092320442199,
      "num_tokens": 41760711.0,
      "step": 18710
    },
    {
      "entropy": 5.4341185092926025,
      "epoch": 2.1018700511034987,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004565177383192577,
      "loss": 5.2115,
      "mean_token_accuracy": 0.19421566128730774,
      "num_tokens": 41772147.0,
      "step": 18715
    },
    {
      "entropy": 5.489456415176392,
      "epoch": 2.1024316280114563,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045649401158948664,
      "loss": 5.0851,
      "mean_token_accuracy": 0.2066085547208786,
      "num_tokens": 41783265.0,
      "step": 18720
    },
    {
      "entropy": 5.4553502082824705,
      "epoch": 2.1029932049194136,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004564702790808758,
      "loss": 5.0709,
      "mean_token_accuracy": 0.20633191168308257,
      "num_tokens": 41793977.0,
      "step": 18725
    },
    {
      "entropy": 5.431276035308838,
      "epoch": 2.103554781827371,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004564465407941809,
      "loss": 5.0972,
      "mean_token_accuracy": 0.20670644640922547,
      "num_tokens": 41804790.0,
      "step": 18730
    },
    {
      "entropy": 5.326567792892456,
      "epoch": 2.104116358735329,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045642279673015787,
      "loss": 5.0885,
      "mean_token_accuracy": 0.19919336289167405,
      "num_tokens": 41815611.0,
      "step": 18735
    },
    {
      "entropy": 5.374368286132812,
      "epoch": 2.1046779356432865,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004563990468895625,
      "loss": 5.1267,
      "mean_token_accuracy": 0.19925709962844848,
      "num_tokens": 41826935.0,
      "step": 18740
    },
    {
      "entropy": 5.438326406478882,
      "epoch": 2.1052395125512438,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004563752912731512,
      "loss": 5.0905,
      "mean_token_accuracy": 0.20564910024404526,
      "num_tokens": 41838932.0,
      "step": 18745
    },
    {
      "entropy": 5.342984628677368,
      "epoch": 2.1058010894592014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004563515298816803,
      "loss": 5.1491,
      "mean_token_accuracy": 0.1996524453163147,
      "num_tokens": 41850771.0,
      "step": 18750
    },
    {
      "entropy": 5.364787626266479,
      "epoch": 2.106362666367159,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045632776271590624,
      "loss": 5.0641,
      "mean_token_accuracy": 0.20850711017847062,
      "num_tokens": 41861515.0,
      "step": 18755
    },
    {
      "entropy": 5.412940645217896,
      "epoch": 2.1069242432751167,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004563039897765859,
      "loss": 4.9982,
      "mean_token_accuracy": 0.21647325456142424,
      "num_tokens": 41872195.0,
      "step": 18760
    },
    {
      "entropy": 5.370642948150635,
      "epoch": 2.107485820183074,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004562802110644764,
      "loss": 5.0935,
      "mean_token_accuracy": 0.2083800658583641,
      "num_tokens": 41882496.0,
      "step": 18765
    },
    {
      "entropy": 5.51829423904419,
      "epoch": 2.1080473970910316,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045625642658033447,
      "loss": 5.3176,
      "mean_token_accuracy": 0.1927931562066078,
      "num_tokens": 41893078.0,
      "step": 18770
    },
    {
      "entropy": 5.539067697525025,
      "epoch": 2.1086089739989893,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045623263632491766,
      "loss": 5.1821,
      "mean_token_accuracy": 0.20135900229215623,
      "num_tokens": 41905062.0,
      "step": 18775
    },
    {
      "entropy": 5.409191608428955,
      "epoch": 2.1091705509069465,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045620884029898344,
      "loss": 5.1486,
      "mean_token_accuracy": 0.20454021692276,
      "num_tokens": 41916086.0,
      "step": 18780
    },
    {
      "entropy": 5.423175621032715,
      "epoch": 2.109732127814904,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045618503850328954,
      "loss": 5.199,
      "mean_token_accuracy": 0.2041832149028778,
      "num_tokens": 41925466.0,
      "step": 18785
    },
    {
      "entropy": 5.495348691940308,
      "epoch": 2.110293704722862,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004561612309385936,
      "loss": 5.2355,
      "mean_token_accuracy": 0.19579385071992875,
      "num_tokens": 41936567.0,
      "step": 18790
    },
    {
      "entropy": 5.391596174240112,
      "epoch": 2.1108552816308195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045613741760565385,
      "loss": 5.0818,
      "mean_token_accuracy": 0.2067500650882721,
      "num_tokens": 41947021.0,
      "step": 18795
    },
    {
      "entropy": 5.491127681732178,
      "epoch": 2.1114168585387767,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004561135985052285,
      "loss": 5.1489,
      "mean_token_accuracy": 0.19715440422296523,
      "num_tokens": 41957929.0,
      "step": 18800
    },
    {
      "entropy": 5.431910514831543,
      "epoch": 2.1119784354467344,
      "grad_norm": 1.125,
      "learning_rate": 0.000456089773638076,
      "loss": 5.0673,
      "mean_token_accuracy": 0.2119723841547966,
      "num_tokens": 41968677.0,
      "step": 18805
    },
    {
      "entropy": 5.430100631713867,
      "epoch": 2.112540012354692,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004560659430049548,
      "loss": 5.2284,
      "mean_token_accuracy": 0.200178100168705,
      "num_tokens": 41980330.0,
      "step": 18810
    },
    {
      "entropy": 5.422392892837524,
      "epoch": 2.1131015892626497,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004560421066066237,
      "loss": 5.1262,
      "mean_token_accuracy": 0.20078005194664,
      "num_tokens": 41991258.0,
      "step": 18815
    },
    {
      "entropy": 5.518226575851441,
      "epoch": 2.113663166170607,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045601826444384175,
      "loss": 5.2183,
      "mean_token_accuracy": 0.20122957080602646,
      "num_tokens": 42002679.0,
      "step": 18820
    },
    {
      "entropy": 5.434555959701538,
      "epoch": 2.1142247430785646,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045599441651736805,
      "loss": 5.1004,
      "mean_token_accuracy": 0.20533353090286255,
      "num_tokens": 42012845.0,
      "step": 18825
    },
    {
      "entropy": 5.458740615844727,
      "epoch": 2.1147863199865222,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045597056282796196,
      "loss": 5.1744,
      "mean_token_accuracy": 0.19988572001457214,
      "num_tokens": 42024461.0,
      "step": 18830
    },
    {
      "entropy": 5.472439956665039,
      "epoch": 2.11534789689448,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004559467033763829,
      "loss": 5.1641,
      "mean_token_accuracy": 0.19594511538743972,
      "num_tokens": 42036996.0,
      "step": 18835
    },
    {
      "entropy": 5.492838907241821,
      "epoch": 2.115909473802437,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004559228381633907,
      "loss": 5.2189,
      "mean_token_accuracy": 0.20225370973348616,
      "num_tokens": 42048731.0,
      "step": 18840
    },
    {
      "entropy": 5.508620548248291,
      "epoch": 2.1164710507103948,
      "grad_norm": 1.125,
      "learning_rate": 0.00045589896718974517,
      "loss": 5.1735,
      "mean_token_accuracy": 0.20715158581733703,
      "num_tokens": 42060396.0,
      "step": 18845
    },
    {
      "entropy": 5.34435806274414,
      "epoch": 2.1170326276183524,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004558750904562063,
      "loss": 5.065,
      "mean_token_accuracy": 0.2060753136873245,
      "num_tokens": 42072037.0,
      "step": 18850
    },
    {
      "entropy": 5.374680376052856,
      "epoch": 2.11759420452631,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004558512079635345,
      "loss": 5.0208,
      "mean_token_accuracy": 0.21232302635908126,
      "num_tokens": 42082577.0,
      "step": 18855
    },
    {
      "entropy": 5.351109647750855,
      "epoch": 2.1181557814342673,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004558273197124899,
      "loss": 5.048,
      "mean_token_accuracy": 0.21000571250915528,
      "num_tokens": 42093978.0,
      "step": 18860
    },
    {
      "entropy": 5.437092351913452,
      "epoch": 2.118717358342225,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045580342570383347,
      "loss": 5.2465,
      "mean_token_accuracy": 0.1936510145664215,
      "num_tokens": 42105773.0,
      "step": 18865
    },
    {
      "entropy": 5.434174871444702,
      "epoch": 2.1192789352501826,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004557795259383258,
      "loss": 5.1604,
      "mean_token_accuracy": 0.20186442285776138,
      "num_tokens": 42116232.0,
      "step": 18870
    },
    {
      "entropy": 5.438861465454101,
      "epoch": 2.11984051215814,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045575562041672794,
      "loss": 5.1665,
      "mean_token_accuracy": 0.20004648268222808,
      "num_tokens": 42127813.0,
      "step": 18875
    },
    {
      "entropy": 5.392910099029541,
      "epoch": 2.1204020890660975,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045573170913980094,
      "loss": 5.0934,
      "mean_token_accuracy": 0.20497957020998,
      "num_tokens": 42138054.0,
      "step": 18880
    },
    {
      "entropy": 5.48740520477295,
      "epoch": 2.120963665974055,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045570779210830627,
      "loss": 5.1723,
      "mean_token_accuracy": 0.1970529392361641,
      "num_tokens": 42149575.0,
      "step": 18885
    },
    {
      "entropy": 5.44726448059082,
      "epoch": 2.121525242882013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004556838693230054,
      "loss": 5.1766,
      "mean_token_accuracy": 0.2020699843764305,
      "num_tokens": 42160824.0,
      "step": 18890
    },
    {
      "entropy": 5.456457281112671,
      "epoch": 2.12208681978997,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045565994078466005,
      "loss": 5.1995,
      "mean_token_accuracy": 0.20040919482707978,
      "num_tokens": 42171088.0,
      "step": 18895
    },
    {
      "entropy": 5.397271776199341,
      "epoch": 2.1226483966979277,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045563600649403213,
      "loss": 5.1521,
      "mean_token_accuracy": 0.2001300647854805,
      "num_tokens": 42182112.0,
      "step": 18900
    },
    {
      "entropy": 5.45868763923645,
      "epoch": 2.1232099736058854,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004556120664518838,
      "loss": 5.0438,
      "mean_token_accuracy": 0.21389970928430557,
      "num_tokens": 42193563.0,
      "step": 18905
    },
    {
      "entropy": 5.448945569992065,
      "epoch": 2.123771550513843,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004555881206589771,
      "loss": 5.1176,
      "mean_token_accuracy": 0.20248157680034637,
      "num_tokens": 42204891.0,
      "step": 18910
    },
    {
      "entropy": 5.459609270095825,
      "epoch": 2.1243331274218002,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004555641691160746,
      "loss": 5.1825,
      "mean_token_accuracy": 0.20771038085222243,
      "num_tokens": 42217226.0,
      "step": 18915
    },
    {
      "entropy": 5.480522537231446,
      "epoch": 2.124894704329758,
      "grad_norm": 1.15625,
      "learning_rate": 0.000455540211823939,
      "loss": 5.1579,
      "mean_token_accuracy": 0.19973548352718354,
      "num_tokens": 42228134.0,
      "step": 18920
    },
    {
      "entropy": 5.466844320297241,
      "epoch": 2.1254562812377156,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045551624878333296,
      "loss": 5.1708,
      "mean_token_accuracy": 0.19729188978672027,
      "num_tokens": 42239868.0,
      "step": 18925
    },
    {
      "entropy": 5.501995801925659,
      "epoch": 2.1260178581456732,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045549227999501957,
      "loss": 5.2359,
      "mean_token_accuracy": 0.19998051822185517,
      "num_tokens": 42250297.0,
      "step": 18930
    },
    {
      "entropy": 5.418631935119629,
      "epoch": 2.1265794350536305,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045546830545976203,
      "loss": 4.9213,
      "mean_token_accuracy": 0.21900530755519867,
      "num_tokens": 42261263.0,
      "step": 18935
    },
    {
      "entropy": 5.44773416519165,
      "epoch": 2.127141011961588,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045544432517832356,
      "loss": 5.1931,
      "mean_token_accuracy": 0.20379654169082642,
      "num_tokens": 42272357.0,
      "step": 18940
    },
    {
      "entropy": 5.463847923278808,
      "epoch": 2.1277025888695458,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045542033915146787,
      "loss": 5.2276,
      "mean_token_accuracy": 0.19887278974056244,
      "num_tokens": 42283630.0,
      "step": 18945
    },
    {
      "entropy": 5.468612623214722,
      "epoch": 2.1282641657775034,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045539634737995857,
      "loss": 5.2215,
      "mean_token_accuracy": 0.19014732539653778,
      "num_tokens": 42294981.0,
      "step": 18950
    },
    {
      "entropy": 5.440510082244873,
      "epoch": 2.1288257426854607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045537234986455964,
      "loss": 5.1128,
      "mean_token_accuracy": 0.20265614837408066,
      "num_tokens": 42304877.0,
      "step": 18955
    },
    {
      "entropy": 5.449815654754639,
      "epoch": 2.1293873195934183,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004553483466060351,
      "loss": 5.2044,
      "mean_token_accuracy": 0.19880924224853516,
      "num_tokens": 42316325.0,
      "step": 18960
    },
    {
      "entropy": 5.424876499176025,
      "epoch": 2.129948896501376,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045532433760514925,
      "loss": 5.1529,
      "mean_token_accuracy": 0.1977984607219696,
      "num_tokens": 42327437.0,
      "step": 18965
    },
    {
      "entropy": 5.327878475189209,
      "epoch": 2.130510473409333,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004553003228626665,
      "loss": 5.0563,
      "mean_token_accuracy": 0.2110733762383461,
      "num_tokens": 42337742.0,
      "step": 18970
    },
    {
      "entropy": 5.424053001403808,
      "epoch": 2.131072050317291,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004552763023793517,
      "loss": 5.1295,
      "mean_token_accuracy": 0.2034108117222786,
      "num_tokens": 42348511.0,
      "step": 18975
    },
    {
      "entropy": 5.41993522644043,
      "epoch": 2.1316336272252485,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004552522761559694,
      "loss": 5.1041,
      "mean_token_accuracy": 0.2035752847790718,
      "num_tokens": 42358977.0,
      "step": 18980
    },
    {
      "entropy": 5.47494535446167,
      "epoch": 2.132195204133206,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045522824419328475,
      "loss": 5.2033,
      "mean_token_accuracy": 0.2011194109916687,
      "num_tokens": 42371609.0,
      "step": 18985
    },
    {
      "entropy": 5.487996816635132,
      "epoch": 2.1327567810411634,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045520420649206283,
      "loss": 5.1523,
      "mean_token_accuracy": 0.20524742007255553,
      "num_tokens": 42383150.0,
      "step": 18990
    },
    {
      "entropy": 5.456073999404907,
      "epoch": 2.133318357949121,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004551801630530692,
      "loss": 5.0887,
      "mean_token_accuracy": 0.20199889689683914,
      "num_tokens": 42395150.0,
      "step": 18995
    },
    {
      "entropy": 5.459486722946167,
      "epoch": 2.1338799348570787,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004551561138770693,
      "loss": 5.1704,
      "mean_token_accuracy": 0.19895248115062714,
      "num_tokens": 42406348.0,
      "step": 19000
    },
    {
      "entropy": 5.495809078216553,
      "epoch": 2.1344415117650364,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045513205896482875,
      "loss": 5.1709,
      "mean_token_accuracy": 0.1946457713842392,
      "num_tokens": 42417183.0,
      "step": 19005
    },
    {
      "entropy": 5.447253894805908,
      "epoch": 2.1350030886729936,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045510799831711366,
      "loss": 5.1003,
      "mean_token_accuracy": 0.21330567449331284,
      "num_tokens": 42427805.0,
      "step": 19010
    },
    {
      "entropy": 5.44368577003479,
      "epoch": 2.1355646655809513,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045508393193469,
      "loss": 5.107,
      "mean_token_accuracy": 0.19952069073915482,
      "num_tokens": 42439521.0,
      "step": 19015
    },
    {
      "entropy": 5.526159000396729,
      "epoch": 2.136126242488909,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004550598598183241,
      "loss": 5.2336,
      "mean_token_accuracy": 0.19805433452129365,
      "num_tokens": 42451201.0,
      "step": 19020
    },
    {
      "entropy": 5.472752714157105,
      "epoch": 2.1366878193968666,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004550357819687825,
      "loss": 5.121,
      "mean_token_accuracy": 0.2027232512831688,
      "num_tokens": 42462315.0,
      "step": 19025
    },
    {
      "entropy": 5.591591310501099,
      "epoch": 2.137249396304824,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004550116983868317,
      "loss": 5.319,
      "mean_token_accuracy": 0.18874753266572952,
      "num_tokens": 42473993.0,
      "step": 19030
    },
    {
      "entropy": 5.590986633300782,
      "epoch": 2.1378109732127815,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004549876090732386,
      "loss": 5.3004,
      "mean_token_accuracy": 0.19695394486188889,
      "num_tokens": 42484307.0,
      "step": 19035
    },
    {
      "entropy": 5.490777206420899,
      "epoch": 2.138372550120739,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004549635140287703,
      "loss": 5.1978,
      "mean_token_accuracy": 0.19738707989454268,
      "num_tokens": 42497305.0,
      "step": 19040
    },
    {
      "entropy": 5.482597923278808,
      "epoch": 2.138934127028697,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045493941325419385,
      "loss": 5.1918,
      "mean_token_accuracy": 0.1993209093809128,
      "num_tokens": 42508307.0,
      "step": 19045
    },
    {
      "entropy": 5.511053085327148,
      "epoch": 2.139495703936654,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004549153067502768,
      "loss": 5.2602,
      "mean_token_accuracy": 0.19159839451313018,
      "num_tokens": 42519946.0,
      "step": 19050
    },
    {
      "entropy": 5.40810375213623,
      "epoch": 2.1400572808446117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045489119451778645,
      "loss": 5.0389,
      "mean_token_accuracy": 0.20282476991415024,
      "num_tokens": 42530737.0,
      "step": 19055
    },
    {
      "entropy": 5.332473802566528,
      "epoch": 2.1406188577525693,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004548670765574908,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20721573382616043,
      "num_tokens": 42542184.0,
      "step": 19060
    },
    {
      "entropy": 5.4416738986969,
      "epoch": 2.1411804346605265,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045484295287015757,
      "loss": 5.1605,
      "mean_token_accuracy": 0.19836343228816986,
      "num_tokens": 42553816.0,
      "step": 19065
    },
    {
      "entropy": 5.536497449874878,
      "epoch": 2.141742011568484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004548188234565551,
      "loss": 5.1809,
      "mean_token_accuracy": 0.2028401955962181,
      "num_tokens": 42564685.0,
      "step": 19070
    },
    {
      "entropy": 5.420281267166137,
      "epoch": 2.142303588476442,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045479468831745144,
      "loss": 5.1138,
      "mean_token_accuracy": 0.19693809300661086,
      "num_tokens": 42574864.0,
      "step": 19075
    },
    {
      "entropy": 5.410591459274292,
      "epoch": 2.1428651653843995,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004547705474536152,
      "loss": 5.1006,
      "mean_token_accuracy": 0.19923099428415297,
      "num_tokens": 42585214.0,
      "step": 19080
    },
    {
      "entropy": 5.495586776733399,
      "epoch": 2.1434267422923567,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045474640086581506,
      "loss": 5.2265,
      "mean_token_accuracy": 0.19812824130058287,
      "num_tokens": 42596017.0,
      "step": 19085
    },
    {
      "entropy": 5.453445863723755,
      "epoch": 2.1439883192003144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045472224855481975,
      "loss": 5.12,
      "mean_token_accuracy": 0.19889673888683318,
      "num_tokens": 42607903.0,
      "step": 19090
    },
    {
      "entropy": 5.411392307281494,
      "epoch": 2.144549896108272,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045469809052139844,
      "loss": 5.1341,
      "mean_token_accuracy": 0.20624203085899354,
      "num_tokens": 42618690.0,
      "step": 19095
    },
    {
      "entropy": 5.427911186218262,
      "epoch": 2.1451114730162297,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045467392676632016,
      "loss": 5.1055,
      "mean_token_accuracy": 0.2038321852684021,
      "num_tokens": 42629655.0,
      "step": 19100
    },
    {
      "entropy": 5.436651659011841,
      "epoch": 2.145673049924187,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004546497572903543,
      "loss": 5.2117,
      "mean_token_accuracy": 0.20069251209497452,
      "num_tokens": 42641214.0,
      "step": 19105
    },
    {
      "entropy": 5.506593036651611,
      "epoch": 2.1462346268321446,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004546255820942706,
      "loss": 5.2505,
      "mean_token_accuracy": 0.19528359025716782,
      "num_tokens": 42653829.0,
      "step": 19110
    },
    {
      "entropy": 5.501069784164429,
      "epoch": 2.1467962037401023,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004546014011788386,
      "loss": 5.1765,
      "mean_token_accuracy": 0.1994260460138321,
      "num_tokens": 42665245.0,
      "step": 19115
    },
    {
      "entropy": 5.4487401962280275,
      "epoch": 2.14735778064806,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004545772145448284,
      "loss": 5.1172,
      "mean_token_accuracy": 0.2001232087612152,
      "num_tokens": 42676409.0,
      "step": 19120
    },
    {
      "entropy": 5.3757185459136965,
      "epoch": 2.147919357556017,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045455302219301,
      "loss": 4.9753,
      "mean_token_accuracy": 0.21238219887018203,
      "num_tokens": 42686586.0,
      "step": 19125
    },
    {
      "entropy": 5.359950590133667,
      "epoch": 2.148480934463975,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004545288241241538,
      "loss": 5.0593,
      "mean_token_accuracy": 0.205304753780365,
      "num_tokens": 42696597.0,
      "step": 19130
    },
    {
      "entropy": 5.510161638259888,
      "epoch": 2.1490425113719325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004545046203390302,
      "loss": 5.2398,
      "mean_token_accuracy": 0.20513799488544465,
      "num_tokens": 42707447.0,
      "step": 19135
    },
    {
      "entropy": 5.569417095184326,
      "epoch": 2.14960408827989,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045448041083840973,
      "loss": 5.2434,
      "mean_token_accuracy": 0.19274359047412873,
      "num_tokens": 42717533.0,
      "step": 19140
    },
    {
      "entropy": 5.52343339920044,
      "epoch": 2.1501656651878474,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045445619562306354,
      "loss": 5.2365,
      "mean_token_accuracy": 0.19718028604984283,
      "num_tokens": 42730563.0,
      "step": 19145
    },
    {
      "entropy": 5.430177402496338,
      "epoch": 2.150727242095805,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045443197469376243,
      "loss": 5.1106,
      "mean_token_accuracy": 0.20820065587759018,
      "num_tokens": 42740301.0,
      "step": 19150
    },
    {
      "entropy": 5.434264612197876,
      "epoch": 2.1512888190037627,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004544077480512776,
      "loss": 5.1681,
      "mean_token_accuracy": 0.2004183828830719,
      "num_tokens": 42752296.0,
      "step": 19155
    },
    {
      "entropy": 5.372303581237793,
      "epoch": 2.15185039591172,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045438351569638055,
      "loss": 5.0297,
      "mean_token_accuracy": 0.2091466799378395,
      "num_tokens": 42762530.0,
      "step": 19160
    },
    {
      "entropy": 5.421622610092163,
      "epoch": 2.1524119728196776,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045435927762984277,
      "loss": 5.0937,
      "mean_token_accuracy": 0.20492661148309707,
      "num_tokens": 42773059.0,
      "step": 19165
    },
    {
      "entropy": 5.432879638671875,
      "epoch": 2.152973549727635,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000454335033852436,
      "loss": 5.1195,
      "mean_token_accuracy": 0.20628607273101807,
      "num_tokens": 42783832.0,
      "step": 19170
    },
    {
      "entropy": 5.458415174484253,
      "epoch": 2.153535126635593,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004543107843649322,
      "loss": 5.2261,
      "mean_token_accuracy": 0.20346362590789796,
      "num_tokens": 42794565.0,
      "step": 19175
    },
    {
      "entropy": 5.5305615901947025,
      "epoch": 2.15409670354355,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004542865291681035,
      "loss": 5.2683,
      "mean_token_accuracy": 0.19414251148700715,
      "num_tokens": 42805536.0,
      "step": 19180
    },
    {
      "entropy": 5.45842490196228,
      "epoch": 2.1546582804515078,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045426226826272213,
      "loss": 5.1709,
      "mean_token_accuracy": 0.2008275121450424,
      "num_tokens": 42816718.0,
      "step": 19185
    },
    {
      "entropy": 5.465261745452881,
      "epoch": 2.1552198573594654,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004542380016495607,
      "loss": 5.1533,
      "mean_token_accuracy": 0.20271989703178406,
      "num_tokens": 42830183.0,
      "step": 19190
    },
    {
      "entropy": 5.423596811294556,
      "epoch": 2.155781434267423,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045421372932939177,
      "loss": 5.1564,
      "mean_token_accuracy": 0.20599999725818635,
      "num_tokens": 42841241.0,
      "step": 19195
    },
    {
      "entropy": 5.577798175811767,
      "epoch": 2.1563430111753803,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004541894513029881,
      "loss": 5.2192,
      "mean_token_accuracy": 0.20121305137872697,
      "num_tokens": 42852775.0,
      "step": 19200
    },
    {
      "entropy": 5.624221086502075,
      "epoch": 2.156904588083338,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045416516757112286,
      "loss": 5.3359,
      "mean_token_accuracy": 0.19158112108707429,
      "num_tokens": 42863947.0,
      "step": 19205
    },
    {
      "entropy": 5.430298233032227,
      "epoch": 2.1574661649912956,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045414087813456925,
      "loss": 5.184,
      "mean_token_accuracy": 0.20596070289611818,
      "num_tokens": 42874794.0,
      "step": 19210
    },
    {
      "entropy": 5.457150936126709,
      "epoch": 2.1580277418992533,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004541165829941006,
      "loss": 5.1634,
      "mean_token_accuracy": 0.201410935819149,
      "num_tokens": 42886038.0,
      "step": 19215
    },
    {
      "entropy": 5.454669713973999,
      "epoch": 2.1585893188072105,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004540922821504904,
      "loss": 5.1661,
      "mean_token_accuracy": 0.19576012194156647,
      "num_tokens": 42897243.0,
      "step": 19220
    },
    {
      "entropy": 5.437304878234864,
      "epoch": 2.159150895715168,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004540679756045125,
      "loss": 5.0714,
      "mean_token_accuracy": 0.20804495215415955,
      "num_tokens": 42908802.0,
      "step": 19225
    },
    {
      "entropy": 5.512308835983276,
      "epoch": 2.159712472623126,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004540436633569407,
      "loss": 5.2601,
      "mean_token_accuracy": 0.19065648466348648,
      "num_tokens": 42921156.0,
      "step": 19230
    },
    {
      "entropy": 5.3409342765808105,
      "epoch": 2.1602740495310835,
      "grad_norm": 1.125,
      "learning_rate": 0.00045401934540854937,
      "loss": 5.0646,
      "mean_token_accuracy": 0.20856190621852874,
      "num_tokens": 42931336.0,
      "step": 19235
    },
    {
      "entropy": 5.44011697769165,
      "epoch": 2.1608356264390407,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004539950217601127,
      "loss": 5.1567,
      "mean_token_accuracy": 0.20694223344326018,
      "num_tokens": 42942312.0,
      "step": 19240
    },
    {
      "entropy": 5.463801765441895,
      "epoch": 2.1613972033469984,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045397069241240497,
      "loss": 5.215,
      "mean_token_accuracy": 0.20644889920949935,
      "num_tokens": 42953429.0,
      "step": 19245
    },
    {
      "entropy": 5.376381969451904,
      "epoch": 2.161958780254956,
      "grad_norm": 1.171875,
      "learning_rate": 0.000453946357366201,
      "loss": 5.0667,
      "mean_token_accuracy": 0.20865806192159653,
      "num_tokens": 42963354.0,
      "step": 19250
    },
    {
      "entropy": 5.529055786132813,
      "epoch": 2.1625203571629132,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045392201662227565,
      "loss": 5.221,
      "mean_token_accuracy": 0.20140518099069596,
      "num_tokens": 42974643.0,
      "step": 19255
    },
    {
      "entropy": 5.463166189193726,
      "epoch": 2.163081934070871,
      "grad_norm": 1.125,
      "learning_rate": 0.00045389767018140395,
      "loss": 5.1989,
      "mean_token_accuracy": 0.2009876325726509,
      "num_tokens": 42985821.0,
      "step": 19260
    },
    {
      "entropy": 5.470652151107788,
      "epoch": 2.1636435109788286,
      "grad_norm": 1.0,
      "learning_rate": 0.00045387331804436096,
      "loss": 5.2401,
      "mean_token_accuracy": 0.19452736228704454,
      "num_tokens": 42998258.0,
      "step": 19265
    },
    {
      "entropy": 5.449978971481324,
      "epoch": 2.1642050878867862,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004538489602119222,
      "loss": 5.125,
      "mean_token_accuracy": 0.20073011368513108,
      "num_tokens": 43009117.0,
      "step": 19270
    },
    {
      "entropy": 5.408582592010498,
      "epoch": 2.1647666647947434,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004538245966848632,
      "loss": 5.1316,
      "mean_token_accuracy": 0.2023191884160042,
      "num_tokens": 43020437.0,
      "step": 19275
    },
    {
      "entropy": 5.489286994934082,
      "epoch": 2.165328241702701,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045380022746395953,
      "loss": 5.1882,
      "mean_token_accuracy": 0.19996050745248795,
      "num_tokens": 43031722.0,
      "step": 19280
    },
    {
      "entropy": 5.434236526489258,
      "epoch": 2.1658898186106588,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045377585254998743,
      "loss": 5.1086,
      "mean_token_accuracy": 0.21206972002983093,
      "num_tokens": 43042957.0,
      "step": 19285
    },
    {
      "entropy": 5.409184885025025,
      "epoch": 2.1664513955186164,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004537514719437228,
      "loss": 5.1764,
      "mean_token_accuracy": 0.19841236174106597,
      "num_tokens": 43054086.0,
      "step": 19290
    },
    {
      "entropy": 5.557629632949829,
      "epoch": 2.1670129724265736,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045372708564594196,
      "loss": 5.277,
      "mean_token_accuracy": 0.19409868568181993,
      "num_tokens": 43066259.0,
      "step": 19295
    },
    {
      "entropy": 5.37485933303833,
      "epoch": 2.1675745493345313,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045370269365742143,
      "loss": 5.0038,
      "mean_token_accuracy": 0.2103622615337372,
      "num_tokens": 43076837.0,
      "step": 19300
    },
    {
      "entropy": 5.363848924636841,
      "epoch": 2.168136126242489,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045367829597893776,
      "loss": 5.0653,
      "mean_token_accuracy": 0.20678746104240417,
      "num_tokens": 43087802.0,
      "step": 19305
    },
    {
      "entropy": 5.423171138763427,
      "epoch": 2.1686977031504466,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045365389261126793,
      "loss": 5.0939,
      "mean_token_accuracy": 0.20238826423883438,
      "num_tokens": 43097562.0,
      "step": 19310
    },
    {
      "entropy": 5.345750617980957,
      "epoch": 2.169259280058404,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045362948355518883,
      "loss": 5.0839,
      "mean_token_accuracy": 0.20817989110946655,
      "num_tokens": 43107973.0,
      "step": 19315
    },
    {
      "entropy": 5.322268772125244,
      "epoch": 2.1698208569663615,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004536050688114777,
      "loss": 5.0374,
      "mean_token_accuracy": 0.2148093357682228,
      "num_tokens": 43118923.0,
      "step": 19320
    },
    {
      "entropy": 5.391568899154663,
      "epoch": 2.170382433874319,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045358064838091187,
      "loss": 5.1144,
      "mean_token_accuracy": 0.20451484322547914,
      "num_tokens": 43130904.0,
      "step": 19325
    },
    {
      "entropy": 5.472601747512817,
      "epoch": 2.170944010782277,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045355622226426903,
      "loss": 5.1179,
      "mean_token_accuracy": 0.1967015326023102,
      "num_tokens": 43142293.0,
      "step": 19330
    },
    {
      "entropy": 5.474063920974731,
      "epoch": 2.171505587690234,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045353179046232674,
      "loss": 5.1882,
      "mean_token_accuracy": 0.20847147852182388,
      "num_tokens": 43154340.0,
      "step": 19335
    },
    {
      "entropy": 5.421504926681519,
      "epoch": 2.1720671645981917,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000453507352975863,
      "loss": 5.1276,
      "mean_token_accuracy": 0.2093738466501236,
      "num_tokens": 43165924.0,
      "step": 19340
    },
    {
      "entropy": 5.33529577255249,
      "epoch": 2.1726287415061494,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045348290980565595,
      "loss": 5.0823,
      "mean_token_accuracy": 0.2024005651473999,
      "num_tokens": 43177394.0,
      "step": 19345
    },
    {
      "entropy": 5.402228498458863,
      "epoch": 2.1731903184141066,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045345846095248376,
      "loss": 5.0617,
      "mean_token_accuracy": 0.2059154614806175,
      "num_tokens": 43188043.0,
      "step": 19350
    },
    {
      "entropy": 5.49878249168396,
      "epoch": 2.1737518953220643,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045343400641712503,
      "loss": 5.2633,
      "mean_token_accuracy": 0.1958798125386238,
      "num_tokens": 43199466.0,
      "step": 19355
    },
    {
      "entropy": 5.389993667602539,
      "epoch": 2.174313472230022,
      "grad_norm": 1.25,
      "learning_rate": 0.00045340954620035827,
      "loss": 5.0844,
      "mean_token_accuracy": 0.20503222346305847,
      "num_tokens": 43210673.0,
      "step": 19360
    },
    {
      "entropy": 5.459617710113525,
      "epoch": 2.1748750491379796,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045338508030296246,
      "loss": 5.1768,
      "mean_token_accuracy": 0.20423915684223176,
      "num_tokens": 43221263.0,
      "step": 19365
    },
    {
      "entropy": 5.477347183227539,
      "epoch": 2.175436626045937,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045336060872571646,
      "loss": 5.1782,
      "mean_token_accuracy": 0.20773029774427415,
      "num_tokens": 43232590.0,
      "step": 19370
    },
    {
      "entropy": 5.52472882270813,
      "epoch": 2.1759982029538945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004533361314693994,
      "loss": 5.2094,
      "mean_token_accuracy": 0.19878347367048263,
      "num_tokens": 43243693.0,
      "step": 19375
    },
    {
      "entropy": 5.485237693786621,
      "epoch": 2.176559779861852,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045331164853479084,
      "loss": 5.1264,
      "mean_token_accuracy": 0.2032419040799141,
      "num_tokens": 43254526.0,
      "step": 19380
    },
    {
      "entropy": 5.555027914047241,
      "epoch": 2.17712135676981,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045328715992267024,
      "loss": 5.2165,
      "mean_token_accuracy": 0.20115191340446473,
      "num_tokens": 43264490.0,
      "step": 19385
    },
    {
      "entropy": 5.448733234405518,
      "epoch": 2.177682933677767,
      "grad_norm": 1.125,
      "learning_rate": 0.00045326266563381734,
      "loss": 5.1307,
      "mean_token_accuracy": 0.20167769938707353,
      "num_tokens": 43275012.0,
      "step": 19390
    },
    {
      "entropy": 5.423792457580566,
      "epoch": 2.1782445105857247,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045323816566901205,
      "loss": 5.1271,
      "mean_token_accuracy": 0.20693261325359344,
      "num_tokens": 43285194.0,
      "step": 19395
    },
    {
      "entropy": 5.436718845367432,
      "epoch": 2.1788060874936823,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004532136600290343,
      "loss": 5.1946,
      "mean_token_accuracy": 0.20652735829353333,
      "num_tokens": 43298159.0,
      "step": 19400
    },
    {
      "entropy": 5.476078653335572,
      "epoch": 2.17936766440164,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004531891487146646,
      "loss": 5.2199,
      "mean_token_accuracy": 0.20245183110237122,
      "num_tokens": 43309664.0,
      "step": 19405
    },
    {
      "entropy": 5.466586685180664,
      "epoch": 2.179929241309597,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004531646317266833,
      "loss": 5.1953,
      "mean_token_accuracy": 0.19685198366641998,
      "num_tokens": 43320399.0,
      "step": 19410
    },
    {
      "entropy": 5.406252002716064,
      "epoch": 2.180490818217555,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045314010906587096,
      "loss": 5.0155,
      "mean_token_accuracy": 0.20388790518045424,
      "num_tokens": 43331510.0,
      "step": 19415
    },
    {
      "entropy": 5.513053512573242,
      "epoch": 2.1810523951255125,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045311558073300853,
      "loss": 5.2401,
      "mean_token_accuracy": 0.20349526554346084,
      "num_tokens": 43342844.0,
      "step": 19420
    },
    {
      "entropy": 5.371338891983032,
      "epoch": 2.18161397203347,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004530910467288769,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19907493740320206,
      "num_tokens": 43352556.0,
      "step": 19425
    },
    {
      "entropy": 5.382122325897217,
      "epoch": 2.1821755489414274,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045306650705425727,
      "loss": 5.0782,
      "mean_token_accuracy": 0.20880690962076187,
      "num_tokens": 43363357.0,
      "step": 19430
    },
    {
      "entropy": 5.4958654880523685,
      "epoch": 2.182737125849385,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045304196170993093,
      "loss": 5.1447,
      "mean_token_accuracy": 0.20085732787847518,
      "num_tokens": 43374697.0,
      "step": 19435
    },
    {
      "entropy": 5.522832632064819,
      "epoch": 2.1832987027573427,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004530174106966795,
      "loss": 5.2239,
      "mean_token_accuracy": 0.19652022123336793,
      "num_tokens": 43386826.0,
      "step": 19440
    },
    {
      "entropy": 5.41835298538208,
      "epoch": 2.1838602796653,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045299285401528475,
      "loss": 5.1823,
      "mean_token_accuracy": 0.19774671792984008,
      "num_tokens": 43398352.0,
      "step": 19445
    },
    {
      "entropy": 5.46236629486084,
      "epoch": 2.1844218565732576,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004529682916665285,
      "loss": 5.1447,
      "mean_token_accuracy": 0.20672448724508286,
      "num_tokens": 43409509.0,
      "step": 19450
    },
    {
      "entropy": 5.403712463378906,
      "epoch": 2.1849834334812153,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045294372365119276,
      "loss": 5.0949,
      "mean_token_accuracy": 0.20152269005775453,
      "num_tokens": 43420195.0,
      "step": 19455
    },
    {
      "entropy": 5.454810953140258,
      "epoch": 2.185545010389173,
      "grad_norm": 1.25,
      "learning_rate": 0.0004529191499700598,
      "loss": 5.1976,
      "mean_token_accuracy": 0.20226840525865555,
      "num_tokens": 43430599.0,
      "step": 19460
    },
    {
      "entropy": 5.382679986953735,
      "epoch": 2.18610658729713,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004528945706239122,
      "loss": 5.0317,
      "mean_token_accuracy": 0.2047211453318596,
      "num_tokens": 43442212.0,
      "step": 19465
    },
    {
      "entropy": 5.39938530921936,
      "epoch": 2.186668164205088,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045286998561353236,
      "loss": 5.0954,
      "mean_token_accuracy": 0.20168321579694748,
      "num_tokens": 43452631.0,
      "step": 19470
    },
    {
      "entropy": 5.393381214141845,
      "epoch": 2.1872297411130455,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004528453949397033,
      "loss": 5.135,
      "mean_token_accuracy": 0.2065647765994072,
      "num_tokens": 43463323.0,
      "step": 19475
    },
    {
      "entropy": 5.411081790924072,
      "epoch": 2.187791318021003,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045282079860320786,
      "loss": 5.1906,
      "mean_token_accuracy": 0.19413543343544007,
      "num_tokens": 43474451.0,
      "step": 19480
    },
    {
      "entropy": 5.40665373802185,
      "epoch": 2.1883528949289603,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045279619660482925,
      "loss": 5.049,
      "mean_token_accuracy": 0.2050679862499237,
      "num_tokens": 43485948.0,
      "step": 19485
    },
    {
      "entropy": 5.513670444488525,
      "epoch": 2.188914471836918,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045277158894535075,
      "loss": 5.1395,
      "mean_token_accuracy": 0.2081505224108696,
      "num_tokens": 43498955.0,
      "step": 19490
    },
    {
      "entropy": 5.462107992172241,
      "epoch": 2.1894760487448757,
      "grad_norm": 1.125,
      "learning_rate": 0.00045274697562555584,
      "loss": 5.0986,
      "mean_token_accuracy": 0.21066556572914125,
      "num_tokens": 43509416.0,
      "step": 19495
    },
    {
      "entropy": 5.3991227626800535,
      "epoch": 2.1900376256528333,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004527223566462284,
      "loss": 5.1346,
      "mean_token_accuracy": 0.2058756023645401,
      "num_tokens": 43520387.0,
      "step": 19500
    },
    {
      "entropy": 5.378425455093383,
      "epoch": 2.1905992025607905,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004526977320081522,
      "loss": 5.1408,
      "mean_token_accuracy": 0.2011588528752327,
      "num_tokens": 43532006.0,
      "step": 19505
    },
    {
      "entropy": 5.381918573379517,
      "epoch": 2.191160779468748,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004526731017121111,
      "loss": 5.0554,
      "mean_token_accuracy": 0.2088074192404747,
      "num_tokens": 43542303.0,
      "step": 19510
    },
    {
      "entropy": 5.509841871261597,
      "epoch": 2.191722356376706,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045264846575888973,
      "loss": 5.2338,
      "mean_token_accuracy": 0.19600335359573365,
      "num_tokens": 43553711.0,
      "step": 19515
    },
    {
      "entropy": 5.386949729919434,
      "epoch": 2.1922839332846635,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045262382414927235,
      "loss": 5.087,
      "mean_token_accuracy": 0.20398693531751633,
      "num_tokens": 43564488.0,
      "step": 19520
    },
    {
      "entropy": 5.415289497375488,
      "epoch": 2.1928455101926208,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045259917688404336,
      "loss": 5.1071,
      "mean_token_accuracy": 0.20185879468917847,
      "num_tokens": 43575414.0,
      "step": 19525
    },
    {
      "entropy": 5.369929265975952,
      "epoch": 2.1934070871005784,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004525745239639877,
      "loss": 5.0815,
      "mean_token_accuracy": 0.20495269149541856,
      "num_tokens": 43586789.0,
      "step": 19530
    },
    {
      "entropy": 5.4115245819091795,
      "epoch": 2.193968664008536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045254986538989045,
      "loss": 5.0789,
      "mean_token_accuracy": 0.2147873729467392,
      "num_tokens": 43597423.0,
      "step": 19535
    },
    {
      "entropy": 5.412087249755859,
      "epoch": 2.1945302409164933,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004525252011625366,
      "loss": 5.0939,
      "mean_token_accuracy": 0.20439665019512177,
      "num_tokens": 43606703.0,
      "step": 19540
    },
    {
      "entropy": 5.481432294845581,
      "epoch": 2.195091817824451,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004525005312827114,
      "loss": 5.2359,
      "mean_token_accuracy": 0.19984448403120042,
      "num_tokens": 43617869.0,
      "step": 19545
    },
    {
      "entropy": 5.532707166671753,
      "epoch": 2.1956533947324086,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004524758557512005,
      "loss": 5.2931,
      "mean_token_accuracy": 0.20148752629756927,
      "num_tokens": 43629102.0,
      "step": 19550
    },
    {
      "entropy": 5.465566396713257,
      "epoch": 2.1962149716403663,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004524511745687895,
      "loss": 5.182,
      "mean_token_accuracy": 0.20886197239160537,
      "num_tokens": 43640421.0,
      "step": 19555
    },
    {
      "entropy": 5.457819938659668,
      "epoch": 2.1967765485483235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004524264877362643,
      "loss": 5.1186,
      "mean_token_accuracy": 0.20673491507768632,
      "num_tokens": 43650516.0,
      "step": 19560
    },
    {
      "entropy": 5.469805765151977,
      "epoch": 2.197338125456281,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045240179525441094,
      "loss": 5.1107,
      "mean_token_accuracy": 0.2096590980887413,
      "num_tokens": 43661439.0,
      "step": 19565
    },
    {
      "entropy": 5.436346387863159,
      "epoch": 2.197899702364239,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004523770971240156,
      "loss": 5.185,
      "mean_token_accuracy": 0.2025914564728737,
      "num_tokens": 43672137.0,
      "step": 19570
    },
    {
      "entropy": 5.444410514831543,
      "epoch": 2.1984612792721965,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004523523933458647,
      "loss": 5.1727,
      "mean_token_accuracy": 0.19978250563144684,
      "num_tokens": 43683612.0,
      "step": 19575
    },
    {
      "entropy": 5.432463073730469,
      "epoch": 2.1990228561801537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004523276839207448,
      "loss": 5.1131,
      "mean_token_accuracy": 0.2059452474117279,
      "num_tokens": 43694331.0,
      "step": 19580
    },
    {
      "entropy": 5.478057479858398,
      "epoch": 2.1995844330881114,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045230296884944264,
      "loss": 5.2169,
      "mean_token_accuracy": 0.19790075719356537,
      "num_tokens": 43706572.0,
      "step": 19585
    },
    {
      "entropy": 5.391623020172119,
      "epoch": 2.200146009996069,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045227824813274517,
      "loss": 5.0849,
      "mean_token_accuracy": 0.2133317321538925,
      "num_tokens": 43718472.0,
      "step": 19590
    },
    {
      "entropy": 5.469793748855591,
      "epoch": 2.2007075869040267,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004522535217714396,
      "loss": 5.2013,
      "mean_token_accuracy": 0.20271628350019455,
      "num_tokens": 43729642.0,
      "step": 19595
    },
    {
      "entropy": 5.479317569732666,
      "epoch": 2.201269163811984,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004522287897663131,
      "loss": 5.1999,
      "mean_token_accuracy": 0.19823446571826936,
      "num_tokens": 43739846.0,
      "step": 19600
    },
    {
      "entropy": 5.382539463043213,
      "epoch": 2.2018307407199416,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045220405211815327,
      "loss": 5.0376,
      "mean_token_accuracy": 0.20500393807888032,
      "num_tokens": 43750265.0,
      "step": 19605
    },
    {
      "entropy": 5.440049457550049,
      "epoch": 2.202392317627899,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045217930882774754,
      "loss": 5.1407,
      "mean_token_accuracy": 0.1977613463997841,
      "num_tokens": 43761245.0,
      "step": 19610
    },
    {
      "entropy": 5.522872447967529,
      "epoch": 2.202953894535857,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045215455989588396,
      "loss": 5.151,
      "mean_token_accuracy": 0.20322937816381453,
      "num_tokens": 43772677.0,
      "step": 19615
    },
    {
      "entropy": 5.490469741821289,
      "epoch": 2.203515471443814,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045212980532335055,
      "loss": 5.2224,
      "mean_token_accuracy": 0.19638148099184036,
      "num_tokens": 43783128.0,
      "step": 19620
    },
    {
      "entropy": 5.378346920013428,
      "epoch": 2.2040770483517718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045210504511093534,
      "loss": 5.1505,
      "mean_token_accuracy": 0.2051555410027504,
      "num_tokens": 43795323.0,
      "step": 19625
    },
    {
      "entropy": 5.495058298110962,
      "epoch": 2.2046386252597294,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004520802792594269,
      "loss": 5.2309,
      "mean_token_accuracy": 0.19798728674650193,
      "num_tokens": 43806917.0,
      "step": 19630
    },
    {
      "entropy": 5.578366374969482,
      "epoch": 2.2052002021676866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045205550776961366,
      "loss": 5.2118,
      "mean_token_accuracy": 0.2005753293633461,
      "num_tokens": 43818963.0,
      "step": 19635
    },
    {
      "entropy": 5.39679536819458,
      "epoch": 2.2057617790756443,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004520307306422844,
      "loss": 5.0733,
      "mean_token_accuracy": 0.20616210103034974,
      "num_tokens": 43830163.0,
      "step": 19640
    },
    {
      "entropy": 5.359626770019531,
      "epoch": 2.206323355983602,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045200594787822795,
      "loss": 5.0932,
      "mean_token_accuracy": 0.20827651470899583,
      "num_tokens": 43840434.0,
      "step": 19645
    },
    {
      "entropy": 5.326594877243042,
      "epoch": 2.2068849328915596,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004519811594782335,
      "loss": 4.9504,
      "mean_token_accuracy": 0.21284874975681306,
      "num_tokens": 43851086.0,
      "step": 19650
    },
    {
      "entropy": 5.438270330429077,
      "epoch": 2.207446509799517,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045195636544309033,
      "loss": 5.1779,
      "mean_token_accuracy": 0.1964749351143837,
      "num_tokens": 43862834.0,
      "step": 19655
    },
    {
      "entropy": 5.41281156539917,
      "epoch": 2.2080080867074745,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045193156577358784,
      "loss": 5.1716,
      "mean_token_accuracy": 0.19788086116313935,
      "num_tokens": 43874868.0,
      "step": 19660
    },
    {
      "entropy": 5.498284864425659,
      "epoch": 2.208569663615432,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004519067604705157,
      "loss": 5.1537,
      "mean_token_accuracy": 0.20734975934028627,
      "num_tokens": 43886175.0,
      "step": 19665
    },
    {
      "entropy": 5.468722200393676,
      "epoch": 2.20913124052339,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045188194953466363,
      "loss": 5.1127,
      "mean_token_accuracy": 0.20881798565387727,
      "num_tokens": 43896489.0,
      "step": 19670
    },
    {
      "entropy": 5.39941668510437,
      "epoch": 2.209692817431347,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045185713296682175,
      "loss": 5.1052,
      "mean_token_accuracy": 0.20671319514513015,
      "num_tokens": 43907759.0,
      "step": 19675
    },
    {
      "entropy": 5.396782732009887,
      "epoch": 2.2102543943393047,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045183231076778005,
      "loss": 5.1138,
      "mean_token_accuracy": 0.201300710439682,
      "num_tokens": 43918136.0,
      "step": 19680
    },
    {
      "entropy": 5.474825096130371,
      "epoch": 2.2108159712472624,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045180748293832906,
      "loss": 5.1749,
      "mean_token_accuracy": 0.19686809033155442,
      "num_tokens": 43929662.0,
      "step": 19685
    },
    {
      "entropy": 5.466724920272827,
      "epoch": 2.21137754815522,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045178264947925924,
      "loss": 5.1458,
      "mean_token_accuracy": 0.20807749927043914,
      "num_tokens": 43939599.0,
      "step": 19690
    },
    {
      "entropy": 5.400431394577026,
      "epoch": 2.2119391250631772,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045175781039136133,
      "loss": 5.0902,
      "mean_token_accuracy": 0.2035491183400154,
      "num_tokens": 43949728.0,
      "step": 19695
    },
    {
      "entropy": 5.467465591430664,
      "epoch": 2.212500701971135,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004517329656754261,
      "loss": 5.2018,
      "mean_token_accuracy": 0.20216406881809235,
      "num_tokens": 43961147.0,
      "step": 19700
    },
    {
      "entropy": 5.495803165435791,
      "epoch": 2.2130622788790926,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004517081153322448,
      "loss": 5.1897,
      "mean_token_accuracy": 0.20225719809532167,
      "num_tokens": 43972541.0,
      "step": 19705
    },
    {
      "entropy": 5.446324872970581,
      "epoch": 2.2136238557870502,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004516832593626085,
      "loss": 5.2021,
      "mean_token_accuracy": 0.19894807785749435,
      "num_tokens": 43983540.0,
      "step": 19710
    },
    {
      "entropy": 5.404163122177124,
      "epoch": 2.2141854326950074,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004516583977673087,
      "loss": 5.0725,
      "mean_token_accuracy": 0.21014162600040437,
      "num_tokens": 43994255.0,
      "step": 19715
    },
    {
      "entropy": 5.527474355697632,
      "epoch": 2.214747009602965,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045163353054713704,
      "loss": 5.2451,
      "mean_token_accuracy": 0.19958435446023942,
      "num_tokens": 44004364.0,
      "step": 19720
    },
    {
      "entropy": 5.458494710922241,
      "epoch": 2.2153085865109228,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004516086577028852,
      "loss": 5.2164,
      "mean_token_accuracy": 0.20162972211837768,
      "num_tokens": 44015487.0,
      "step": 19725
    },
    {
      "entropy": 5.521120691299439,
      "epoch": 2.21587016341888,
      "grad_norm": 1.0,
      "learning_rate": 0.0004515837792353453,
      "loss": 5.188,
      "mean_token_accuracy": 0.20811489671468736,
      "num_tokens": 44026974.0,
      "step": 19730
    },
    {
      "entropy": 5.482464218139649,
      "epoch": 2.2164317403268377,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004515588951453093,
      "loss": 5.1441,
      "mean_token_accuracy": 0.20208192765712737,
      "num_tokens": 44038718.0,
      "step": 19735
    },
    {
      "entropy": 5.390679168701172,
      "epoch": 2.2169933172347953,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004515340054335696,
      "loss": 5.0713,
      "mean_token_accuracy": 0.2050388216972351,
      "num_tokens": 44048821.0,
      "step": 19740
    },
    {
      "entropy": 5.472933292388916,
      "epoch": 2.217554894142753,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004515091101009188,
      "loss": 5.2278,
      "mean_token_accuracy": 0.19522123485803605,
      "num_tokens": 44060973.0,
      "step": 19745
    },
    {
      "entropy": 5.593734645843506,
      "epoch": 2.21811647105071,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004514842091481493,
      "loss": 5.1967,
      "mean_token_accuracy": 0.20391546189785004,
      "num_tokens": 44073595.0,
      "step": 19750
    },
    {
      "entropy": 5.401270771026612,
      "epoch": 2.218678047958668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004514593025760543,
      "loss": 5.1146,
      "mean_token_accuracy": 0.2045004889369011,
      "num_tokens": 44084860.0,
      "step": 19755
    },
    {
      "entropy": 5.425442790985107,
      "epoch": 2.2192396248666255,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045143439038542653,
      "loss": 5.224,
      "mean_token_accuracy": 0.1957052990794182,
      "num_tokens": 44096026.0,
      "step": 19760
    },
    {
      "entropy": 5.473255777359009,
      "epoch": 2.219801201774583,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004514094725770594,
      "loss": 5.1392,
      "mean_token_accuracy": 0.2014969527721405,
      "num_tokens": 44107981.0,
      "step": 19765
    },
    {
      "entropy": 5.4604541778564455,
      "epoch": 2.2203627786825404,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045138454915174625,
      "loss": 5.1949,
      "mean_token_accuracy": 0.19833221584558486,
      "num_tokens": 44119589.0,
      "step": 19770
    },
    {
      "entropy": 5.420510101318359,
      "epoch": 2.220924355590498,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045135962011028064,
      "loss": 5.1749,
      "mean_token_accuracy": 0.20289915651082993,
      "num_tokens": 44131094.0,
      "step": 19775
    },
    {
      "entropy": 5.441696643829346,
      "epoch": 2.2214859324984557,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004513346854534564,
      "loss": 5.1622,
      "mean_token_accuracy": 0.19940573573112488,
      "num_tokens": 44141348.0,
      "step": 19780
    },
    {
      "entropy": 5.579134464263916,
      "epoch": 2.2220475094064134,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004513097451820673,
      "loss": 5.2968,
      "mean_token_accuracy": 0.19988732784986496,
      "num_tokens": 44152085.0,
      "step": 19785
    },
    {
      "entropy": 5.468767595291138,
      "epoch": 2.2226090863143706,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045128479929690756,
      "loss": 5.088,
      "mean_token_accuracy": 0.21069650053977967,
      "num_tokens": 44163518.0,
      "step": 19790
    },
    {
      "entropy": 5.380663156509399,
      "epoch": 2.2231706632223283,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004512598477987714,
      "loss": 5.1334,
      "mean_token_accuracy": 0.19633735716342926,
      "num_tokens": 44175513.0,
      "step": 19795
    },
    {
      "entropy": 5.469863700866699,
      "epoch": 2.223732240130286,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045123489068845334,
      "loss": 5.1319,
      "mean_token_accuracy": 0.2018109068274498,
      "num_tokens": 44186660.0,
      "step": 19800
    },
    {
      "entropy": 5.483963251113892,
      "epoch": 2.2242938170382436,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045120992796674804,
      "loss": 5.213,
      "mean_token_accuracy": 0.19745038002729415,
      "num_tokens": 44197613.0,
      "step": 19805
    },
    {
      "entropy": 5.407143449783325,
      "epoch": 2.224855393946201,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004511849596344503,
      "loss": 5.1335,
      "mean_token_accuracy": 0.1987305238842964,
      "num_tokens": 44209469.0,
      "step": 19810
    },
    {
      "entropy": 5.473362445831299,
      "epoch": 2.2254169708541585,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045115998569235507,
      "loss": 5.1522,
      "mean_token_accuracy": 0.1981020152568817,
      "num_tokens": 44220672.0,
      "step": 19815
    },
    {
      "entropy": 5.480352973937988,
      "epoch": 2.225978547762116,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045113500614125756,
      "loss": 5.1724,
      "mean_token_accuracy": 0.1967556044459343,
      "num_tokens": 44232867.0,
      "step": 19820
    },
    {
      "entropy": 5.4493903636932375,
      "epoch": 2.2265401246700733,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004511100209819532,
      "loss": 5.1772,
      "mean_token_accuracy": 0.20165464133024216,
      "num_tokens": 44245567.0,
      "step": 19825
    },
    {
      "entropy": 5.435141754150391,
      "epoch": 2.227101701578031,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045108503021523737,
      "loss": 5.1281,
      "mean_token_accuracy": 0.20610821545124053,
      "num_tokens": 44257449.0,
      "step": 19830
    },
    {
      "entropy": 5.459766530990601,
      "epoch": 2.2276632784859887,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045106003384190604,
      "loss": 5.2096,
      "mean_token_accuracy": 0.197022745013237,
      "num_tokens": 44268890.0,
      "step": 19835
    },
    {
      "entropy": 5.532184362411499,
      "epoch": 2.2282248553939463,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045103503186275486,
      "loss": 5.1714,
      "mean_token_accuracy": 0.19489402323961258,
      "num_tokens": 44280719.0,
      "step": 19840
    },
    {
      "entropy": 5.549978542327881,
      "epoch": 2.2287864323019035,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045101002427857997,
      "loss": 5.2616,
      "mean_token_accuracy": 0.19264942705631255,
      "num_tokens": 44292567.0,
      "step": 19845
    },
    {
      "entropy": 5.465829467773437,
      "epoch": 2.229348009209861,
      "grad_norm": 1.25,
      "learning_rate": 0.00045098501109017765,
      "loss": 5.1984,
      "mean_token_accuracy": 0.1947641223669052,
      "num_tokens": 44304177.0,
      "step": 19850
    },
    {
      "entropy": 5.437126684188843,
      "epoch": 2.229909586117819,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045095999229834437,
      "loss": 5.0856,
      "mean_token_accuracy": 0.2043955832719803,
      "num_tokens": 44315142.0,
      "step": 19855
    },
    {
      "entropy": 5.431200265884399,
      "epoch": 2.2304711630257765,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004509349679038767,
      "loss": 5.249,
      "mean_token_accuracy": 0.20043786913156508,
      "num_tokens": 44327457.0,
      "step": 19860
    },
    {
      "entropy": 5.552132654190063,
      "epoch": 2.2310327399337337,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045090993790757137,
      "loss": 5.23,
      "mean_token_accuracy": 0.19535116851329803,
      "num_tokens": 44338978.0,
      "step": 19865
    },
    {
      "entropy": 5.5329993724822994,
      "epoch": 2.2315943168416914,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004508849023102253,
      "loss": 5.2273,
      "mean_token_accuracy": 0.201108855009079,
      "num_tokens": 44349121.0,
      "step": 19870
    },
    {
      "entropy": 5.397964572906494,
      "epoch": 2.232155893749649,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004508598611126359,
      "loss": 5.1004,
      "mean_token_accuracy": 0.2088798239827156,
      "num_tokens": 44360753.0,
      "step": 19875
    },
    {
      "entropy": 5.479589796066284,
      "epoch": 2.2327174706576067,
      "grad_norm": 1.125,
      "learning_rate": 0.0004508348143156002,
      "loss": 5.1646,
      "mean_token_accuracy": 0.19993786960840226,
      "num_tokens": 44371977.0,
      "step": 19880
    },
    {
      "entropy": 5.42446346282959,
      "epoch": 2.233279047565564,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045080976191991587,
      "loss": 5.0802,
      "mean_token_accuracy": 0.2039664462208748,
      "num_tokens": 44383077.0,
      "step": 19885
    },
    {
      "entropy": 5.457240438461303,
      "epoch": 2.2338406244735216,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004507847039263805,
      "loss": 5.1136,
      "mean_token_accuracy": 0.2009204939007759,
      "num_tokens": 44394827.0,
      "step": 19890
    },
    {
      "entropy": 5.401480388641358,
      "epoch": 2.2344022013814793,
      "grad_norm": 1.203125,
      "learning_rate": 0.000450759640335792,
      "loss": 5.0796,
      "mean_token_accuracy": 0.208608914911747,
      "num_tokens": 44406594.0,
      "step": 19895
    },
    {
      "entropy": 5.3781944751739506,
      "epoch": 2.234963778289437,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004507345711489484,
      "loss": 5.0789,
      "mean_token_accuracy": 0.2009037658572197,
      "num_tokens": 44417842.0,
      "step": 19900
    },
    {
      "entropy": 5.47165584564209,
      "epoch": 2.235525355197394,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045070949636664785,
      "loss": 5.123,
      "mean_token_accuracy": 0.20028728544712066,
      "num_tokens": 44428234.0,
      "step": 19905
    },
    {
      "entropy": 5.384031295776367,
      "epoch": 2.236086932105352,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004506844159896888,
      "loss": 5.0782,
      "mean_token_accuracy": 0.20572000443935395,
      "num_tokens": 44437504.0,
      "step": 19910
    },
    {
      "entropy": 5.457863998413086,
      "epoch": 2.2366485090133095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004506593300188698,
      "loss": 5.1303,
      "mean_token_accuracy": 0.2080046519637108,
      "num_tokens": 44447873.0,
      "step": 19915
    },
    {
      "entropy": 5.464411640167237,
      "epoch": 2.2372100859212667,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004506342384549897,
      "loss": 5.2087,
      "mean_token_accuracy": 0.20213390290737152,
      "num_tokens": 44458990.0,
      "step": 19920
    },
    {
      "entropy": 5.466749906539917,
      "epoch": 2.2377716628292244,
      "grad_norm": 1.125,
      "learning_rate": 0.0004506091412988472,
      "loss": 5.1104,
      "mean_token_accuracy": 0.20302632004022597,
      "num_tokens": 44469374.0,
      "step": 19925
    },
    {
      "entropy": 5.447497749328614,
      "epoch": 2.238333239737182,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045058403855124165,
      "loss": 5.1346,
      "mean_token_accuracy": 0.20520556420087815,
      "num_tokens": 44479629.0,
      "step": 19930
    },
    {
      "entropy": 5.50403060913086,
      "epoch": 2.2388948166451397,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004505589302129721,
      "loss": 5.207,
      "mean_token_accuracy": 0.1963550940155983,
      "num_tokens": 44490603.0,
      "step": 19935
    },
    {
      "entropy": 5.3765510559082035,
      "epoch": 2.239456393553097,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004505338162848382,
      "loss": 5.073,
      "mean_token_accuracy": 0.2119968131184578,
      "num_tokens": 44501527.0,
      "step": 19940
    },
    {
      "entropy": 5.514767837524414,
      "epoch": 2.2400179704610546,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004505086967676395,
      "loss": 5.2819,
      "mean_token_accuracy": 0.19897997081279756,
      "num_tokens": 44512619.0,
      "step": 19945
    },
    {
      "entropy": 5.569184970855713,
      "epoch": 2.240579547369012,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004504835716621757,
      "loss": 5.2666,
      "mean_token_accuracy": 0.19585637599229813,
      "num_tokens": 44524639.0,
      "step": 19950
    },
    {
      "entropy": 5.417506980895996,
      "epoch": 2.24114112427697,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004504584409692471,
      "loss": 5.1243,
      "mean_token_accuracy": 0.21007796376943588,
      "num_tokens": 44535013.0,
      "step": 19955
    },
    {
      "entropy": 5.557058954238892,
      "epoch": 2.241702701184927,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045043330468965364,
      "loss": 5.233,
      "mean_token_accuracy": 0.19973451793193817,
      "num_tokens": 44546838.0,
      "step": 19960
    },
    {
      "entropy": 5.443006944656372,
      "epoch": 2.2422642780928848,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045040816282419566,
      "loss": 5.0991,
      "mean_token_accuracy": 0.20768779814243316,
      "num_tokens": 44557984.0,
      "step": 19965
    },
    {
      "entropy": 5.442243766784668,
      "epoch": 2.2428258550008424,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045038301537367374,
      "loss": 5.078,
      "mean_token_accuracy": 0.20558004230260848,
      "num_tokens": 44569029.0,
      "step": 19970
    },
    {
      "entropy": 5.426868629455567,
      "epoch": 2.2433874319088,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045035786233888864,
      "loss": 5.1766,
      "mean_token_accuracy": 0.206084306538105,
      "num_tokens": 44578804.0,
      "step": 19975
    },
    {
      "entropy": 5.4826044082641605,
      "epoch": 2.2439490088167573,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045033270372064123,
      "loss": 5.0925,
      "mean_token_accuracy": 0.20685162395238876,
      "num_tokens": 44589362.0,
      "step": 19980
    },
    {
      "entropy": 5.4772521495819095,
      "epoch": 2.244510585724715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045030753951973243,
      "loss": 5.1628,
      "mean_token_accuracy": 0.20179336816072463,
      "num_tokens": 44600137.0,
      "step": 19985
    },
    {
      "entropy": 5.413036584854126,
      "epoch": 2.2450721626326726,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045028236973696363,
      "loss": 5.1514,
      "mean_token_accuracy": 0.21040593832731247,
      "num_tokens": 44611707.0,
      "step": 19990
    },
    {
      "entropy": 5.448164176940918,
      "epoch": 2.2456337395406303,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004502571943731362,
      "loss": 5.1368,
      "mean_token_accuracy": 0.2031902328133583,
      "num_tokens": 44621141.0,
      "step": 19995
    },
    {
      "entropy": 5.486722564697265,
      "epoch": 2.2461953164485875,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004502320134290517,
      "loss": 5.1966,
      "mean_token_accuracy": 0.20007037818431855,
      "num_tokens": 44632043.0,
      "step": 20000
    },
    {
      "entropy": 5.515257024765015,
      "epoch": 2.246756893356545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045020682690551203,
      "loss": 5.211,
      "mean_token_accuracy": 0.19908837527036666,
      "num_tokens": 44643905.0,
      "step": 20005
    },
    {
      "entropy": 5.40823450088501,
      "epoch": 2.247318470264503,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045018163480331887,
      "loss": 5.1291,
      "mean_token_accuracy": 0.19696681797504426,
      "num_tokens": 44655105.0,
      "step": 20010
    },
    {
      "entropy": 5.408110475540161,
      "epoch": 2.24788004717246,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045015643712327465,
      "loss": 5.0951,
      "mean_token_accuracy": 0.20303869098424912,
      "num_tokens": 44667096.0,
      "step": 20015
    },
    {
      "entropy": 5.4923171043396,
      "epoch": 2.2484416240804177,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004501312338661815,
      "loss": 5.157,
      "mean_token_accuracy": 0.20112112462520598,
      "num_tokens": 44678206.0,
      "step": 20020
    },
    {
      "entropy": 5.425653123855591,
      "epoch": 2.2490032009883754,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004501060250328419,
      "loss": 5.0795,
      "mean_token_accuracy": 0.20635404735803603,
      "num_tokens": 44688619.0,
      "step": 20025
    },
    {
      "entropy": 5.392659759521484,
      "epoch": 2.249564777896333,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004500808106240586,
      "loss": 5.1362,
      "mean_token_accuracy": 0.20224724858999252,
      "num_tokens": 44700757.0,
      "step": 20030
    },
    {
      "entropy": 5.34596815109253,
      "epoch": 2.2501263548042907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004500555906406343,
      "loss": 5.003,
      "mean_token_accuracy": 0.21299721002578736,
      "num_tokens": 44710822.0,
      "step": 20035
    },
    {
      "entropy": 5.329952096939087,
      "epoch": 2.250687931712248,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004500303650833722,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2127438023686409,
      "num_tokens": 44722163.0,
      "step": 20040
    },
    {
      "entropy": 5.501678800582885,
      "epoch": 2.2512495086202056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004500051339530754,
      "loss": 5.2003,
      "mean_token_accuracy": 0.1995450437068939,
      "num_tokens": 44734271.0,
      "step": 20045
    },
    {
      "entropy": 5.520967864990235,
      "epoch": 2.2518110855281632,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004499798972505472,
      "loss": 5.153,
      "mean_token_accuracy": 0.20573473870754241,
      "num_tokens": 44745345.0,
      "step": 20050
    },
    {
      "entropy": 5.448700714111328,
      "epoch": 2.2523726624361204,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044995465497659127,
      "loss": 5.1245,
      "mean_token_accuracy": 0.1958379179239273,
      "num_tokens": 44756986.0,
      "step": 20055
    },
    {
      "entropy": 5.366703033447266,
      "epoch": 2.252934239344078,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004499294071320112,
      "loss": 5.0696,
      "mean_token_accuracy": 0.20842144787311553,
      "num_tokens": 44767259.0,
      "step": 20060
    },
    {
      "entropy": 5.403236007690429,
      "epoch": 2.2534958162520358,
      "grad_norm": 1.125,
      "learning_rate": 0.000449904153717611,
      "loss": 5.1658,
      "mean_token_accuracy": 0.2043090581893921,
      "num_tokens": 44778884.0,
      "step": 20065
    },
    {
      "entropy": 5.445444107055664,
      "epoch": 2.2540573931599934,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004498788947341947,
      "loss": 5.0721,
      "mean_token_accuracy": 0.2082044005393982,
      "num_tokens": 44789216.0,
      "step": 20070
    },
    {
      "entropy": 5.337964963912964,
      "epoch": 2.2546189700679506,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004498536301825665,
      "loss": 5.024,
      "mean_token_accuracy": 0.2110508233308792,
      "num_tokens": 44799357.0,
      "step": 20075
    },
    {
      "entropy": 5.377265071868896,
      "epoch": 2.2551805469759083,
      "grad_norm": 1.03125,
      "learning_rate": 0.000449828360063531,
      "loss": 5.1641,
      "mean_token_accuracy": 0.20232641696929932,
      "num_tokens": 44811040.0,
      "step": 20080
    },
    {
      "entropy": 5.475870895385742,
      "epoch": 2.255742123883866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044980308437789267,
      "loss": 5.2257,
      "mean_token_accuracy": 0.19621861577033997,
      "num_tokens": 44822103.0,
      "step": 20085
    },
    {
      "entropy": 5.451181411743164,
      "epoch": 2.2563037007918236,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004497778031264564,
      "loss": 5.127,
      "mean_token_accuracy": 0.19825687557458876,
      "num_tokens": 44834712.0,
      "step": 20090
    },
    {
      "entropy": 5.434058809280396,
      "epoch": 2.256865277699781,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044975251631002694,
      "loss": 5.0669,
      "mean_token_accuracy": 0.20046408474445343,
      "num_tokens": 44845412.0,
      "step": 20095
    },
    {
      "entropy": 5.388689088821411,
      "epoch": 2.2574268546077385,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004497272239294097,
      "loss": 5.1421,
      "mean_token_accuracy": 0.2022656410932541,
      "num_tokens": 44857720.0,
      "step": 20100
    },
    {
      "entropy": 5.388620758056641,
      "epoch": 2.257988431515696,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004497019259854099,
      "loss": 5.113,
      "mean_token_accuracy": 0.20880619436502457,
      "num_tokens": 44868253.0,
      "step": 20105
    },
    {
      "entropy": 5.472629404067993,
      "epoch": 2.2585500084236534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044967662247883295,
      "loss": 5.0992,
      "mean_token_accuracy": 0.1986072525382042,
      "num_tokens": 44878525.0,
      "step": 20110
    },
    {
      "entropy": 5.365305233001709,
      "epoch": 2.259111585331611,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044965131341048454,
      "loss": 5.079,
      "mean_token_accuracy": 0.20988816767930984,
      "num_tokens": 44889565.0,
      "step": 20115
    },
    {
      "entropy": 5.3405157089233395,
      "epoch": 2.2596731622395687,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004496259987811707,
      "loss": 5.0632,
      "mean_token_accuracy": 0.20656754821538925,
      "num_tokens": 44901191.0,
      "step": 20120
    },
    {
      "entropy": 5.412547588348389,
      "epoch": 2.2602347391475264,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004496006785916973,
      "loss": 5.0423,
      "mean_token_accuracy": 0.21028562039136886,
      "num_tokens": 44911483.0,
      "step": 20125
    },
    {
      "entropy": 5.5000035762786865,
      "epoch": 2.260796316055484,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044957535284287055,
      "loss": 5.2279,
      "mean_token_accuracy": 0.2033305749297142,
      "num_tokens": 44923537.0,
      "step": 20130
    },
    {
      "entropy": 5.494990682601928,
      "epoch": 2.2613578929634413,
      "grad_norm": 1.125,
      "learning_rate": 0.00044955002153549686,
      "loss": 5.1746,
      "mean_token_accuracy": 0.19797872006893158,
      "num_tokens": 44935780.0,
      "step": 20135
    },
    {
      "entropy": 5.489409494400024,
      "epoch": 2.261919469871399,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004495246846703828,
      "loss": 5.1553,
      "mean_token_accuracy": 0.20243533551692963,
      "num_tokens": 44946879.0,
      "step": 20140
    },
    {
      "entropy": 5.4829188823699955,
      "epoch": 2.2624810467793566,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044949934224833505,
      "loss": 5.1437,
      "mean_token_accuracy": 0.20465155690908432,
      "num_tokens": 44958385.0,
      "step": 20145
    },
    {
      "entropy": 5.450573396682739,
      "epoch": 2.263042623687314,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004494739942701606,
      "loss": 5.2525,
      "mean_token_accuracy": 0.19533410370349885,
      "num_tokens": 44970597.0,
      "step": 20150
    },
    {
      "entropy": 5.454568576812744,
      "epoch": 2.2636042005952715,
      "grad_norm": 1.125,
      "learning_rate": 0.00044944864073666655,
      "loss": 5.178,
      "mean_token_accuracy": 0.20250202864408492,
      "num_tokens": 44983123.0,
      "step": 20155
    },
    {
      "entropy": 5.4483579158782955,
      "epoch": 2.264165777503229,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044942328164866,
      "loss": 5.0974,
      "mean_token_accuracy": 0.19975200891494752,
      "num_tokens": 44993378.0,
      "step": 20160
    },
    {
      "entropy": 5.399605560302734,
      "epoch": 2.2647273544111868,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004493979170069486,
      "loss": 5.0704,
      "mean_token_accuracy": 0.1999097764492035,
      "num_tokens": 45004905.0,
      "step": 20165
    },
    {
      "entropy": 5.380496597290039,
      "epoch": 2.265288931319144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044937254681233984,
      "loss": 5.1204,
      "mean_token_accuracy": 0.20543215721845626,
      "num_tokens": 45016640.0,
      "step": 20170
    },
    {
      "entropy": 5.487663221359253,
      "epoch": 2.2658505082271017,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044934717106564164,
      "loss": 5.2602,
      "mean_token_accuracy": 0.19706982225179673,
      "num_tokens": 45028285.0,
      "step": 20175
    },
    {
      "entropy": 5.501610612869262,
      "epoch": 2.2664120851350593,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044932178976766183,
      "loss": 5.1345,
      "mean_token_accuracy": 0.19833078980445862,
      "num_tokens": 45038924.0,
      "step": 20180
    },
    {
      "entropy": 5.482291555404663,
      "epoch": 2.266973662043017,
      "grad_norm": 1.125,
      "learning_rate": 0.00044929640291920865,
      "loss": 5.0943,
      "mean_token_accuracy": 0.20344796925783157,
      "num_tokens": 45050483.0,
      "step": 20185
    },
    {
      "entropy": 5.513018989562989,
      "epoch": 2.267535238950974,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004492710105210904,
      "loss": 5.2948,
      "mean_token_accuracy": 0.19606169760227204,
      "num_tokens": 45063242.0,
      "step": 20190
    },
    {
      "entropy": 5.3799660205841064,
      "epoch": 2.268096815858932,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044924561257411554,
      "loss": 5.155,
      "mean_token_accuracy": 0.2003306970000267,
      "num_tokens": 45074057.0,
      "step": 20195
    },
    {
      "entropy": 5.414817190170288,
      "epoch": 2.2686583927668895,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004492202090790928,
      "loss": 5.2061,
      "mean_token_accuracy": 0.19339478462934495,
      "num_tokens": 45085488.0,
      "step": 20200
    },
    {
      "entropy": 5.51557149887085,
      "epoch": 2.2692199696748467,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004491948000368311,
      "loss": 5.2609,
      "mean_token_accuracy": 0.19702999889850617,
      "num_tokens": 45096303.0,
      "step": 20205
    },
    {
      "entropy": 5.461765432357788,
      "epoch": 2.2697815465828044,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044916938544813936,
      "loss": 5.0331,
      "mean_token_accuracy": 0.20492334961891173,
      "num_tokens": 45106411.0,
      "step": 20210
    },
    {
      "entropy": 5.462426328659058,
      "epoch": 2.270343123490762,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044914396531382686,
      "loss": 5.2059,
      "mean_token_accuracy": 0.20202705413103103,
      "num_tokens": 45117355.0,
      "step": 20215
    },
    {
      "entropy": 5.434532356262207,
      "epoch": 2.2709047003987197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044911853963470295,
      "loss": 5.074,
      "mean_token_accuracy": 0.21141235232353212,
      "num_tokens": 45127612.0,
      "step": 20220
    },
    {
      "entropy": 5.5082543849945065,
      "epoch": 2.2714662773066774,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004490931084115772,
      "loss": 5.2505,
      "mean_token_accuracy": 0.20010817795991898,
      "num_tokens": 45138495.0,
      "step": 20225
    },
    {
      "entropy": 5.426121997833252,
      "epoch": 2.2720278542146346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044906767164525936,
      "loss": 5.0897,
      "mean_token_accuracy": 0.21036437153816223,
      "num_tokens": 45149727.0,
      "step": 20230
    },
    {
      "entropy": 5.414049911499023,
      "epoch": 2.2725894311225923,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004490422293365594,
      "loss": 5.1466,
      "mean_token_accuracy": 0.20095557421445848,
      "num_tokens": 45160921.0,
      "step": 20235
    },
    {
      "entropy": 5.396750402450562,
      "epoch": 2.27315100803055,
      "grad_norm": 1.0,
      "learning_rate": 0.0004490167814862872,
      "loss": 5.068,
      "mean_token_accuracy": 0.20542089641094208,
      "num_tokens": 45172607.0,
      "step": 20240
    },
    {
      "entropy": 5.5406591415405275,
      "epoch": 2.273712584938507,
      "grad_norm": 1.125,
      "learning_rate": 0.0004489913280952533,
      "loss": 5.2551,
      "mean_token_accuracy": 0.20312247425317764,
      "num_tokens": 45186330.0,
      "step": 20245
    },
    {
      "entropy": 5.329312181472778,
      "epoch": 2.274274161846465,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004489658691642681,
      "loss": 5.0133,
      "mean_token_accuracy": 0.2148903951048851,
      "num_tokens": 45198239.0,
      "step": 20250
    },
    {
      "entropy": 5.314783096313477,
      "epoch": 2.2748357387544225,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044894040469414207,
      "loss": 5.0203,
      "mean_token_accuracy": 0.20988162457942963,
      "num_tokens": 45209295.0,
      "step": 20255
    },
    {
      "entropy": 5.430111074447632,
      "epoch": 2.27539731566238,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044891493468568607,
      "loss": 5.1488,
      "mean_token_accuracy": 0.20581330060958863,
      "num_tokens": 45220210.0,
      "step": 20260
    },
    {
      "entropy": 5.513519191741944,
      "epoch": 2.2759588925703373,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044888945913971106,
      "loss": 5.1752,
      "mean_token_accuracy": 0.20108020156621934,
      "num_tokens": 45231809.0,
      "step": 20265
    },
    {
      "entropy": 5.445365953445434,
      "epoch": 2.276520469478295,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044886397805702835,
      "loss": 5.1269,
      "mean_token_accuracy": 0.21129992455244065,
      "num_tokens": 45242847.0,
      "step": 20270
    },
    {
      "entropy": 5.338790941238403,
      "epoch": 2.2770820463862527,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044883849143844914,
      "loss": 5.0733,
      "mean_token_accuracy": 0.21474968791007995,
      "num_tokens": 45253357.0,
      "step": 20275
    },
    {
      "entropy": 5.320821952819824,
      "epoch": 2.2776436232942103,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004488129992847848,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20881016701459884,
      "num_tokens": 45264607.0,
      "step": 20280
    },
    {
      "entropy": 5.493633079528808,
      "epoch": 2.2782052002021675,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004487875015968473,
      "loss": 5.1753,
      "mean_token_accuracy": 0.20363058149814606,
      "num_tokens": 45275211.0,
      "step": 20285
    },
    {
      "entropy": 5.396355867385864,
      "epoch": 2.278766777110125,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004487619983754483,
      "loss": 5.0553,
      "mean_token_accuracy": 0.20995334386825562,
      "num_tokens": 45286665.0,
      "step": 20290
    },
    {
      "entropy": 5.397709560394287,
      "epoch": 2.279328354018083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044873648962139983,
      "loss": 5.1663,
      "mean_token_accuracy": 0.2040874645113945,
      "num_tokens": 45298066.0,
      "step": 20295
    },
    {
      "entropy": 5.401922607421875,
      "epoch": 2.27988993092604,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004487109753355142,
      "loss": 5.1513,
      "mean_token_accuracy": 0.1969730570912361,
      "num_tokens": 45308799.0,
      "step": 20300
    },
    {
      "entropy": 5.440746736526489,
      "epoch": 2.2804515078339977,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004486854555186037,
      "loss": 5.0144,
      "mean_token_accuracy": 0.21088502556085587,
      "num_tokens": 45319932.0,
      "step": 20305
    },
    {
      "entropy": 5.4936261653900145,
      "epoch": 2.2810130847419554,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004486599301714809,
      "loss": 5.16,
      "mean_token_accuracy": 0.20557641834020615,
      "num_tokens": 45330008.0,
      "step": 20310
    },
    {
      "entropy": 5.432768106460571,
      "epoch": 2.281574661649913,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004486343992949587,
      "loss": 5.1782,
      "mean_token_accuracy": 0.19589841216802598,
      "num_tokens": 45340877.0,
      "step": 20315
    },
    {
      "entropy": 5.430566358566284,
      "epoch": 2.2821362385578707,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044860886288984977,
      "loss": 5.1169,
      "mean_token_accuracy": 0.2034818485379219,
      "num_tokens": 45351956.0,
      "step": 20320
    },
    {
      "entropy": 5.505562162399292,
      "epoch": 2.282697815465828,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004485833209569674,
      "loss": 5.219,
      "mean_token_accuracy": 0.19854362457990646,
      "num_tokens": 45363133.0,
      "step": 20325
    },
    {
      "entropy": 5.402228689193725,
      "epoch": 2.2832593923737856,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004485577734971247,
      "loss": 5.0771,
      "mean_token_accuracy": 0.2046315237879753,
      "num_tokens": 45373917.0,
      "step": 20330
    },
    {
      "entropy": 5.471015977859497,
      "epoch": 2.2838209692817433,
      "grad_norm": 1.125,
      "learning_rate": 0.00044853222051113525,
      "loss": 5.1818,
      "mean_token_accuracy": 0.202048559486866,
      "num_tokens": 45385575.0,
      "step": 20335
    },
    {
      "entropy": 5.419047737121582,
      "epoch": 2.2843825461897005,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044850666199981244,
      "loss": 5.0747,
      "mean_token_accuracy": 0.20632156878709793,
      "num_tokens": 45396615.0,
      "step": 20340
    },
    {
      "entropy": 5.452276515960693,
      "epoch": 2.284944123097658,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044848109796397035,
      "loss": 5.0964,
      "mean_token_accuracy": 0.20425934344530106,
      "num_tokens": 45408931.0,
      "step": 20345
    },
    {
      "entropy": 5.493478202819825,
      "epoch": 2.285505700005616,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044845552840442276,
      "loss": 5.1806,
      "mean_token_accuracy": 0.19871591329574584,
      "num_tokens": 45419993.0,
      "step": 20350
    },
    {
      "entropy": 5.45655255317688,
      "epoch": 2.2860672769135735,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044842995332198393,
      "loss": 5.2168,
      "mean_token_accuracy": 0.19570968747138978,
      "num_tokens": 45431889.0,
      "step": 20355
    },
    {
      "entropy": 5.385016918182373,
      "epoch": 2.2866288538215307,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044840437271746807,
      "loss": 5.1425,
      "mean_token_accuracy": 0.2062792956829071,
      "num_tokens": 45444233.0,
      "step": 20360
    },
    {
      "entropy": 5.399848318099975,
      "epoch": 2.2871904307294884,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004483787865916898,
      "loss": 5.1015,
      "mean_token_accuracy": 0.20416233390569688,
      "num_tokens": 45455743.0,
      "step": 20365
    },
    {
      "entropy": 5.389883375167846,
      "epoch": 2.287752007637446,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044835319494546366,
      "loss": 5.0441,
      "mean_token_accuracy": 0.20174830704927443,
      "num_tokens": 45466272.0,
      "step": 20370
    },
    {
      "entropy": 5.4742837905883786,
      "epoch": 2.2883135845454037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044832759777960456,
      "loss": 5.1164,
      "mean_token_accuracy": 0.2006078764796257,
      "num_tokens": 45477828.0,
      "step": 20375
    },
    {
      "entropy": 5.4513427257537845,
      "epoch": 2.288875161453361,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004483019950949276,
      "loss": 5.1228,
      "mean_token_accuracy": 0.19857906848192214,
      "num_tokens": 45489315.0,
      "step": 20380
    },
    {
      "entropy": 5.347049140930176,
      "epoch": 2.2894367383613186,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004482763868922478,
      "loss": 5.0652,
      "mean_token_accuracy": 0.21041259169578552,
      "num_tokens": 45499836.0,
      "step": 20385
    },
    {
      "entropy": 5.4062036037445065,
      "epoch": 2.289998315269276,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004482507731723807,
      "loss": 5.1589,
      "mean_token_accuracy": 0.20286243855953218,
      "num_tokens": 45511106.0,
      "step": 20390
    },
    {
      "entropy": 5.409636354446411,
      "epoch": 2.2905598921772334,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004482251539361418,
      "loss": 5.1086,
      "mean_token_accuracy": 0.2008928507566452,
      "num_tokens": 45521314.0,
      "step": 20395
    },
    {
      "entropy": 5.532837772369385,
      "epoch": 2.291121469085191,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004481995291843467,
      "loss": 5.1464,
      "mean_token_accuracy": 0.20846298038959504,
      "num_tokens": 45531675.0,
      "step": 20400
    },
    {
      "entropy": 5.4842345237731935,
      "epoch": 2.2916830459931488,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004481738989178115,
      "loss": 5.182,
      "mean_token_accuracy": 0.20274983048439027,
      "num_tokens": 45542064.0,
      "step": 20405
    },
    {
      "entropy": 5.406715536117554,
      "epoch": 2.2922446229011064,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004481482631373522,
      "loss": 5.1992,
      "mean_token_accuracy": 0.20416887253522872,
      "num_tokens": 45553764.0,
      "step": 20410
    },
    {
      "entropy": 5.527762603759766,
      "epoch": 2.292806199809064,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044812262184378503,
      "loss": 5.1981,
      "mean_token_accuracy": 0.2025418683886528,
      "num_tokens": 45565302.0,
      "step": 20415
    },
    {
      "entropy": 5.339626455307007,
      "epoch": 2.2933677767170213,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044809697503792647,
      "loss": 5.0301,
      "mean_token_accuracy": 0.21230603456497193,
      "num_tokens": 45577165.0,
      "step": 20420
    },
    {
      "entropy": 5.41700119972229,
      "epoch": 2.293929353624979,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044807132272059306,
      "loss": 5.0864,
      "mean_token_accuracy": 0.20804390907287598,
      "num_tokens": 45587771.0,
      "step": 20425
    },
    {
      "entropy": 5.439433813095093,
      "epoch": 2.2944909305329366,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004480456648926016,
      "loss": 5.1223,
      "mean_token_accuracy": 0.21467435508966445,
      "num_tokens": 45598815.0,
      "step": 20430
    },
    {
      "entropy": 5.4032196521759035,
      "epoch": 2.295052507440894,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004480200015547691,
      "loss": 5.0677,
      "mean_token_accuracy": 0.20760321021080017,
      "num_tokens": 45609007.0,
      "step": 20435
    },
    {
      "entropy": 5.419494676589966,
      "epoch": 2.2956140843488515,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044799433270791255,
      "loss": 5.1907,
      "mean_token_accuracy": 0.19920454621315004,
      "num_tokens": 45621323.0,
      "step": 20440
    },
    {
      "entropy": 5.36867995262146,
      "epoch": 2.296175661256809,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004479686583528494,
      "loss": 4.9543,
      "mean_token_accuracy": 0.2121911406517029,
      "num_tokens": 45631895.0,
      "step": 20445
    },
    {
      "entropy": 5.476343727111816,
      "epoch": 2.296737238164767,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004479429784903971,
      "loss": 5.1216,
      "mean_token_accuracy": 0.20749529153108598,
      "num_tokens": 45643204.0,
      "step": 20450
    },
    {
      "entropy": 5.462383794784546,
      "epoch": 2.297298815072724,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004479172931213733,
      "loss": 5.1571,
      "mean_token_accuracy": 0.20414634495973588,
      "num_tokens": 45654607.0,
      "step": 20455
    },
    {
      "entropy": 5.343553733825684,
      "epoch": 2.2978603919806817,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004478916022465958,
      "loss": 5.0717,
      "mean_token_accuracy": 0.20763460099697112,
      "num_tokens": 45666746.0,
      "step": 20460
    },
    {
      "entropy": 5.3788408756256105,
      "epoch": 2.2984219688886394,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004478659058668827,
      "loss": 5.0256,
      "mean_token_accuracy": 0.20930723398923873,
      "num_tokens": 45676663.0,
      "step": 20465
    },
    {
      "entropy": 5.3930426120758055,
      "epoch": 2.298983545796597,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044784020398305204,
      "loss": 5.0583,
      "mean_token_accuracy": 0.20310064256191254,
      "num_tokens": 45688746.0,
      "step": 20470
    },
    {
      "entropy": 5.534788799285889,
      "epoch": 2.2995451227045542,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004478144965959222,
      "loss": 5.2655,
      "mean_token_accuracy": 0.19727590233087539,
      "num_tokens": 45700199.0,
      "step": 20475
    },
    {
      "entropy": 5.461199808120727,
      "epoch": 2.300106699612512,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044778878370631183,
      "loss": 5.0975,
      "mean_token_accuracy": 0.197740438580513,
      "num_tokens": 45710771.0,
      "step": 20480
    },
    {
      "entropy": 5.477123880386353,
      "epoch": 2.3006682765204696,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044776306531503957,
      "loss": 5.2738,
      "mean_token_accuracy": 0.20002095252275467,
      "num_tokens": 45721957.0,
      "step": 20485
    },
    {
      "entropy": 5.410528373718262,
      "epoch": 2.301229853428427,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044773734142292425,
      "loss": 5.1345,
      "mean_token_accuracy": 0.19851591885089875,
      "num_tokens": 45733904.0,
      "step": 20490
    },
    {
      "entropy": 5.4797749519348145,
      "epoch": 2.3017914303363844,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004477116120307851,
      "loss": 5.1455,
      "mean_token_accuracy": 0.20586278438568115,
      "num_tokens": 45744004.0,
      "step": 20495
    },
    {
      "entropy": 5.429679298400879,
      "epoch": 2.302353007244342,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004476858771394412,
      "loss": 5.0512,
      "mean_token_accuracy": 0.21419067978858947,
      "num_tokens": 45755130.0,
      "step": 20500
    },
    {
      "entropy": 5.43867506980896,
      "epoch": 2.3029145841522998,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004476601367497119,
      "loss": 5.2285,
      "mean_token_accuracy": 0.2003798231482506,
      "num_tokens": 45767353.0,
      "step": 20505
    },
    {
      "entropy": 5.520587825775147,
      "epoch": 2.3034761610602574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000447634390862417,
      "loss": 5.2957,
      "mean_token_accuracy": 0.19349124580621718,
      "num_tokens": 45779266.0,
      "step": 20510
    },
    {
      "entropy": 5.560569190979004,
      "epoch": 2.3040377379682146,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044760863947837607,
      "loss": 5.2447,
      "mean_token_accuracy": 0.19961845576763154,
      "num_tokens": 45790601.0,
      "step": 20515
    },
    {
      "entropy": 5.432533597946167,
      "epoch": 2.3045993148761723,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044758288259840904,
      "loss": 5.1172,
      "mean_token_accuracy": 0.2029760792851448,
      "num_tokens": 45801449.0,
      "step": 20520
    },
    {
      "entropy": 5.330210494995117,
      "epoch": 2.30516089178413,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004475571202233363,
      "loss": 5.0558,
      "mean_token_accuracy": 0.20565571039915084,
      "num_tokens": 45811737.0,
      "step": 20525
    },
    {
      "entropy": 5.436011457443238,
      "epoch": 2.305722468692087,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004475313523539778,
      "loss": 5.1017,
      "mean_token_accuracy": 0.2086130827665329,
      "num_tokens": 45822955.0,
      "step": 20530
    },
    {
      "entropy": 5.455626344680786,
      "epoch": 2.306284045600045,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044750557899115407,
      "loss": 5.1957,
      "mean_token_accuracy": 0.20599274784326554,
      "num_tokens": 45834697.0,
      "step": 20535
    },
    {
      "entropy": 5.403077077865601,
      "epoch": 2.3068456225080025,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004474798001356859,
      "loss": 5.1165,
      "mean_token_accuracy": 0.19859466701745987,
      "num_tokens": 45846869.0,
      "step": 20540
    },
    {
      "entropy": 5.401674222946167,
      "epoch": 2.30740719941596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000447454015788394,
      "loss": 5.0825,
      "mean_token_accuracy": 0.20572948902845384,
      "num_tokens": 45858171.0,
      "step": 20545
    },
    {
      "entropy": 5.42176570892334,
      "epoch": 2.3079687763239174,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004474282259500992,
      "loss": 5.0928,
      "mean_token_accuracy": 0.20348020046949386,
      "num_tokens": 45868715.0,
      "step": 20550
    },
    {
      "entropy": 5.4375592231750485,
      "epoch": 2.308530353231875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000447402430621623,
      "loss": 5.0823,
      "mean_token_accuracy": 0.21181593537330629,
      "num_tokens": 45880569.0,
      "step": 20555
    },
    {
      "entropy": 5.443147325515747,
      "epoch": 2.3090919301398327,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004473766298037865,
      "loss": 5.0969,
      "mean_token_accuracy": 0.20375711023807525,
      "num_tokens": 45890452.0,
      "step": 20560
    },
    {
      "entropy": 5.499129438400269,
      "epoch": 2.3096535070477904,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044735082349741116,
      "loss": 5.1987,
      "mean_token_accuracy": 0.1998823270201683,
      "num_tokens": 45903125.0,
      "step": 20565
    },
    {
      "entropy": 5.407380628585815,
      "epoch": 2.3102150839557476,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044732501170331886,
      "loss": 5.0801,
      "mean_token_accuracy": 0.20521142929792405,
      "num_tokens": 45914961.0,
      "step": 20570
    },
    {
      "entropy": 5.4881196975708,
      "epoch": 2.3107766608637053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004472991944223313,
      "loss": 5.2333,
      "mean_token_accuracy": 0.2012341096997261,
      "num_tokens": 45927198.0,
      "step": 20575
    },
    {
      "entropy": 5.484736537933349,
      "epoch": 2.311338237771663,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044727337165527056,
      "loss": 5.193,
      "mean_token_accuracy": 0.20932626724243164,
      "num_tokens": 45937937.0,
      "step": 20580
    },
    {
      "entropy": 5.538526439666748,
      "epoch": 2.31189981467962,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044724754340295887,
      "loss": 5.1459,
      "mean_token_accuracy": 0.19765277504920958,
      "num_tokens": 45948983.0,
      "step": 20585
    },
    {
      "entropy": 5.451184225082398,
      "epoch": 2.312461391587578,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004472217096662186,
      "loss": 5.2091,
      "mean_token_accuracy": 0.1980422854423523,
      "num_tokens": 45960674.0,
      "step": 20590
    },
    {
      "entropy": 5.428760814666748,
      "epoch": 2.3130229684955355,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004471958704458723,
      "loss": 5.1023,
      "mean_token_accuracy": 0.1996896296739578,
      "num_tokens": 45971260.0,
      "step": 20595
    },
    {
      "entropy": 5.5292823791503904,
      "epoch": 2.313584545403493,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044717002574274267,
      "loss": 5.1505,
      "mean_token_accuracy": 0.2068763941526413,
      "num_tokens": 45983942.0,
      "step": 20600
    },
    {
      "entropy": 5.475135469436646,
      "epoch": 2.314146122311451,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004471441755576527,
      "loss": 5.189,
      "mean_token_accuracy": 0.1958257347345352,
      "num_tokens": 45995983.0,
      "step": 20605
    },
    {
      "entropy": 5.418174982070923,
      "epoch": 2.314707699219408,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044711831989142534,
      "loss": 5.092,
      "mean_token_accuracy": 0.20370739102363586,
      "num_tokens": 46007074.0,
      "step": 20610
    },
    {
      "entropy": 5.4555549144744875,
      "epoch": 2.3152692761273657,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044709245874488405,
      "loss": 5.1555,
      "mean_token_accuracy": 0.2058824837207794,
      "num_tokens": 46018376.0,
      "step": 20615
    },
    {
      "entropy": 5.444838523864746,
      "epoch": 2.3158308530353233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004470665921188519,
      "loss": 5.1074,
      "mean_token_accuracy": 0.20470893383026123,
      "num_tokens": 46029192.0,
      "step": 20620
    },
    {
      "entropy": 5.393720579147339,
      "epoch": 2.3163924299432805,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044704072001415285,
      "loss": 5.1548,
      "mean_token_accuracy": 0.2058490812778473,
      "num_tokens": 46040735.0,
      "step": 20625
    },
    {
      "entropy": 5.363713550567627,
      "epoch": 2.316954006851238,
      "grad_norm": 1.125,
      "learning_rate": 0.00044701484243161055,
      "loss": 5.0262,
      "mean_token_accuracy": 0.2093999281525612,
      "num_tokens": 46050859.0,
      "step": 20630
    },
    {
      "entropy": 5.349782466888428,
      "epoch": 2.317515583759196,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000446988959372049,
      "loss": 5.1024,
      "mean_token_accuracy": 0.20917047411203385,
      "num_tokens": 46060704.0,
      "step": 20635
    },
    {
      "entropy": 5.362698698043824,
      "epoch": 2.3180771606671535,
      "grad_norm": 1.125,
      "learning_rate": 0.0004469630708362922,
      "loss": 5.1182,
      "mean_token_accuracy": 0.2047714814543724,
      "num_tokens": 46072317.0,
      "step": 20640
    },
    {
      "entropy": 5.519372749328613,
      "epoch": 2.3186387375751107,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004469371768251646,
      "loss": 5.2482,
      "mean_token_accuracy": 0.1977435290813446,
      "num_tokens": 46083823.0,
      "step": 20645
    },
    {
      "entropy": 5.547029685974121,
      "epoch": 2.3192003144830684,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004469112773394906,
      "loss": 5.1796,
      "mean_token_accuracy": 0.20172752439975739,
      "num_tokens": 46093920.0,
      "step": 20650
    },
    {
      "entropy": 5.424487447738647,
      "epoch": 2.319761891391026,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004468853723800948,
      "loss": 5.1026,
      "mean_token_accuracy": 0.21028678268194198,
      "num_tokens": 46104618.0,
      "step": 20655
    },
    {
      "entropy": 5.381516170501709,
      "epoch": 2.3203234682989837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004468594619478021,
      "loss": 4.995,
      "mean_token_accuracy": 0.21107428073883056,
      "num_tokens": 46114748.0,
      "step": 20660
    },
    {
      "entropy": 5.436009788513184,
      "epoch": 2.320885045206941,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004468335460434375,
      "loss": 5.1681,
      "mean_token_accuracy": 0.20895057618618013,
      "num_tokens": 46125125.0,
      "step": 20665
    },
    {
      "entropy": 5.323436641693116,
      "epoch": 2.3214466221148986,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004468076246678261,
      "loss": 5.0548,
      "mean_token_accuracy": 0.2040531173348427,
      "num_tokens": 46135877.0,
      "step": 20670
    },
    {
      "entropy": 5.485434103012085,
      "epoch": 2.3220081990228563,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044678169782179334,
      "loss": 5.2621,
      "mean_token_accuracy": 0.19618668854236604,
      "num_tokens": 46146769.0,
      "step": 20675
    },
    {
      "entropy": 5.43642225265503,
      "epoch": 2.3225697759308135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044675576550616466,
      "loss": 5.1101,
      "mean_token_accuracy": 0.20431993901729584,
      "num_tokens": 46158408.0,
      "step": 20680
    },
    {
      "entropy": 5.533440589904785,
      "epoch": 2.323131352838771,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044672982772176584,
      "loss": 5.2539,
      "mean_token_accuracy": 0.19789442867040635,
      "num_tokens": 46169961.0,
      "step": 20685
    },
    {
      "entropy": 5.5149883270263675,
      "epoch": 2.323692929746729,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004467038844694227,
      "loss": 5.1894,
      "mean_token_accuracy": 0.20298723876476288,
      "num_tokens": 46181750.0,
      "step": 20690
    },
    {
      "entropy": 5.432737255096436,
      "epoch": 2.3242545066546865,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004466779357499612,
      "loss": 5.0748,
      "mean_token_accuracy": 0.20784169733524321,
      "num_tokens": 46192745.0,
      "step": 20695
    },
    {
      "entropy": 5.385699129104614,
      "epoch": 2.324816083562644,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044665198156420774,
      "loss": 5.1443,
      "mean_token_accuracy": 0.20713905692100526,
      "num_tokens": 46203267.0,
      "step": 20700
    },
    {
      "entropy": 5.425318336486816,
      "epoch": 2.3253776604706013,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044662602191298856,
      "loss": 5.1223,
      "mean_token_accuracy": 0.2082781419157982,
      "num_tokens": 46214966.0,
      "step": 20705
    },
    {
      "entropy": 5.37998251914978,
      "epoch": 2.325939237378559,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044660005679713027,
      "loss": 5.0729,
      "mean_token_accuracy": 0.20619067698717117,
      "num_tokens": 46225755.0,
      "step": 20710
    },
    {
      "entropy": 5.43026328086853,
      "epoch": 2.3265008142865167,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044657408621745964,
      "loss": 5.1657,
      "mean_token_accuracy": 0.2067641317844391,
      "num_tokens": 46237091.0,
      "step": 20715
    },
    {
      "entropy": 5.465959119796753,
      "epoch": 2.327062391194474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004465481101748036,
      "loss": 5.1562,
      "mean_token_accuracy": 0.20297016501426696,
      "num_tokens": 46249250.0,
      "step": 20720
    },
    {
      "entropy": 5.441180801391601,
      "epoch": 2.3276239681024316,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044652212866998914,
      "loss": 5.0925,
      "mean_token_accuracy": 0.20590269863605498,
      "num_tokens": 46259866.0,
      "step": 20725
    },
    {
      "entropy": 5.41418309211731,
      "epoch": 2.328185545010389,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044649614170384354,
      "loss": 5.0675,
      "mean_token_accuracy": 0.21372539848089217,
      "num_tokens": 46270888.0,
      "step": 20730
    },
    {
      "entropy": 5.40954999923706,
      "epoch": 2.328747121918347,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044647014927719436,
      "loss": 5.136,
      "mean_token_accuracy": 0.20617228895425796,
      "num_tokens": 46281109.0,
      "step": 20735
    },
    {
      "entropy": 5.406164693832397,
      "epoch": 2.329308698826304,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044644415139086913,
      "loss": 4.9957,
      "mean_token_accuracy": 0.2161534085869789,
      "num_tokens": 46292171.0,
      "step": 20740
    },
    {
      "entropy": 5.413169527053833,
      "epoch": 2.3298702757342618,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004464181480456955,
      "loss": 5.1012,
      "mean_token_accuracy": 0.2029654785990715,
      "num_tokens": 46302919.0,
      "step": 20745
    },
    {
      "entropy": 5.311221170425415,
      "epoch": 2.3304318526422194,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004463921392425017,
      "loss": 5.0063,
      "mean_token_accuracy": 0.21227649748325347,
      "num_tokens": 46313199.0,
      "step": 20750
    },
    {
      "entropy": 5.490975570678711,
      "epoch": 2.330993429550177,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044636612498211566,
      "loss": 5.1682,
      "mean_token_accuracy": 0.20500044524669647,
      "num_tokens": 46324316.0,
      "step": 20755
    },
    {
      "entropy": 5.42360668182373,
      "epoch": 2.3315550064581343,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044634010526536575,
      "loss": 5.1333,
      "mean_token_accuracy": 0.19894029945135117,
      "num_tokens": 46335767.0,
      "step": 20760
    },
    {
      "entropy": 5.372499132156372,
      "epoch": 2.332116583366092,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004463140800930804,
      "loss": 5.0705,
      "mean_token_accuracy": 0.2019847124814987,
      "num_tokens": 46347840.0,
      "step": 20765
    },
    {
      "entropy": 5.401471328735352,
      "epoch": 2.3326781602740496,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004462880494660883,
      "loss": 5.2156,
      "mean_token_accuracy": 0.20026473551988602,
      "num_tokens": 46359766.0,
      "step": 20770
    },
    {
      "entropy": 5.496323347091675,
      "epoch": 2.333239737182007,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044626201338521834,
      "loss": 5.1211,
      "mean_token_accuracy": 0.2147686406970024,
      "num_tokens": 46371426.0,
      "step": 20775
    },
    {
      "entropy": 5.483458709716797,
      "epoch": 2.3338013140899645,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004462359718512993,
      "loss": 5.1667,
      "mean_token_accuracy": 0.20436545163393022,
      "num_tokens": 46382257.0,
      "step": 20780
    },
    {
      "entropy": 5.438762044906616,
      "epoch": 2.334362890997922,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044620992486516073,
      "loss": 5.1385,
      "mean_token_accuracy": 0.20005451142787933,
      "num_tokens": 46393689.0,
      "step": 20785
    },
    {
      "entropy": 5.419998693466186,
      "epoch": 2.33492446790588,
      "grad_norm": 1.125,
      "learning_rate": 0.0004461838724276316,
      "loss": 5.1764,
      "mean_token_accuracy": 0.20407554209232331,
      "num_tokens": 46404856.0,
      "step": 20790
    },
    {
      "entropy": 5.570212554931641,
      "epoch": 2.3354860448138375,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044615781453954164,
      "loss": 5.1697,
      "mean_token_accuracy": 0.197558256983757,
      "num_tokens": 46415395.0,
      "step": 20795
    },
    {
      "entropy": 5.564279842376709,
      "epoch": 2.3360476217217947,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044613175120172035,
      "loss": 5.2006,
      "mean_token_accuracy": 0.1974046438932419,
      "num_tokens": 46426773.0,
      "step": 20800
    },
    {
      "entropy": 5.355666637420654,
      "epoch": 2.3366091986297524,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004461056824149979,
      "loss": 5.0473,
      "mean_token_accuracy": 0.2046341374516487,
      "num_tokens": 46438206.0,
      "step": 20805
    },
    {
      "entropy": 5.401529598236084,
      "epoch": 2.33717077553771,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000446079608180204,
      "loss": 5.1575,
      "mean_token_accuracy": 0.20468377321958542,
      "num_tokens": 46449922.0,
      "step": 20810
    },
    {
      "entropy": 5.438352012634278,
      "epoch": 2.3377323524456672,
      "grad_norm": 1.125,
      "learning_rate": 0.0004460535284981692,
      "loss": 5.0886,
      "mean_token_accuracy": 0.2068081885576248,
      "num_tokens": 46461715.0,
      "step": 20815
    },
    {
      "entropy": 5.434967184066773,
      "epoch": 2.338293929353625,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004460274433697235,
      "loss": 5.0867,
      "mean_token_accuracy": 0.20653095543384553,
      "num_tokens": 46473136.0,
      "step": 20820
    },
    {
      "entropy": 5.37169508934021,
      "epoch": 2.3388555062615826,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044600135279569784,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21270499229431153,
      "num_tokens": 46483644.0,
      "step": 20825
    },
    {
      "entropy": 5.464639043807983,
      "epoch": 2.3394170831695402,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044597525677692273,
      "loss": 5.1614,
      "mean_token_accuracy": 0.2070650801062584,
      "num_tokens": 46494896.0,
      "step": 20830
    },
    {
      "entropy": 5.470511770248413,
      "epoch": 2.3399786600774974,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004459491553142291,
      "loss": 5.1559,
      "mean_token_accuracy": 0.20189064741134644,
      "num_tokens": 46505415.0,
      "step": 20835
    },
    {
      "entropy": 5.409118700027466,
      "epoch": 2.340540236985455,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044592304840844813,
      "loss": 5.1278,
      "mean_token_accuracy": 0.2101516142487526,
      "num_tokens": 46516280.0,
      "step": 20840
    },
    {
      "entropy": 5.339461994171143,
      "epoch": 2.3411018138934128,
      "grad_norm": 1.109375,
      "learning_rate": 0.000445896936060411,
      "loss": 5.0047,
      "mean_token_accuracy": 0.21017466485500336,
      "num_tokens": 46526456.0,
      "step": 20845
    },
    {
      "entropy": 5.474007940292358,
      "epoch": 2.3416633908013704,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044587081827094906,
      "loss": 5.1585,
      "mean_token_accuracy": 0.20065757185220717,
      "num_tokens": 46537547.0,
      "step": 20850
    },
    {
      "entropy": 5.374297952651977,
      "epoch": 2.3422249677093276,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044584469504089407,
      "loss": 5.0514,
      "mean_token_accuracy": 0.2099015712738037,
      "num_tokens": 46549424.0,
      "step": 20855
    },
    {
      "entropy": 5.443913841247559,
      "epoch": 2.3427865446172853,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004458185663710777,
      "loss": 5.0837,
      "mean_token_accuracy": 0.20638689547777175,
      "num_tokens": 46560234.0,
      "step": 20860
    },
    {
      "entropy": 5.443848466873169,
      "epoch": 2.343348121525243,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004457924322623319,
      "loss": 5.1186,
      "mean_token_accuracy": 0.201383376121521,
      "num_tokens": 46570847.0,
      "step": 20865
    },
    {
      "entropy": 5.340955400466919,
      "epoch": 2.3439096984332,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044576629271548884,
      "loss": 5.1095,
      "mean_token_accuracy": 0.2021559715270996,
      "num_tokens": 46582379.0,
      "step": 20870
    },
    {
      "entropy": 5.408187294006348,
      "epoch": 2.344471275341158,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004457401477313808,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19638585895299912,
      "num_tokens": 46594669.0,
      "step": 20875
    },
    {
      "entropy": 5.299769401550293,
      "epoch": 2.3450328522491155,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004457139973108403,
      "loss": 4.9937,
      "mean_token_accuracy": 0.21712994277477266,
      "num_tokens": 46605834.0,
      "step": 20880
    },
    {
      "entropy": 5.386976528167724,
      "epoch": 2.345594429157073,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004456878414546998,
      "loss": 5.115,
      "mean_token_accuracy": 0.20739735066890716,
      "num_tokens": 46616684.0,
      "step": 20885
    },
    {
      "entropy": 5.450810432434082,
      "epoch": 2.346156006065031,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044566168016379234,
      "loss": 5.1495,
      "mean_token_accuracy": 0.20875076949596405,
      "num_tokens": 46628045.0,
      "step": 20890
    },
    {
      "entropy": 5.401967525482178,
      "epoch": 2.346717582972988,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004456355134389507,
      "loss": 5.1111,
      "mean_token_accuracy": 0.20266646146774292,
      "num_tokens": 46639455.0,
      "step": 20895
    },
    {
      "entropy": 5.559433126449585,
      "epoch": 2.3472791598809457,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044560934128100814,
      "loss": 5.3275,
      "mean_token_accuracy": 0.19512481838464737,
      "num_tokens": 46651025.0,
      "step": 20900
    },
    {
      "entropy": 5.488357830047607,
      "epoch": 2.3478407367889034,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044558316369079805,
      "loss": 5.1212,
      "mean_token_accuracy": 0.20403787791728972,
      "num_tokens": 46661859.0,
      "step": 20905
    },
    {
      "entropy": 5.472838401794434,
      "epoch": 2.3484023136968606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044555698066915385,
      "loss": 5.1046,
      "mean_token_accuracy": 0.2062048837542534,
      "num_tokens": 46673441.0,
      "step": 20910
    },
    {
      "entropy": 5.4575324058532715,
      "epoch": 2.3489638906048182,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004455307922169092,
      "loss": 5.185,
      "mean_token_accuracy": 0.20046877562999726,
      "num_tokens": 46684959.0,
      "step": 20915
    },
    {
      "entropy": 5.463108539581299,
      "epoch": 2.349525467512776,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004455045983348979,
      "loss": 5.1897,
      "mean_token_accuracy": 0.21050703525543213,
      "num_tokens": 46694940.0,
      "step": 20920
    },
    {
      "entropy": 5.426689386367798,
      "epoch": 2.3500870444207336,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004454783990239542,
      "loss": 5.0636,
      "mean_token_accuracy": 0.21039489954710006,
      "num_tokens": 46706184.0,
      "step": 20925
    },
    {
      "entropy": 5.438544273376465,
      "epoch": 2.350648621328691,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044545219428491215,
      "loss": 5.1314,
      "mean_token_accuracy": 0.20452527850866317,
      "num_tokens": 46718267.0,
      "step": 20930
    },
    {
      "entropy": 5.434673738479614,
      "epoch": 2.3512101982366485,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044542598411860605,
      "loss": 5.1412,
      "mean_token_accuracy": 0.2029553547501564,
      "num_tokens": 46729293.0,
      "step": 20935
    },
    {
      "entropy": 5.418290853500366,
      "epoch": 2.351771775144606,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004453997685258706,
      "loss": 5.0982,
      "mean_token_accuracy": 0.20700334012508392,
      "num_tokens": 46740214.0,
      "step": 20940
    },
    {
      "entropy": 5.487683725357056,
      "epoch": 2.3523333520525638,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044537354750754033,
      "loss": 5.1649,
      "mean_token_accuracy": 0.20408000648021699,
      "num_tokens": 46751269.0,
      "step": 20945
    },
    {
      "entropy": 5.374183988571167,
      "epoch": 2.352894928960521,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004453473210644503,
      "loss": 4.9544,
      "mean_token_accuracy": 0.2161452889442444,
      "num_tokens": 46761851.0,
      "step": 20950
    },
    {
      "entropy": 5.3502898693084715,
      "epoch": 2.3534565058684787,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004453210891974355,
      "loss": 5.0943,
      "mean_token_accuracy": 0.20845457017421723,
      "num_tokens": 46773441.0,
      "step": 20955
    },
    {
      "entropy": 5.328681230545044,
      "epoch": 2.3540180827764363,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044529485190733113,
      "loss": 5.0807,
      "mean_token_accuracy": 0.20720956325531006,
      "num_tokens": 46785016.0,
      "step": 20960
    },
    {
      "entropy": 5.452505350112915,
      "epoch": 2.3545796596843935,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004452686091949727,
      "loss": 5.1403,
      "mean_token_accuracy": 0.19733494967222215,
      "num_tokens": 46796059.0,
      "step": 20965
    },
    {
      "entropy": 5.37146406173706,
      "epoch": 2.355141236592351,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044524236106119563,
      "loss": 5.0512,
      "mean_token_accuracy": 0.20761397182941438,
      "num_tokens": 46805725.0,
      "step": 20970
    },
    {
      "entropy": 5.402361822128296,
      "epoch": 2.355702813500309,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044521610750683586,
      "loss": 5.0597,
      "mean_token_accuracy": 0.20883367508649825,
      "num_tokens": 46816300.0,
      "step": 20975
    },
    {
      "entropy": 5.411523389816284,
      "epoch": 2.3562643904082665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004451898485327291,
      "loss": 5.0879,
      "mean_token_accuracy": 0.19941720962524415,
      "num_tokens": 46827498.0,
      "step": 20980
    },
    {
      "entropy": 5.410050582885742,
      "epoch": 2.356825967316224,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004451635841397116,
      "loss": 5.1061,
      "mean_token_accuracy": 0.19964056015014647,
      "num_tokens": 46838416.0,
      "step": 20985
    },
    {
      "entropy": 5.456202602386474,
      "epoch": 2.3573875442241814,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004451373143286195,
      "loss": 5.1931,
      "mean_token_accuracy": 0.20411793291568756,
      "num_tokens": 46849371.0,
      "step": 20990
    },
    {
      "entropy": 5.40569429397583,
      "epoch": 2.357949121132139,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044511103910028945,
      "loss": 5.1362,
      "mean_token_accuracy": 0.20762857049703598,
      "num_tokens": 46860095.0,
      "step": 20995
    },
    {
      "entropy": 5.49809455871582,
      "epoch": 2.3585106980400967,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004450847584555579,
      "loss": 5.2129,
      "mean_token_accuracy": 0.1987141713500023,
      "num_tokens": 46871217.0,
      "step": 21000
    },
    {
      "epoch": 2.3585106980400967,
      "eval_entropy": 5.279021420797368,
      "eval_loss": 5.261969089508057,
      "eval_mean_token_accuracy": 0.206022280616446,
      "eval_num_tokens": 46871217.0,
      "eval_runtime": 59.594,
      "eval_samples_per_second": 520.405,
      "eval_steps_per_second": 65.057,
      "step": 21000
    }
  ],
  "logging_steps": 5,
  "max_steps": 89030,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.3389513544704e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}