eng-latn-100mb-after-ppt-shuff-dyck-disjoint-tok-ckpt500_seed3407_seed3407/checkpoint-9000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.010782276632785,
  "eval_steps": 3000,
  "global_step": 9000,
  "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
    }
  ],
  "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": 3.146523558912e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}