{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.788741860953581,
  "eval_steps": 3000,
  "global_step": 57000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.253151750564575,
      "epoch": 0.0004200798151648813,
      "grad_norm": 1.2578125,
      "learning_rate": 2e-06,
      "loss": 6.8984,
      "mean_token_accuracy": 0.09029425084590911,
      "num_tokens": 8348.0,
      "step": 5
    },
    {
      "entropy": 7.219348526000976,
      "epoch": 0.0008401596303297626,
      "grad_norm": 1.3359375,
      "learning_rate": 4.5e-06,
      "loss": 6.9563,
      "mean_token_accuracy": 0.0939315177500248,
      "num_tokens": 17465.0,
      "step": 10
    },
    {
      "entropy": 7.203219985961914,
      "epoch": 0.001260239445494644,
      "grad_norm": 1.1171875,
      "learning_rate": 7e-06,
      "loss": 6.9004,
      "mean_token_accuracy": 0.08940818756818772,
      "num_tokens": 26627.0,
      "step": 15
    },
    {
      "entropy": 7.224360418319702,
      "epoch": 0.0016803192606595252,
      "grad_norm": 1.1640625,
      "learning_rate": 9.5e-06,
      "loss": 6.9259,
      "mean_token_accuracy": 0.09512931853532791,
      "num_tokens": 36069.0,
      "step": 20
    },
    {
      "entropy": 7.186052703857422,
      "epoch": 0.002100399075824407,
      "grad_norm": 1.2734375,
      "learning_rate": 1.2e-05,
      "loss": 6.8641,
      "mean_token_accuracy": 0.09390396177768708,
      "num_tokens": 44967.0,
      "step": 25
    },
    {
      "entropy": 7.151274633407593,
      "epoch": 0.002520478890989288,
      "grad_norm": 1.1875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 6.964,
      "mean_token_accuracy": 0.08688623458147049,
      "num_tokens": 55132.0,
      "step": 30
    },
    {
      "entropy": 7.053877639770508,
      "epoch": 0.0029405587061541692,
      "grad_norm": 1.0546875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 6.7637,
      "mean_token_accuracy": 0.09662552699446678,
      "num_tokens": 65141.0,
      "step": 35
    },
    {
      "entropy": 7.099168729782105,
      "epoch": 0.0033606385213190504,
      "grad_norm": 1.0625,
      "learning_rate": 1.95e-05,
      "loss": 6.8446,
      "mean_token_accuracy": 0.09567792639136315,
      "num_tokens": 74007.0,
      "step": 40
    },
    {
      "entropy": 7.140014743804931,
      "epoch": 0.003780718336483932,
      "grad_norm": 1.078125,
      "learning_rate": 2.2e-05,
      "loss": 6.8569,
      "mean_token_accuracy": 0.09555562734603881,
      "num_tokens": 83736.0,
      "step": 45
    },
    {
      "entropy": 7.1646524429321286,
      "epoch": 0.004200798151648814,
      "grad_norm": 1.0390625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.8505,
      "mean_token_accuracy": 0.09592381715774537,
      "num_tokens": 92525.0,
      "step": 50
    },
    {
      "entropy": 7.107996654510498,
      "epoch": 0.004620877966813695,
      "grad_norm": 1.0390625,
      "learning_rate": 2.7e-05,
      "loss": 6.7755,
      "mean_token_accuracy": 0.09664912968873977,
      "num_tokens": 102015.0,
      "step": 55
    },
    {
      "entropy": 7.137420606613159,
      "epoch": 0.005040957781978576,
      "grad_norm": 1.0859375,
      "learning_rate": 2.95e-05,
      "loss": 6.9768,
      "mean_token_accuracy": 0.0867392435669899,
      "num_tokens": 110887.0,
      "step": 60
    },
    {
      "entropy": 7.134914398193359,
      "epoch": 0.005461037597143457,
      "grad_norm": 1.1171875,
      "learning_rate": 3.2e-05,
      "loss": 6.9024,
      "mean_token_accuracy": 0.08935272470116615,
      "num_tokens": 120442.0,
      "step": 65
    },
    {
      "entropy": 7.1443713188171385,
      "epoch": 0.0058811174123083385,
      "grad_norm": 1.1953125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.8989,
      "mean_token_accuracy": 0.09712273105978966,
      "num_tokens": 129297.0,
      "step": 70
    },
    {
      "entropy": 7.160055351257324,
      "epoch": 0.00630119722747322,
      "grad_norm": 1.0390625,
      "learning_rate": 3.7e-05,
      "loss": 6.8507,
      "mean_token_accuracy": 0.09877990111708641,
      "num_tokens": 138305.0,
      "step": 75
    },
    {
      "entropy": 7.141705083847046,
      "epoch": 0.006721277042638101,
      "grad_norm": 1.0859375,
      "learning_rate": 3.95e-05,
      "loss": 6.8372,
      "mean_token_accuracy": 0.09292416870594025,
      "num_tokens": 147640.0,
      "step": 80
    },
    {
      "entropy": 7.002462768554688,
      "epoch": 0.007141356857802983,
      "grad_norm": 1.078125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.75,
      "mean_token_accuracy": 0.09611514061689377,
      "num_tokens": 157633.0,
      "step": 85
    },
    {
      "entropy": 7.100855731964112,
      "epoch": 0.007561436672967864,
      "grad_norm": 0.96875,
      "learning_rate": 4.45e-05,
      "loss": 6.7673,
      "mean_token_accuracy": 0.08827547580003739,
      "num_tokens": 167984.0,
      "step": 90
    },
    {
      "entropy": 7.129158973693848,
      "epoch": 0.007981516488132745,
      "grad_norm": 1.0234375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 6.8948,
      "mean_token_accuracy": 0.099730733782053,
      "num_tokens": 176984.0,
      "step": 95
    },
    {
      "entropy": 7.069277429580689,
      "epoch": 0.008401596303297627,
      "grad_norm": 1.1796875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.7542,
      "mean_token_accuracy": 0.0975344181060791,
      "num_tokens": 185931.0,
      "step": 100
    },
    {
      "entropy": 7.183401107788086,
      "epoch": 0.008821676118462508,
      "grad_norm": 1.125,
      "learning_rate": 5.2e-05,
      "loss": 6.8312,
      "mean_token_accuracy": 0.09061248749494552,
      "num_tokens": 195065.0,
      "step": 105
    },
    {
      "entropy": 7.100264024734497,
      "epoch": 0.00924175593362739,
      "grad_norm": 1.1640625,
      "learning_rate": 5.45e-05,
      "loss": 6.8219,
      "mean_token_accuracy": 0.09686729088425636,
      "num_tokens": 203687.0,
      "step": 110
    },
    {
      "entropy": 7.042130661010742,
      "epoch": 0.00966183574879227,
      "grad_norm": 1.109375,
      "learning_rate": 5.7e-05,
      "loss": 6.8491,
      "mean_token_accuracy": 0.09541062936186791,
      "num_tokens": 212847.0,
      "step": 115
    },
    {
      "entropy": 7.127160167694091,
      "epoch": 0.010081915563957152,
      "grad_norm": 1.203125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.8242,
      "mean_token_accuracy": 0.0955698125064373,
      "num_tokens": 222593.0,
      "step": 120
    },
    {
      "entropy": 7.0331028461456295,
      "epoch": 0.010501995379122032,
      "grad_norm": 1.1875,
      "learning_rate": 6.2e-05,
      "loss": 6.711,
      "mean_token_accuracy": 0.0981454961001873,
      "num_tokens": 231174.0,
      "step": 125
    },
    {
      "entropy": 6.9479889392852785,
      "epoch": 0.010922075194286915,
      "grad_norm": 1.140625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.663,
      "mean_token_accuracy": 0.10388810336589813,
      "num_tokens": 239833.0,
      "step": 130
    },
    {
      "entropy": 7.190658855438232,
      "epoch": 0.011342155009451797,
      "grad_norm": 1.125,
      "learning_rate": 6.7e-05,
      "loss": 6.951,
      "mean_token_accuracy": 0.0949922852218151,
      "num_tokens": 248794.0,
      "step": 135
    },
    {
      "entropy": 7.251979541778565,
      "epoch": 0.011762234824616677,
      "grad_norm": 1.2734375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.9449,
      "mean_token_accuracy": 0.09376412332057953,
      "num_tokens": 257123.0,
      "step": 140
    },
    {
      "entropy": 7.008625459671021,
      "epoch": 0.012182314639781559,
      "grad_norm": 1.1171875,
      "learning_rate": 7.2e-05,
      "loss": 6.6678,
      "mean_token_accuracy": 0.10136394873261452,
      "num_tokens": 266088.0,
      "step": 145
    },
    {
      "entropy": 7.118732166290283,
      "epoch": 0.01260239445494644,
      "grad_norm": 1.0625,
      "learning_rate": 7.45e-05,
      "loss": 6.9922,
      "mean_token_accuracy": 0.08856561928987502,
      "num_tokens": 276074.0,
      "step": 150
    },
    {
      "entropy": 7.265724229812622,
      "epoch": 0.013022474270111321,
      "grad_norm": 1.078125,
      "learning_rate": 7.7e-05,
      "loss": 6.8415,
      "mean_token_accuracy": 0.09575222656130791,
      "num_tokens": 285280.0,
      "step": 155
    },
    {
      "entropy": 7.09201340675354,
      "epoch": 0.013442554085276202,
      "grad_norm": 1.09375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.9508,
      "mean_token_accuracy": 0.08855971023440361,
      "num_tokens": 296115.0,
      "step": 160
    },
    {
      "entropy": 7.172540521621704,
      "epoch": 0.013862633900441084,
      "grad_norm": 1.03125,
      "learning_rate": 8.2e-05,
      "loss": 6.7833,
      "mean_token_accuracy": 0.09475113973021507,
      "num_tokens": 305483.0,
      "step": 165
    },
    {
      "entropy": 7.142146444320678,
      "epoch": 0.014282713715605966,
      "grad_norm": 1.0859375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.798,
      "mean_token_accuracy": 0.09611742347478866,
      "num_tokens": 314000.0,
      "step": 170
    },
    {
      "entropy": 7.191121196746826,
      "epoch": 0.014702793530770846,
      "grad_norm": 1.1171875,
      "learning_rate": 8.7e-05,
      "loss": 6.8719,
      "mean_token_accuracy": 0.09523176699876786,
      "num_tokens": 323667.0,
      "step": 175
    },
    {
      "entropy": 7.111018800735474,
      "epoch": 0.015122873345935728,
      "grad_norm": 1.1640625,
      "learning_rate": 8.95e-05,
      "loss": 6.9126,
      "mean_token_accuracy": 0.09633364677429199,
      "num_tokens": 332695.0,
      "step": 180
    },
    {
      "entropy": 7.152137660980225,
      "epoch": 0.015542953161100609,
      "grad_norm": 1.265625,
      "learning_rate": 9.2e-05,
      "loss": 6.7081,
      "mean_token_accuracy": 0.09874472171068191,
      "num_tokens": 342428.0,
      "step": 185
    },
    {
      "entropy": 7.018444013595581,
      "epoch": 0.01596303297626549,
      "grad_norm": 1.0390625,
      "learning_rate": 9.45e-05,
      "loss": 6.9196,
      "mean_token_accuracy": 0.09128761291503906,
      "num_tokens": 353587.0,
      "step": 190
    },
    {
      "entropy": 7.154490900039673,
      "epoch": 0.01638311279143037,
      "grad_norm": 1.234375,
      "learning_rate": 9.7e-05,
      "loss": 6.8259,
      "mean_token_accuracy": 0.09795344024896621,
      "num_tokens": 362997.0,
      "step": 195
    },
    {
      "entropy": 7.170652723312378,
      "epoch": 0.016803192606595255,
      "grad_norm": 1.0703125,
      "learning_rate": 9.95e-05,
      "loss": 6.8563,
      "mean_token_accuracy": 0.09546189531683921,
      "num_tokens": 372346.0,
      "step": 200
    },
    {
      "entropy": 7.110174036026001,
      "epoch": 0.017223272421760135,
      "grad_norm": 1.328125,
      "learning_rate": 0.000102,
      "loss": 6.741,
      "mean_token_accuracy": 0.09936807751655578,
      "num_tokens": 381575.0,
      "step": 205
    },
    {
      "entropy": 7.106418037414551,
      "epoch": 0.017643352236925015,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.8292,
      "mean_token_accuracy": 0.09433561563491821,
      "num_tokens": 390706.0,
      "step": 210
    },
    {
      "entropy": 7.192156839370727,
      "epoch": 0.018063432052089896,
      "grad_norm": 1.140625,
      "learning_rate": 0.000107,
      "loss": 6.8516,
      "mean_token_accuracy": 0.0938378892838955,
      "num_tokens": 400000.0,
      "step": 215
    },
    {
      "entropy": 6.9766045093536375,
      "epoch": 0.01848351186725478,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001095,
      "loss": 6.8445,
      "mean_token_accuracy": 0.10025399252772331,
      "num_tokens": 409447.0,
      "step": 220
    },
    {
      "entropy": 7.228094434738159,
      "epoch": 0.01890359168241966,
      "grad_norm": 1.34375,
      "learning_rate": 0.000112,
      "loss": 6.7628,
      "mean_token_accuracy": 0.09841207265853882,
      "num_tokens": 418417.0,
      "step": 225
    },
    {
      "entropy": 6.980171537399292,
      "epoch": 0.01932367149758454,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001145,
      "loss": 6.7865,
      "mean_token_accuracy": 0.10002906545996666,
      "num_tokens": 427619.0,
      "step": 230
    },
    {
      "entropy": 7.174957370758056,
      "epoch": 0.019743751312749424,
      "grad_norm": 1.09375,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.8595,
      "mean_token_accuracy": 0.09506258964538575,
      "num_tokens": 437931.0,
      "step": 235
    },
    {
      "entropy": 7.124919366836548,
      "epoch": 0.020163831127914304,
      "grad_norm": 1.203125,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.8624,
      "mean_token_accuracy": 0.10153809040784836,
      "num_tokens": 447595.0,
      "step": 240
    },
    {
      "entropy": 7.076248693466186,
      "epoch": 0.020583910943079185,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000122,
      "loss": 6.7534,
      "mean_token_accuracy": 0.09595257192850112,
      "num_tokens": 457062.0,
      "step": 245
    },
    {
      "entropy": 7.11194372177124,
      "epoch": 0.021003990758244065,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001245,
      "loss": 6.8486,
      "mean_token_accuracy": 0.09663526639342308,
      "num_tokens": 466191.0,
      "step": 250
    },
    {
      "entropy": 7.119431734085083,
      "epoch": 0.02142407057340895,
      "grad_norm": 1.40625,
      "learning_rate": 0.000127,
      "loss": 6.8411,
      "mean_token_accuracy": 0.09689914286136628,
      "num_tokens": 475693.0,
      "step": 255
    },
    {
      "entropy": 7.101942634582519,
      "epoch": 0.02184415038857383,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001295,
      "loss": 6.8733,
      "mean_token_accuracy": 0.0926995851099491,
      "num_tokens": 485173.0,
      "step": 260
    },
    {
      "entropy": 7.02587628364563,
      "epoch": 0.02226423020373871,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000132,
      "loss": 6.7261,
      "mean_token_accuracy": 0.1030467577278614,
      "num_tokens": 493985.0,
      "step": 265
    },
    {
      "entropy": 7.204363059997559,
      "epoch": 0.022684310018903593,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.8283,
      "mean_token_accuracy": 0.09744107499718666,
      "num_tokens": 502837.0,
      "step": 270
    },
    {
      "entropy": 7.018724775314331,
      "epoch": 0.023104389834068473,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.7299,
      "mean_token_accuracy": 0.10230938643217087,
      "num_tokens": 511503.0,
      "step": 275
    },
    {
      "entropy": 7.203450679779053,
      "epoch": 0.023524469649233354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001395,
      "loss": 6.9591,
      "mean_token_accuracy": 0.09394779950380325,
      "num_tokens": 521499.0,
      "step": 280
    },
    {
      "entropy": 7.085290002822876,
      "epoch": 0.023944549464398234,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.7304,
      "mean_token_accuracy": 0.09568566456437111,
      "num_tokens": 530067.0,
      "step": 285
    },
    {
      "entropy": 7.13321099281311,
      "epoch": 0.024364629279563118,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001445,
      "loss": 6.7422,
      "mean_token_accuracy": 0.10064690634608268,
      "num_tokens": 538559.0,
      "step": 290
    },
    {
      "entropy": 7.135575151443481,
      "epoch": 0.024784709094728,
      "grad_norm": 1.25,
      "learning_rate": 0.000147,
      "loss": 6.9403,
      "mean_token_accuracy": 0.09277286529541015,
      "num_tokens": 547288.0,
      "step": 295
    },
    {
      "entropy": 7.132419538497925,
      "epoch": 0.02520478890989288,
      "grad_norm": 1.125,
      "learning_rate": 0.0001495,
      "loss": 6.8165,
      "mean_token_accuracy": 0.09548124819993972,
      "num_tokens": 557269.0,
      "step": 300
    },
    {
      "entropy": 7.107891798019409,
      "epoch": 0.025624868725057762,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000152,
      "loss": 6.8392,
      "mean_token_accuracy": 0.09125733524560928,
      "num_tokens": 567280.0,
      "step": 305
    },
    {
      "entropy": 7.009502935409546,
      "epoch": 0.026044948540222643,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.6443,
      "mean_token_accuracy": 0.0992837019264698,
      "num_tokens": 576609.0,
      "step": 310
    },
    {
      "entropy": 6.9137735843658445,
      "epoch": 0.026465028355387523,
      "grad_norm": 1.234375,
      "learning_rate": 0.000157,
      "loss": 6.627,
      "mean_token_accuracy": 0.1031131848692894,
      "num_tokens": 586053.0,
      "step": 315
    },
    {
      "entropy": 7.00633134841919,
      "epoch": 0.026885108170552403,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001595,
      "loss": 6.8295,
      "mean_token_accuracy": 0.09749070778489113,
      "num_tokens": 594649.0,
      "step": 320
    },
    {
      "entropy": 7.084966850280762,
      "epoch": 0.027305187985717287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000162,
      "loss": 6.7142,
      "mean_token_accuracy": 0.09451463893055916,
      "num_tokens": 603445.0,
      "step": 325
    },
    {
      "entropy": 7.088579750061035,
      "epoch": 0.027725267800882167,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.8658,
      "mean_token_accuracy": 0.09315285831689835,
      "num_tokens": 613611.0,
      "step": 330
    },
    {
      "entropy": 7.3623809814453125,
      "epoch": 0.028145347616047048,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.0713,
      "mean_token_accuracy": 0.0919966921210289,
      "num_tokens": 623024.0,
      "step": 335
    },
    {
      "entropy": 6.999190092086792,
      "epoch": 0.02856542743121193,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.7385,
      "mean_token_accuracy": 0.1017606370151043,
      "num_tokens": 631624.0,
      "step": 340
    },
    {
      "entropy": 6.994140338897705,
      "epoch": 0.028985507246376812,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.6633,
      "mean_token_accuracy": 0.10479742139577866,
      "num_tokens": 640473.0,
      "step": 345
    },
    {
      "entropy": 7.0762580871582035,
      "epoch": 0.029405587061541692,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9006,
      "mean_token_accuracy": 0.09429975003004074,
      "num_tokens": 649692.0,
      "step": 350
    },
    {
      "entropy": 7.142873811721802,
      "epoch": 0.029825666876706573,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000177,
      "loss": 6.8377,
      "mean_token_accuracy": 0.09720863476395607,
      "num_tokens": 658236.0,
      "step": 355
    },
    {
      "entropy": 6.952843904495239,
      "epoch": 0.030245746691871456,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001795,
      "loss": 6.6251,
      "mean_token_accuracy": 0.09688405320048332,
      "num_tokens": 667175.0,
      "step": 360
    },
    {
      "entropy": 7.14975872039795,
      "epoch": 0.030665826507036337,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000182,
      "loss": 6.9438,
      "mean_token_accuracy": 0.09247877895832061,
      "num_tokens": 676456.0,
      "step": 365
    },
    {
      "entropy": 7.224644327163697,
      "epoch": 0.031085906322201217,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001845,
      "loss": 6.926,
      "mean_token_accuracy": 0.08845363929867744,
      "num_tokens": 686881.0,
      "step": 370
    },
    {
      "entropy": 6.932204675674439,
      "epoch": 0.0315059861373661,
      "grad_norm": 1.203125,
      "learning_rate": 0.000187,
      "loss": 6.6985,
      "mean_token_accuracy": 0.09394535645842553,
      "num_tokens": 696045.0,
      "step": 375
    },
    {
      "entropy": 6.965549278259277,
      "epoch": 0.03192606595253098,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001895,
      "loss": 6.6653,
      "mean_token_accuracy": 0.10160319805145264,
      "num_tokens": 704729.0,
      "step": 380
    },
    {
      "entropy": 6.985969495773316,
      "epoch": 0.032346145767695865,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000192,
      "loss": 6.7426,
      "mean_token_accuracy": 0.0950203962624073,
      "num_tokens": 714331.0,
      "step": 385
    },
    {
      "entropy": 7.101364660263061,
      "epoch": 0.03276622558286074,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001945,
      "loss": 6.6912,
      "mean_token_accuracy": 0.10189466029405594,
      "num_tokens": 722788.0,
      "step": 390
    },
    {
      "entropy": 7.007110738754273,
      "epoch": 0.033186305398025626,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7849,
      "mean_token_accuracy": 0.10113223120570183,
      "num_tokens": 731417.0,
      "step": 395
    },
    {
      "entropy": 7.0138044357299805,
      "epoch": 0.03360638521319051,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.7534,
      "mean_token_accuracy": 0.09411026313900947,
      "num_tokens": 741034.0,
      "step": 400
    },
    {
      "entropy": 7.00489068031311,
      "epoch": 0.034026465028355386,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000202,
      "loss": 6.7516,
      "mean_token_accuracy": 0.09877323731780052,
      "num_tokens": 749596.0,
      "step": 405
    },
    {
      "entropy": 7.017527103424072,
      "epoch": 0.03444654484352027,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.7218,
      "mean_token_accuracy": 0.09416642934083938,
      "num_tokens": 758931.0,
      "step": 410
    },
    {
      "entropy": 6.871344518661499,
      "epoch": 0.03486662465868515,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000207,
      "loss": 6.6216,
      "mean_token_accuracy": 0.10105381533503532,
      "num_tokens": 767534.0,
      "step": 415
    },
    {
      "entropy": 6.960817480087281,
      "epoch": 0.03528670447385003,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002095,
      "loss": 6.6761,
      "mean_token_accuracy": 0.10064006224274635,
      "num_tokens": 776456.0,
      "step": 420
    },
    {
      "entropy": 7.008622884750366,
      "epoch": 0.035706784289014915,
      "grad_norm": 1.21875,
      "learning_rate": 0.000212,
      "loss": 6.7646,
      "mean_token_accuracy": 0.09697613269090652,
      "num_tokens": 786172.0,
      "step": 425
    },
    {
      "entropy": 6.960406351089477,
      "epoch": 0.03612686410417979,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002145,
      "loss": 6.6658,
      "mean_token_accuracy": 0.10456070601940155,
      "num_tokens": 795081.0,
      "step": 430
    },
    {
      "entropy": 6.916972398757935,
      "epoch": 0.036546943919344675,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6979,
      "mean_token_accuracy": 0.09141508191823959,
      "num_tokens": 804259.0,
      "step": 435
    },
    {
      "entropy": 6.980181550979614,
      "epoch": 0.03696702373450956,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002195,
      "loss": 6.7251,
      "mean_token_accuracy": 0.0985159382224083,
      "num_tokens": 813463.0,
      "step": 440
    },
    {
      "entropy": 6.952114248275757,
      "epoch": 0.037387103549674436,
      "grad_norm": 1.328125,
      "learning_rate": 0.000222,
      "loss": 6.6346,
      "mean_token_accuracy": 0.10426531285047531,
      "num_tokens": 823029.0,
      "step": 445
    },
    {
      "entropy": 6.91859712600708,
      "epoch": 0.03780718336483932,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002245,
      "loss": 6.7327,
      "mean_token_accuracy": 0.0944428451359272,
      "num_tokens": 832902.0,
      "step": 450
    },
    {
      "entropy": 6.92227520942688,
      "epoch": 0.0382272631800042,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.6724,
      "mean_token_accuracy": 0.10073406398296356,
      "num_tokens": 842162.0,
      "step": 455
    },
    {
      "entropy": 6.979531192779541,
      "epoch": 0.03864734299516908,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.7091,
      "mean_token_accuracy": 0.0995998091995716,
      "num_tokens": 852328.0,
      "step": 460
    },
    {
      "entropy": 6.8678590774536135,
      "epoch": 0.039067422810333964,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.6831,
      "mean_token_accuracy": 0.1023336872458458,
      "num_tokens": 860929.0,
      "step": 465
    },
    {
      "entropy": 7.0033402919769285,
      "epoch": 0.03948750262549885,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7492,
      "mean_token_accuracy": 0.09593449011445046,
      "num_tokens": 869144.0,
      "step": 470
    },
    {
      "entropy": 7.003532409667969,
      "epoch": 0.039907582440663725,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000237,
      "loss": 6.705,
      "mean_token_accuracy": 0.10385636389255523,
      "num_tokens": 877447.0,
      "step": 475
    },
    {
      "entropy": 6.86921706199646,
      "epoch": 0.04032766225582861,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002395,
      "loss": 6.6601,
      "mean_token_accuracy": 0.09642177075147629,
      "num_tokens": 887020.0,
      "step": 480
    },
    {
      "entropy": 6.996332120895386,
      "epoch": 0.040747742070993485,
      "grad_norm": 1.359375,
      "learning_rate": 0.000242,
      "loss": 6.7054,
      "mean_token_accuracy": 0.0975713811814785,
      "num_tokens": 895937.0,
      "step": 485
    },
    {
      "entropy": 6.852901887893677,
      "epoch": 0.04116782188615837,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002445,
      "loss": 6.7267,
      "mean_token_accuracy": 0.09650165066123009,
      "num_tokens": 905446.0,
      "step": 490
    },
    {
      "entropy": 6.893301010131836,
      "epoch": 0.04158790170132325,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000247,
      "loss": 6.6036,
      "mean_token_accuracy": 0.10643761828541756,
      "num_tokens": 914547.0,
      "step": 495
    },
    {
      "entropy": 6.915640449523925,
      "epoch": 0.04200798151648813,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002495,
      "loss": 6.6263,
      "mean_token_accuracy": 0.10527556240558625,
      "num_tokens": 922900.0,
      "step": 500
    },
    {
      "entropy": 6.947235059738159,
      "epoch": 0.042428061331653014,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000252,
      "loss": 6.6686,
      "mean_token_accuracy": 0.10355583727359771,
      "num_tokens": 930876.0,
      "step": 505
    },
    {
      "entropy": 6.88210015296936,
      "epoch": 0.0428481411468179,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002545,
      "loss": 6.7087,
      "mean_token_accuracy": 0.10312066823244095,
      "num_tokens": 939871.0,
      "step": 510
    },
    {
      "entropy": 6.947447443008423,
      "epoch": 0.043268220961982774,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000257,
      "loss": 6.695,
      "mean_token_accuracy": 0.10180827602744102,
      "num_tokens": 948673.0,
      "step": 515
    },
    {
      "entropy": 6.816449880599976,
      "epoch": 0.04368830077714766,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002595,
      "loss": 6.6488,
      "mean_token_accuracy": 0.09842450320720672,
      "num_tokens": 957603.0,
      "step": 520
    },
    {
      "entropy": 6.928069686889648,
      "epoch": 0.04410838059231254,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000262,
      "loss": 6.7274,
      "mean_token_accuracy": 0.09575201719999313,
      "num_tokens": 967731.0,
      "step": 525
    },
    {
      "entropy": 6.940513849258423,
      "epoch": 0.04452846040747742,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7098,
      "mean_token_accuracy": 0.10156730636954307,
      "num_tokens": 977427.0,
      "step": 530
    },
    {
      "entropy": 6.883533573150634,
      "epoch": 0.0449485402226423,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.7126,
      "mean_token_accuracy": 0.09694371595978737,
      "num_tokens": 986758.0,
      "step": 535
    },
    {
      "entropy": 7.054216384887695,
      "epoch": 0.045368620037807186,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7073,
      "mean_token_accuracy": 0.10619494765996933,
      "num_tokens": 996377.0,
      "step": 540
    },
    {
      "entropy": 6.823930788040161,
      "epoch": 0.04578869985297206,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.7762,
      "mean_token_accuracy": 0.09864854142069816,
      "num_tokens": 1006483.0,
      "step": 545
    },
    {
      "entropy": 6.839679384231568,
      "epoch": 0.04620877966813695,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002745,
      "loss": 6.6608,
      "mean_token_accuracy": 0.09898171871900559,
      "num_tokens": 1016132.0,
      "step": 550
    },
    {
      "entropy": 6.886328125,
      "epoch": 0.04662885948330183,
      "grad_norm": 1.265625,
      "learning_rate": 0.000277,
      "loss": 6.5878,
      "mean_token_accuracy": 0.10534627884626388,
      "num_tokens": 1024970.0,
      "step": 555
    },
    {
      "entropy": 6.892021656036377,
      "epoch": 0.04704893929846671,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002795,
      "loss": 6.7043,
      "mean_token_accuracy": 0.0967423141002655,
      "num_tokens": 1034335.0,
      "step": 560
    },
    {
      "entropy": 6.914703607559204,
      "epoch": 0.04746901911363159,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.756,
      "mean_token_accuracy": 0.10775318518280982,
      "num_tokens": 1043954.0,
      "step": 565
    },
    {
      "entropy": 6.942829847335815,
      "epoch": 0.04788909892879647,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002845,
      "loss": 6.6882,
      "mean_token_accuracy": 0.10058957412838936,
      "num_tokens": 1053554.0,
      "step": 570
    },
    {
      "entropy": 6.854119396209716,
      "epoch": 0.04830917874396135,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000287,
      "loss": 6.6366,
      "mean_token_accuracy": 0.10385002046823502,
      "num_tokens": 1062008.0,
      "step": 575
    },
    {
      "entropy": 6.868479824066162,
      "epoch": 0.048729258559126236,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002895,
      "loss": 6.7048,
      "mean_token_accuracy": 0.10346106439828873,
      "num_tokens": 1070740.0,
      "step": 580
    },
    {
      "entropy": 6.8440343856811525,
      "epoch": 0.04914933837429111,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000292,
      "loss": 6.7057,
      "mean_token_accuracy": 0.10240900367498398,
      "num_tokens": 1079681.0,
      "step": 585
    },
    {
      "entropy": 6.858892154693604,
      "epoch": 0.049569418189456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002945,
      "loss": 6.5847,
      "mean_token_accuracy": 0.10450911447405815,
      "num_tokens": 1088979.0,
      "step": 590
    },
    {
      "entropy": 6.772767686843872,
      "epoch": 0.04998949800462088,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000297,
      "loss": 6.5832,
      "mean_token_accuracy": 0.10501813441514969,
      "num_tokens": 1097870.0,
      "step": 595
    },
    {
      "entropy": 6.856569433212281,
      "epoch": 0.05040957781978576,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002995,
      "loss": 6.714,
      "mean_token_accuracy": 0.09948427230119705,
      "num_tokens": 1107948.0,
      "step": 600
    },
    {
      "entropy": 6.876928329467773,
      "epoch": 0.05082965763495064,
      "grad_norm": 1.21875,
      "learning_rate": 0.000302,
      "loss": 6.6226,
      "mean_token_accuracy": 0.1076712541282177,
      "num_tokens": 1117032.0,
      "step": 605
    },
    {
      "entropy": 6.769250106811524,
      "epoch": 0.051249737450115525,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003045,
      "loss": 6.5928,
      "mean_token_accuracy": 0.10671919211745262,
      "num_tokens": 1127834.0,
      "step": 610
    },
    {
      "entropy": 6.948690032958984,
      "epoch": 0.0516698172652804,
      "grad_norm": 1.40625,
      "learning_rate": 0.000307,
      "loss": 6.695,
      "mean_token_accuracy": 0.11499854996800422,
      "num_tokens": 1137382.0,
      "step": 615
    },
    {
      "entropy": 6.77532000541687,
      "epoch": 0.052089897080445285,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003095,
      "loss": 6.5422,
      "mean_token_accuracy": 0.11223937124013901,
      "num_tokens": 1146095.0,
      "step": 620
    },
    {
      "entropy": 6.723394155502319,
      "epoch": 0.05250997689561017,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000312,
      "loss": 6.5822,
      "mean_token_accuracy": 0.10564726367592811,
      "num_tokens": 1154981.0,
      "step": 625
    },
    {
      "entropy": 6.776411151885986,
      "epoch": 0.052930056710775046,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0003145,
      "loss": 6.6153,
      "mean_token_accuracy": 0.10753953084349632,
      "num_tokens": 1164939.0,
      "step": 630
    },
    {
      "entropy": 6.936794233322144,
      "epoch": 0.05335013652593993,
      "grad_norm": 1.203125,
      "learning_rate": 0.000317,
      "loss": 6.7291,
      "mean_token_accuracy": 0.09790047407150268,
      "num_tokens": 1174991.0,
      "step": 635
    },
    {
      "entropy": 6.818718576431275,
      "epoch": 0.05377021634110481,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003195,
      "loss": 6.78,
      "mean_token_accuracy": 0.09581352695822716,
      "num_tokens": 1184885.0,
      "step": 640
    },
    {
      "entropy": 6.859689378738404,
      "epoch": 0.05419029615626969,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000322,
      "loss": 6.6652,
      "mean_token_accuracy": 0.10177846625447273,
      "num_tokens": 1193637.0,
      "step": 645
    },
    {
      "entropy": 6.70958137512207,
      "epoch": 0.054610375971434574,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4505,
      "mean_token_accuracy": 0.11398516818881035,
      "num_tokens": 1202188.0,
      "step": 650
    },
    {
      "entropy": 6.731061363220215,
      "epoch": 0.05503045578659945,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.5923,
      "mean_token_accuracy": 0.10111142173409463,
      "num_tokens": 1210768.0,
      "step": 655
    },
    {
      "entropy": 6.75755124092102,
      "epoch": 0.055450535601764335,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5885,
      "mean_token_accuracy": 0.10299575850367546,
      "num_tokens": 1219819.0,
      "step": 660
    },
    {
      "entropy": 6.8775472164154055,
      "epoch": 0.05587061541692922,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.6507,
      "mean_token_accuracy": 0.09766614213585853,
      "num_tokens": 1229703.0,
      "step": 665
    },
    {
      "entropy": 6.829215049743652,
      "epoch": 0.056290695232094096,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.6863,
      "mean_token_accuracy": 0.09930930510163308,
      "num_tokens": 1238942.0,
      "step": 670
    },
    {
      "entropy": 6.886805677413941,
      "epoch": 0.05671077504725898,
      "grad_norm": 1.1875,
      "learning_rate": 0.000337,
      "loss": 6.7475,
      "mean_token_accuracy": 0.09512239620089531,
      "num_tokens": 1248943.0,
      "step": 675
    },
    {
      "entropy": 6.774325275421143,
      "epoch": 0.05713085486242386,
      "grad_norm": 1.125,
      "learning_rate": 0.0003395,
      "loss": 6.6164,
      "mean_token_accuracy": 0.10321223735809326,
      "num_tokens": 1257761.0,
      "step": 680
    },
    {
      "entropy": 6.6621216297149655,
      "epoch": 0.05755093467758874,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000342,
      "loss": 6.5622,
      "mean_token_accuracy": 0.10228212624788284,
      "num_tokens": 1267216.0,
      "step": 685
    },
    {
      "entropy": 6.826507520675659,
      "epoch": 0.057971014492753624,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6587,
      "mean_token_accuracy": 0.1079720102250576,
      "num_tokens": 1277210.0,
      "step": 690
    },
    {
      "entropy": 6.737741279602051,
      "epoch": 0.05839109430791851,
      "grad_norm": 1.21875,
      "learning_rate": 0.000347,
      "loss": 6.5588,
      "mean_token_accuracy": 0.10001136437058449,
      "num_tokens": 1285310.0,
      "step": 695
    },
    {
      "entropy": 6.800521755218506,
      "epoch": 0.058811174123083385,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003495,
      "loss": 6.5887,
      "mean_token_accuracy": 0.10580191239714623,
      "num_tokens": 1294421.0,
      "step": 700
    },
    {
      "entropy": 6.603023052215576,
      "epoch": 0.05923125393824827,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000352,
      "loss": 6.4007,
      "mean_token_accuracy": 0.11451570391654968,
      "num_tokens": 1303281.0,
      "step": 705
    },
    {
      "entropy": 6.694077110290527,
      "epoch": 0.059651333753413145,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003545,
      "loss": 6.5884,
      "mean_token_accuracy": 0.1103569135069847,
      "num_tokens": 1312280.0,
      "step": 710
    },
    {
      "entropy": 6.703026485443115,
      "epoch": 0.06007141356857803,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000357,
      "loss": 6.5455,
      "mean_token_accuracy": 0.10655389800667762,
      "num_tokens": 1321243.0,
      "step": 715
    },
    {
      "entropy": 6.783720779418945,
      "epoch": 0.06049149338374291,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003595,
      "loss": 6.6752,
      "mean_token_accuracy": 0.10890973284840584,
      "num_tokens": 1330324.0,
      "step": 720
    },
    {
      "entropy": 6.716011047363281,
      "epoch": 0.06091157319890779,
      "grad_norm": 1.25,
      "learning_rate": 0.000362,
      "loss": 6.4895,
      "mean_token_accuracy": 0.11130202338099479,
      "num_tokens": 1339485.0,
      "step": 725
    },
    {
      "entropy": 6.763300609588623,
      "epoch": 0.06133165301407267,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003645,
      "loss": 6.6603,
      "mean_token_accuracy": 0.09835303947329521,
      "num_tokens": 1348640.0,
      "step": 730
    },
    {
      "entropy": 6.686880588531494,
      "epoch": 0.06175173282923756,
      "grad_norm": 1.25,
      "learning_rate": 0.000367,
      "loss": 6.5352,
      "mean_token_accuracy": 0.10910931676626205,
      "num_tokens": 1357581.0,
      "step": 735
    },
    {
      "entropy": 6.789766788482666,
      "epoch": 0.062171812644402434,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003695,
      "loss": 6.6092,
      "mean_token_accuracy": 0.10686837136745453,
      "num_tokens": 1367883.0,
      "step": 740
    },
    {
      "entropy": 6.689715576171875,
      "epoch": 0.06259189245956731,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000372,
      "loss": 6.5732,
      "mean_token_accuracy": 0.10044000372290611,
      "num_tokens": 1376936.0,
      "step": 745
    },
    {
      "entropy": 6.619902896881103,
      "epoch": 0.0630119722747322,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003745,
      "loss": 6.5013,
      "mean_token_accuracy": 0.10585071742534638,
      "num_tokens": 1386359.0,
      "step": 750
    },
    {
      "entropy": 6.691353893280029,
      "epoch": 0.06343205208989708,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000377,
      "loss": 6.5614,
      "mean_token_accuracy": 0.10925468727946282,
      "num_tokens": 1395223.0,
      "step": 755
    },
    {
      "entropy": 6.756776332855225,
      "epoch": 0.06385213190506196,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003795,
      "loss": 6.7048,
      "mean_token_accuracy": 0.10100763738155365,
      "num_tokens": 1404917.0,
      "step": 760
    },
    {
      "entropy": 6.7892101287841795,
      "epoch": 0.06427221172022685,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000382,
      "loss": 6.6077,
      "mean_token_accuracy": 0.11203819289803504,
      "num_tokens": 1413348.0,
      "step": 765
    },
    {
      "entropy": 6.617217540740967,
      "epoch": 0.06469229153539173,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003845,
      "loss": 6.5804,
      "mean_token_accuracy": 0.10595368966460228,
      "num_tokens": 1421726.0,
      "step": 770
    },
    {
      "entropy": 6.699965381622315,
      "epoch": 0.0651123713505566,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5984,
      "mean_token_accuracy": 0.10766990706324578,
      "num_tokens": 1430686.0,
      "step": 775
    },
    {
      "entropy": 6.773920488357544,
      "epoch": 0.06553245116572148,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5765,
      "mean_token_accuracy": 0.10770290642976761,
      "num_tokens": 1439499.0,
      "step": 780
    },
    {
      "entropy": 6.685867691040039,
      "epoch": 0.06595253098088637,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.5731,
      "mean_token_accuracy": 0.10584950372576714,
      "num_tokens": 1448220.0,
      "step": 785
    },
    {
      "entropy": 6.635032224655151,
      "epoch": 0.06637261079605125,
      "grad_norm": 1.0,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.5914,
      "mean_token_accuracy": 0.09675629287958146,
      "num_tokens": 1458217.0,
      "step": 790
    },
    {
      "entropy": 6.699159860610962,
      "epoch": 0.06679269061121614,
      "grad_norm": 1.125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4848,
      "mean_token_accuracy": 0.10567129477858543,
      "num_tokens": 1467422.0,
      "step": 795
    },
    {
      "entropy": 6.620410299301147,
      "epoch": 0.06721277042638102,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003995,
      "loss": 6.4767,
      "mean_token_accuracy": 0.11094664260745049,
      "num_tokens": 1476152.0,
      "step": 800
    },
    {
      "entropy": 6.650699758529663,
      "epoch": 0.06763285024154589,
      "grad_norm": 1.265625,
      "learning_rate": 0.000402,
      "loss": 6.5738,
      "mean_token_accuracy": 0.1043787069618702,
      "num_tokens": 1485248.0,
      "step": 805
    },
    {
      "entropy": 6.619400262832642,
      "epoch": 0.06805293005671077,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004045,
      "loss": 6.5511,
      "mean_token_accuracy": 0.10442669913172722,
      "num_tokens": 1494248.0,
      "step": 810
    },
    {
      "entropy": 6.722617673873901,
      "epoch": 0.06847300987187566,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.6573,
      "mean_token_accuracy": 0.10585515722632408,
      "num_tokens": 1503565.0,
      "step": 815
    },
    {
      "entropy": 6.83908371925354,
      "epoch": 0.06889308968704054,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004095,
      "loss": 6.745,
      "mean_token_accuracy": 0.10003346055746079,
      "num_tokens": 1513227.0,
      "step": 820
    },
    {
      "entropy": 6.658945035934448,
      "epoch": 0.06931316950220542,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000412,
      "loss": 6.5346,
      "mean_token_accuracy": 0.10508675500750542,
      "num_tokens": 1522312.0,
      "step": 825
    },
    {
      "entropy": 6.637969160079956,
      "epoch": 0.0697332493173703,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004145,
      "loss": 6.4802,
      "mean_token_accuracy": 0.10670675709843636,
      "num_tokens": 1531720.0,
      "step": 830
    },
    {
      "entropy": 6.6340169429779055,
      "epoch": 0.07015332913253518,
      "grad_norm": 1.140625,
      "learning_rate": 0.000417,
      "loss": 6.5721,
      "mean_token_accuracy": 0.10074454993009567,
      "num_tokens": 1541238.0,
      "step": 835
    },
    {
      "entropy": 6.695564794540405,
      "epoch": 0.07057340894770006,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004195,
      "loss": 6.6648,
      "mean_token_accuracy": 0.10375690832734108,
      "num_tokens": 1550875.0,
      "step": 840
    },
    {
      "entropy": 6.645870971679687,
      "epoch": 0.07099348876286495,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000422,
      "loss": 6.6076,
      "mean_token_accuracy": 0.10648187175393105,
      "num_tokens": 1560287.0,
      "step": 845
    },
    {
      "entropy": 6.6967510223388675,
      "epoch": 0.07141356857802983,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004245,
      "loss": 6.4978,
      "mean_token_accuracy": 0.11105224043130875,
      "num_tokens": 1569043.0,
      "step": 850
    },
    {
      "entropy": 6.554346418380737,
      "epoch": 0.07183364839319471,
      "grad_norm": 1.140625,
      "learning_rate": 0.000427,
      "loss": 6.5154,
      "mean_token_accuracy": 0.11203170269727707,
      "num_tokens": 1578112.0,
      "step": 855
    },
    {
      "entropy": 6.515066003799438,
      "epoch": 0.07225372820835958,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004295,
      "loss": 6.4364,
      "mean_token_accuracy": 0.11132391095161438,
      "num_tokens": 1586587.0,
      "step": 860
    },
    {
      "entropy": 6.742719125747681,
      "epoch": 0.07267380802352447,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000432,
      "loss": 6.5978,
      "mean_token_accuracy": 0.10682642236351966,
      "num_tokens": 1595585.0,
      "step": 865
    },
    {
      "entropy": 6.641049814224243,
      "epoch": 0.07309388783868935,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004345,
      "loss": 6.551,
      "mean_token_accuracy": 0.10661023184657097,
      "num_tokens": 1605355.0,
      "step": 870
    },
    {
      "entropy": 6.638308906555176,
      "epoch": 0.07351396765385423,
      "grad_norm": 1.234375,
      "learning_rate": 0.000437,
      "loss": 6.5889,
      "mean_token_accuracy": 0.10184741988778115,
      "num_tokens": 1613637.0,
      "step": 875
    },
    {
      "entropy": 6.658770608901977,
      "epoch": 0.07393404746901912,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004395,
      "loss": 6.5351,
      "mean_token_accuracy": 0.10806669145822526,
      "num_tokens": 1622731.0,
      "step": 880
    },
    {
      "entropy": 6.610404300689697,
      "epoch": 0.074354127284184,
      "grad_norm": 1.140625,
      "learning_rate": 0.000442,
      "loss": 6.5083,
      "mean_token_accuracy": 0.10660439133644103,
      "num_tokens": 1632098.0,
      "step": 885
    },
    {
      "entropy": 6.581112480163574,
      "epoch": 0.07477420709934887,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004445,
      "loss": 6.5117,
      "mean_token_accuracy": 0.10190015733242035,
      "num_tokens": 1641259.0,
      "step": 890
    },
    {
      "entropy": 6.669602966308593,
      "epoch": 0.07519428691451376,
      "grad_norm": 1.1875,
      "learning_rate": 0.000447,
      "loss": 6.5769,
      "mean_token_accuracy": 0.103199552744627,
      "num_tokens": 1651362.0,
      "step": 895
    },
    {
      "entropy": 6.581767272949219,
      "epoch": 0.07561436672967864,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.4893,
      "mean_token_accuracy": 0.1063641555607319,
      "num_tokens": 1660190.0,
      "step": 900
    },
    {
      "entropy": 6.573458099365235,
      "epoch": 0.07603444654484352,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5473,
      "mean_token_accuracy": 0.10238413438200951,
      "num_tokens": 1669020.0,
      "step": 905
    },
    {
      "entropy": 6.650429916381836,
      "epoch": 0.0764545263600084,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.5657,
      "mean_token_accuracy": 0.1067358523607254,
      "num_tokens": 1678158.0,
      "step": 910
    },
    {
      "entropy": 6.640725135803223,
      "epoch": 0.07687460617517328,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.5723,
      "mean_token_accuracy": 0.10445328801870346,
      "num_tokens": 1687481.0,
      "step": 915
    },
    {
      "entropy": 6.5850495338439945,
      "epoch": 0.07729468599033816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.5327,
      "mean_token_accuracy": 0.11253217458724976,
      "num_tokens": 1696782.0,
      "step": 920
    },
    {
      "entropy": 6.555831384658814,
      "epoch": 0.07771476580550304,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000462,
      "loss": 6.528,
      "mean_token_accuracy": 0.10824255496263505,
      "num_tokens": 1706153.0,
      "step": 925
    },
    {
      "entropy": 6.625135850906372,
      "epoch": 0.07813484562066793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004645,
      "loss": 6.559,
      "mean_token_accuracy": 0.1069357268512249,
      "num_tokens": 1715585.0,
      "step": 930
    },
    {
      "entropy": 6.768569469451904,
      "epoch": 0.07855492543583281,
      "grad_norm": 1.578125,
      "learning_rate": 0.000467,
      "loss": 6.6818,
      "mean_token_accuracy": 0.10121209248900413,
      "num_tokens": 1724857.0,
      "step": 935
    },
    {
      "entropy": 6.553330516815185,
      "epoch": 0.0789750052509977,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004695,
      "loss": 6.4953,
      "mean_token_accuracy": 0.11398640796542167,
      "num_tokens": 1733528.0,
      "step": 940
    },
    {
      "entropy": 6.64405460357666,
      "epoch": 0.07939508506616257,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000472,
      "loss": 6.616,
      "mean_token_accuracy": 0.10737027525901795,
      "num_tokens": 1742953.0,
      "step": 945
    },
    {
      "entropy": 6.612694597244262,
      "epoch": 0.07981516488132745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004745,
      "loss": 6.5605,
      "mean_token_accuracy": 0.11337294653058053,
      "num_tokens": 1752155.0,
      "step": 950
    },
    {
      "entropy": 6.579654312133789,
      "epoch": 0.08023524469649233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000477,
      "loss": 6.4539,
      "mean_token_accuracy": 0.10857293009757996,
      "num_tokens": 1760562.0,
      "step": 955
    },
    {
      "entropy": 6.568000841140747,
      "epoch": 0.08065532451165722,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004795,
      "loss": 6.4953,
      "mean_token_accuracy": 0.10662117302417755,
      "num_tokens": 1769631.0,
      "step": 960
    },
    {
      "entropy": 6.530600309371948,
      "epoch": 0.0810754043268221,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000482,
      "loss": 6.513,
      "mean_token_accuracy": 0.10268357619643212,
      "num_tokens": 1779080.0,
      "step": 965
    },
    {
      "entropy": 6.608699893951416,
      "epoch": 0.08149548414198697,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004845,
      "loss": 6.5062,
      "mean_token_accuracy": 0.1082501009106636,
      "num_tokens": 1787830.0,
      "step": 970
    },
    {
      "entropy": 6.3936583518981935,
      "epoch": 0.08191556395715185,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000487,
      "loss": 6.4399,
      "mean_token_accuracy": 0.10422300174832344,
      "num_tokens": 1796998.0,
      "step": 975
    },
    {
      "entropy": 6.6043681621551515,
      "epoch": 0.08233564377231674,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004895,
      "loss": 6.5165,
      "mean_token_accuracy": 0.1103883646428585,
      "num_tokens": 1806194.0,
      "step": 980
    },
    {
      "entropy": 6.372901821136475,
      "epoch": 0.08275572358748162,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000492,
      "loss": 6.4139,
      "mean_token_accuracy": 0.11206447035074234,
      "num_tokens": 1815751.0,
      "step": 985
    },
    {
      "entropy": 6.442894506454468,
      "epoch": 0.0831758034026465,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004945,
      "loss": 6.4907,
      "mean_token_accuracy": 0.11046408414840699,
      "num_tokens": 1825379.0,
      "step": 990
    },
    {
      "entropy": 6.55560712814331,
      "epoch": 0.08359588321781139,
      "grad_norm": 1.203125,
      "learning_rate": 0.000497,
      "loss": 6.454,
      "mean_token_accuracy": 0.10851948186755181,
      "num_tokens": 1834158.0,
      "step": 995
    },
    {
      "entropy": 6.516812181472778,
      "epoch": 0.08401596303297626,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995,
      "loss": 6.4132,
      "mean_token_accuracy": 0.10896480083465576,
      "num_tokens": 1842724.0,
      "step": 1000
    },
    {
      "entropy": 6.518254280090332,
      "epoch": 0.08443604284814114,
      "grad_norm": 1.125,
      "learning_rate": 0.000499999998724557,
      "loss": 6.4359,
      "mean_token_accuracy": 0.11062911972403526,
      "num_tokens": 1852485.0,
      "step": 1005
    },
    {
      "entropy": 6.472753667831421,
      "epoch": 0.08485612266330603,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999999935430703,
      "loss": 6.4668,
      "mean_token_accuracy": 0.11211320757865906,
      "num_tokens": 1861303.0,
      "step": 1010
    },
    {
      "entropy": 6.340228652954101,
      "epoch": 0.08527620247847091,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999999843758243,
      "loss": 6.4328,
      "mean_token_accuracy": 0.11544003784656524,
      "num_tokens": 1870859.0,
      "step": 1015
    },
    {
      "entropy": 6.671287488937378,
      "epoch": 0.0856962822936358,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999999712228196,
      "loss": 6.6205,
      "mean_token_accuracy": 0.0996169812977314,
      "num_tokens": 1880295.0,
      "step": 1020
    },
    {
      "entropy": 6.63971586227417,
      "epoch": 0.08611636210880068,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999999540840562,
      "loss": 6.5111,
      "mean_token_accuracy": 0.11260380744934081,
      "num_tokens": 1889193.0,
      "step": 1025
    },
    {
      "entropy": 6.493311834335327,
      "epoch": 0.08653644192396555,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999999329595345,
      "loss": 6.6021,
      "mean_token_accuracy": 0.09916243478655815,
      "num_tokens": 1899437.0,
      "step": 1030
    },
    {
      "entropy": 6.5729930877685545,
      "epoch": 0.08695652173913043,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999078492548,
      "loss": 6.5058,
      "mean_token_accuracy": 0.10446131974458694,
      "num_tokens": 1907882.0,
      "step": 1035
    },
    {
      "entropy": 6.492805910110474,
      "epoch": 0.08737660155429532,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999998787532176,
      "loss": 6.4138,
      "mean_token_accuracy": 0.1126400038599968,
      "num_tokens": 1916872.0,
      "step": 1040
    },
    {
      "entropy": 6.506380701065064,
      "epoch": 0.0877966813694602,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999998456714234,
      "loss": 6.5924,
      "mean_token_accuracy": 0.11272363439202308,
      "num_tokens": 1926636.0,
      "step": 1045
    },
    {
      "entropy": 6.5269097805023195,
      "epoch": 0.08821676118462508,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999998086038729,
      "loss": 6.4905,
      "mean_token_accuracy": 0.10973675772547722,
      "num_tokens": 1935962.0,
      "step": 1050
    },
    {
      "entropy": 6.491125774383545,
      "epoch": 0.08863684099978995,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999997675505665,
      "loss": 6.4506,
      "mean_token_accuracy": 0.11091897338628769,
      "num_tokens": 1944600.0,
      "step": 1055
    },
    {
      "entropy": 6.560743236541748,
      "epoch": 0.08905692081495484,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999997225115052,
      "loss": 6.6403,
      "mean_token_accuracy": 0.10566280484199524,
      "num_tokens": 1954234.0,
      "step": 1060
    },
    {
      "entropy": 6.71457724571228,
      "epoch": 0.08947700063011972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999996734866896,
      "loss": 6.5989,
      "mean_token_accuracy": 0.10413395762443542,
      "num_tokens": 1964499.0,
      "step": 1065
    },
    {
      "entropy": 6.307662582397461,
      "epoch": 0.0898970804452846,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999996204761206,
      "loss": 6.3004,
      "mean_token_accuracy": 0.11687865033745766,
      "num_tokens": 1973635.0,
      "step": 1070
    },
    {
      "entropy": 6.497924661636352,
      "epoch": 0.09031716026044949,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999995634797993,
      "loss": 6.4546,
      "mean_token_accuracy": 0.11474235132336616,
      "num_tokens": 1983509.0,
      "step": 1075
    },
    {
      "entropy": 6.480887317657471,
      "epoch": 0.09073724007561437,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999995024977265,
      "loss": 6.4317,
      "mean_token_accuracy": 0.11495376154780387,
      "num_tokens": 1992336.0,
      "step": 1080
    },
    {
      "entropy": 6.518788290023804,
      "epoch": 0.09115731989077924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999994375299034,
      "loss": 6.467,
      "mean_token_accuracy": 0.11141329482197762,
      "num_tokens": 2001931.0,
      "step": 1085
    },
    {
      "entropy": 6.396731853485107,
      "epoch": 0.09157739970594413,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499999368576331,
      "loss": 6.3268,
      "mean_token_accuracy": 0.11944503113627433,
      "num_tokens": 2010935.0,
      "step": 1090
    },
    {
      "entropy": 6.410158395767212,
      "epoch": 0.09199747952110901,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999992956370109,
      "loss": 6.3911,
      "mean_token_accuracy": 0.1134261205792427,
      "num_tokens": 2020587.0,
      "step": 1095
    },
    {
      "entropy": 6.34518141746521,
      "epoch": 0.0924175593362739,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499999218711944,
      "loss": 6.4155,
      "mean_token_accuracy": 0.11149835661053657,
      "num_tokens": 2029743.0,
      "step": 1100
    },
    {
      "entropy": 6.548259019851685,
      "epoch": 0.09283763915143878,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999991378011317,
      "loss": 6.4423,
      "mean_token_accuracy": 0.11465151533484459,
      "num_tokens": 2038468.0,
      "step": 1105
    },
    {
      "entropy": 6.381483364105224,
      "epoch": 0.09325771896660366,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999990529045757,
      "loss": 6.3588,
      "mean_token_accuracy": 0.11419494673609734,
      "num_tokens": 2047456.0,
      "step": 1110
    },
    {
      "entropy": 6.610185289382935,
      "epoch": 0.09367779878176853,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999989640222771,
      "loss": 6.6665,
      "mean_token_accuracy": 0.09994395300745965,
      "num_tokens": 2056691.0,
      "step": 1115
    },
    {
      "entropy": 6.5219189643859865,
      "epoch": 0.09409787859693342,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499998871154238,
      "loss": 6.4644,
      "mean_token_accuracy": 0.10945621505379677,
      "num_tokens": 2066068.0,
      "step": 1120
    },
    {
      "entropy": 6.522899913787842,
      "epoch": 0.0945179584120983,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999987743004597,
      "loss": 6.3814,
      "mean_token_accuracy": 0.11605900377035142,
      "num_tokens": 2075113.0,
      "step": 1125
    },
    {
      "entropy": 6.463716268539429,
      "epoch": 0.09493803822726318,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999986734609438,
      "loss": 6.5424,
      "mean_token_accuracy": 0.11121912077069282,
      "num_tokens": 2084557.0,
      "step": 1130
    },
    {
      "entropy": 6.5195088386535645,
      "epoch": 0.09535811804242807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999985686356923,
      "loss": 6.4293,
      "mean_token_accuracy": 0.10896992832422256,
      "num_tokens": 2093424.0,
      "step": 1135
    },
    {
      "entropy": 6.501124429702759,
      "epoch": 0.09577819785759294,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499998459824707,
      "loss": 6.541,
      "mean_token_accuracy": 0.10720071643590927,
      "num_tokens": 2103066.0,
      "step": 1140
    },
    {
      "entropy": 6.485859060287476,
      "epoch": 0.09619827767275782,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00049999834702799,
      "loss": 6.4218,
      "mean_token_accuracy": 0.11379043385386467,
      "num_tokens": 2112447.0,
      "step": 1145
    },
    {
      "entropy": 6.421380949020386,
      "epoch": 0.0966183574879227,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999982302455431,
      "loss": 6.4471,
      "mean_token_accuracy": 0.11498942598700523,
      "num_tokens": 2121949.0,
      "step": 1150
    },
    {
      "entropy": 6.473872327804566,
      "epoch": 0.09703843730308759,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999981094773683,
      "loss": 6.3538,
      "mean_token_accuracy": 0.11318295449018478,
      "num_tokens": 2130464.0,
      "step": 1155
    },
    {
      "entropy": 6.440775918960571,
      "epoch": 0.09745851711825247,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499997984723468,
      "loss": 6.516,
      "mean_token_accuracy": 0.10744207352399826,
      "num_tokens": 2139577.0,
      "step": 1160
    },
    {
      "entropy": 6.234747648239136,
      "epoch": 0.09787859693341736,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999978559838441,
      "loss": 6.2367,
      "mean_token_accuracy": 0.11980548799037934,
      "num_tokens": 2147919.0,
      "step": 1165
    },
    {
      "entropy": 6.425514888763428,
      "epoch": 0.09829867674858223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999977232584991,
      "loss": 6.411,
      "mean_token_accuracy": 0.11489588171243667,
      "num_tokens": 2156936.0,
      "step": 1170
    },
    {
      "entropy": 6.470327949523925,
      "epoch": 0.09871875656374711,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999975865474354,
      "loss": 6.474,
      "mean_token_accuracy": 0.11181759610772132,
      "num_tokens": 2165362.0,
      "step": 1175
    },
    {
      "entropy": 6.40549669265747,
      "epoch": 0.099138836378912,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999974458506551,
      "loss": 6.3913,
      "mean_token_accuracy": 0.11321083605289459,
      "num_tokens": 2173665.0,
      "step": 1180
    },
    {
      "entropy": 6.484990215301513,
      "epoch": 0.09955891619407688,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499997301168161,
      "loss": 6.3547,
      "mean_token_accuracy": 0.1165615513920784,
      "num_tokens": 2182222.0,
      "step": 1185
    },
    {
      "entropy": 6.47981333732605,
      "epoch": 0.09997899600924176,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999971524999556,
      "loss": 6.4586,
      "mean_token_accuracy": 0.11381722316145897,
      "num_tokens": 2192358.0,
      "step": 1190
    },
    {
      "entropy": 6.44229564666748,
      "epoch": 0.10039907582440663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999969998460414,
      "loss": 6.4519,
      "mean_token_accuracy": 0.11379328668117523,
      "num_tokens": 2201889.0,
      "step": 1195
    },
    {
      "entropy": 6.39982123374939,
      "epoch": 0.10081915563957151,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004999968432064213,
      "loss": 6.4571,
      "mean_token_accuracy": 0.11692695915699006,
      "num_tokens": 2211810.0,
      "step": 1200
    },
    {
      "entropy": 6.344041538238526,
      "epoch": 0.1012392354547364,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999966825810979,
      "loss": 6.401,
      "mean_token_accuracy": 0.11303748413920403,
      "num_tokens": 2221123.0,
      "step": 1205
    },
    {
      "entropy": 6.35064206123352,
      "epoch": 0.10165931526990128,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999965179700742,
      "loss": 6.3287,
      "mean_token_accuracy": 0.11802673861384391,
      "num_tokens": 2230129.0,
      "step": 1210
    },
    {
      "entropy": 6.3616985321044925,
      "epoch": 0.10207939508506617,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499996349373353,
      "loss": 6.3828,
      "mean_token_accuracy": 0.11236807405948639,
      "num_tokens": 2239929.0,
      "step": 1215
    },
    {
      "entropy": 6.4787780284881595,
      "epoch": 0.10249947490023105,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999961767909374,
      "loss": 6.3565,
      "mean_token_accuracy": 0.11246357783675194,
      "num_tokens": 2248078.0,
      "step": 1220
    },
    {
      "entropy": 6.329130172729492,
      "epoch": 0.10291955471539592,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999960002228303,
      "loss": 6.4464,
      "mean_token_accuracy": 0.11457708552479744,
      "num_tokens": 2256975.0,
      "step": 1225
    },
    {
      "entropy": 6.41359543800354,
      "epoch": 0.1033396345305608,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999958196690349,
      "loss": 6.3075,
      "mean_token_accuracy": 0.11755945533514023,
      "num_tokens": 2265797.0,
      "step": 1230
    },
    {
      "entropy": 6.318053436279297,
      "epoch": 0.10375971434572569,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999956351295545,
      "loss": 6.407,
      "mean_token_accuracy": 0.12029065862298012,
      "num_tokens": 2274099.0,
      "step": 1235
    },
    {
      "entropy": 6.369567108154297,
      "epoch": 0.10417979416089057,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999954466043922,
      "loss": 6.3363,
      "mean_token_accuracy": 0.11501603052020073,
      "num_tokens": 2282360.0,
      "step": 1240
    },
    {
      "entropy": 6.357530832290649,
      "epoch": 0.10459987397605545,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999952540935514,
      "loss": 6.427,
      "mean_token_accuracy": 0.10687654167413711,
      "num_tokens": 2292714.0,
      "step": 1245
    },
    {
      "entropy": 6.4097044467926025,
      "epoch": 0.10501995379122034,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999950575970356,
      "loss": 6.3634,
      "mean_token_accuracy": 0.11621319502592087,
      "num_tokens": 2301633.0,
      "step": 1250
    },
    {
      "entropy": 6.369806289672852,
      "epoch": 0.10544003360638521,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999948571148482,
      "loss": 6.3449,
      "mean_token_accuracy": 0.12026969790458679,
      "num_tokens": 2310067.0,
      "step": 1255
    },
    {
      "entropy": 6.404969978332519,
      "epoch": 0.10586011342155009,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999946526469927,
      "loss": 6.4302,
      "mean_token_accuracy": 0.11556015834212303,
      "num_tokens": 2320090.0,
      "step": 1260
    },
    {
      "entropy": 6.372836637496948,
      "epoch": 0.10628019323671498,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999944441934728,
      "loss": 6.3703,
      "mean_token_accuracy": 0.1213706873357296,
      "num_tokens": 2329255.0,
      "step": 1265
    },
    {
      "entropy": 6.461856746673584,
      "epoch": 0.10670027305187986,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999942317542922,
      "loss": 6.4648,
      "mean_token_accuracy": 0.11396320164203644,
      "num_tokens": 2339535.0,
      "step": 1270
    },
    {
      "entropy": 6.327042865753174,
      "epoch": 0.10712035286704474,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999940153294546,
      "loss": 6.3664,
      "mean_token_accuracy": 0.11329737976193428,
      "num_tokens": 2348948.0,
      "step": 1275
    },
    {
      "entropy": 6.429010534286499,
      "epoch": 0.10754043268220961,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499993794918964,
      "loss": 6.3999,
      "mean_token_accuracy": 0.11255929544568062,
      "num_tokens": 2359141.0,
      "step": 1280
    },
    {
      "entropy": 6.310593366622925,
      "epoch": 0.1079605124973745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999935705228241,
      "loss": 6.4541,
      "mean_token_accuracy": 0.11158784702420235,
      "num_tokens": 2368906.0,
      "step": 1285
    },
    {
      "entropy": 6.527987766265869,
      "epoch": 0.10838059231253938,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999933421410389,
      "loss": 6.4033,
      "mean_token_accuracy": 0.11808300763368607,
      "num_tokens": 2377029.0,
      "step": 1290
    },
    {
      "entropy": 6.42291259765625,
      "epoch": 0.10880067212770426,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999931097736125,
      "loss": 6.4774,
      "mean_token_accuracy": 0.11096713319420815,
      "num_tokens": 2387088.0,
      "step": 1295
    },
    {
      "entropy": 6.416878700256348,
      "epoch": 0.10922075194286915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999928734205492,
      "loss": 6.3793,
      "mean_token_accuracy": 0.11391115635633468,
      "num_tokens": 2395596.0,
      "step": 1300
    },
    {
      "entropy": 6.3132000923156735,
      "epoch": 0.10964083175803403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999926330818528,
      "loss": 6.3632,
      "mean_token_accuracy": 0.12021223455667496,
      "num_tokens": 2404506.0,
      "step": 1305
    },
    {
      "entropy": 6.388805866241455,
      "epoch": 0.1100609115731989,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999923887575278,
      "loss": 6.4068,
      "mean_token_accuracy": 0.11463478580117226,
      "num_tokens": 2414342.0,
      "step": 1310
    },
    {
      "entropy": 6.404540014266968,
      "epoch": 0.11048099138836379,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999921404475785,
      "loss": 6.3616,
      "mean_token_accuracy": 0.11288373470306397,
      "num_tokens": 2423076.0,
      "step": 1315
    },
    {
      "entropy": 6.345643043518066,
      "epoch": 0.11090107120352867,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999918881520093,
      "loss": 6.3217,
      "mean_token_accuracy": 0.12194343283772469,
      "num_tokens": 2432492.0,
      "step": 1320
    },
    {
      "entropy": 6.29752926826477,
      "epoch": 0.11132115101869355,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999916318708246,
      "loss": 6.2719,
      "mean_token_accuracy": 0.12224209308624268,
      "num_tokens": 2441916.0,
      "step": 1325
    },
    {
      "entropy": 6.3799408912658695,
      "epoch": 0.11174123083385844,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999913716040291,
      "loss": 6.3145,
      "mean_token_accuracy": 0.12352623343467713,
      "num_tokens": 2450932.0,
      "step": 1330
    },
    {
      "entropy": 6.248048543930054,
      "epoch": 0.11216131064902331,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999911073516272,
      "loss": 6.354,
      "mean_token_accuracy": 0.11721153929829597,
      "num_tokens": 2460058.0,
      "step": 1335
    },
    {
      "entropy": 6.305795478820801,
      "epoch": 0.11258139046418819,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999908391136237,
      "loss": 6.3027,
      "mean_token_accuracy": 0.11657693609595299,
      "num_tokens": 2469607.0,
      "step": 1340
    },
    {
      "entropy": 6.403741073608399,
      "epoch": 0.11300147027935308,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999905668900234,
      "loss": 6.3338,
      "mean_token_accuracy": 0.11997214257717133,
      "num_tokens": 2478345.0,
      "step": 1345
    },
    {
      "entropy": 6.354722595214843,
      "epoch": 0.11342155009451796,
      "grad_norm": 1.234375,
      "learning_rate": 0.000499990290680831,
      "loss": 6.2712,
      "mean_token_accuracy": 0.12181988656520844,
      "num_tokens": 2486662.0,
      "step": 1350
    },
    {
      "entropy": 6.314315986633301,
      "epoch": 0.11384162990968284,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999900104860516,
      "loss": 6.4018,
      "mean_token_accuracy": 0.1190257914364338,
      "num_tokens": 2495392.0,
      "step": 1355
    },
    {
      "entropy": 6.396038579940796,
      "epoch": 0.11426170972484773,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999897263056898,
      "loss": 6.4374,
      "mean_token_accuracy": 0.11400101035833358,
      "num_tokens": 2505254.0,
      "step": 1360
    },
    {
      "entropy": 6.415481328964233,
      "epoch": 0.1146817895400126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499989438139751,
      "loss": 6.242,
      "mean_token_accuracy": 0.12410487607121468,
      "num_tokens": 2514096.0,
      "step": 1365
    },
    {
      "entropy": 6.27006311416626,
      "epoch": 0.11510186935517748,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999891459882401,
      "loss": 6.2694,
      "mean_token_accuracy": 0.11747925281524658,
      "num_tokens": 2523635.0,
      "step": 1370
    },
    {
      "entropy": 6.264933919906616,
      "epoch": 0.11552194917034236,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999888498511624,
      "loss": 6.3385,
      "mean_token_accuracy": 0.11789564937353134,
      "num_tokens": 2532528.0,
      "step": 1375
    },
    {
      "entropy": 6.311061573028565,
      "epoch": 0.11594202898550725,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999885497285229,
      "loss": 6.2494,
      "mean_token_accuracy": 0.1151455245912075,
      "num_tokens": 2541893.0,
      "step": 1380
    },
    {
      "entropy": 6.299208688735962,
      "epoch": 0.11636210880067213,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999882456203273,
      "loss": 6.3013,
      "mean_token_accuracy": 0.12206007465720177,
      "num_tokens": 2551551.0,
      "step": 1385
    },
    {
      "entropy": 6.300058650970459,
      "epoch": 0.11678218861583702,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999879375265806,
      "loss": 6.2409,
      "mean_token_accuracy": 0.11834143102169037,
      "num_tokens": 2560183.0,
      "step": 1390
    },
    {
      "entropy": 6.282702207565308,
      "epoch": 0.11720226843100189,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999876254472886,
      "loss": 6.1445,
      "mean_token_accuracy": 0.12687626332044602,
      "num_tokens": 2568697.0,
      "step": 1395
    },
    {
      "entropy": 6.272518634796143,
      "epoch": 0.11762234824616677,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999873093824565,
      "loss": 6.3599,
      "mean_token_accuracy": 0.11727140471339226,
      "num_tokens": 2578151.0,
      "step": 1400
    },
    {
      "entropy": 6.458877515792847,
      "epoch": 0.11804242806133165,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999869893320902,
      "loss": 6.466,
      "mean_token_accuracy": 0.11793015375733376,
      "num_tokens": 2585901.0,
      "step": 1405
    },
    {
      "entropy": 6.321643972396851,
      "epoch": 0.11846250787649654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999866652961952,
      "loss": 6.3084,
      "mean_token_accuracy": 0.11437714174389839,
      "num_tokens": 2595655.0,
      "step": 1410
    },
    {
      "entropy": 6.353274202346801,
      "epoch": 0.11888258769166142,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999863372747773,
      "loss": 6.2695,
      "mean_token_accuracy": 0.11410242393612861,
      "num_tokens": 2604949.0,
      "step": 1415
    },
    {
      "entropy": 6.335246944427491,
      "epoch": 0.11930266750682629,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999860052678423,
      "loss": 6.3408,
      "mean_token_accuracy": 0.11986073330044747,
      "num_tokens": 2614260.0,
      "step": 1420
    },
    {
      "entropy": 6.268476724624634,
      "epoch": 0.11972274732199117,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999856692753959,
      "loss": 6.3378,
      "mean_token_accuracy": 0.11452461034059525,
      "num_tokens": 2623740.0,
      "step": 1425
    },
    {
      "entropy": 6.361990165710449,
      "epoch": 0.12014282713715606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999853292974444,
      "loss": 6.2447,
      "mean_token_accuracy": 0.11876029148697853,
      "num_tokens": 2631998.0,
      "step": 1430
    },
    {
      "entropy": 6.266280698776245,
      "epoch": 0.12056290695232094,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999849853339936,
      "loss": 6.3713,
      "mean_token_accuracy": 0.12363618090748787,
      "num_tokens": 2641169.0,
      "step": 1435
    },
    {
      "entropy": 6.410887956619263,
      "epoch": 0.12098298676748583,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999846373850497,
      "loss": 6.2122,
      "mean_token_accuracy": 0.12586963474750518,
      "num_tokens": 2650576.0,
      "step": 1440
    },
    {
      "entropy": 6.177606773376465,
      "epoch": 0.12140306658265071,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999842854506186,
      "loss": 6.3172,
      "mean_token_accuracy": 0.11797089278697967,
      "num_tokens": 2660817.0,
      "step": 1445
    },
    {
      "entropy": 6.282220935821533,
      "epoch": 0.12182314639781558,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999839295307069,
      "loss": 6.2561,
      "mean_token_accuracy": 0.1204748846590519,
      "num_tokens": 2669338.0,
      "step": 1450
    },
    {
      "entropy": 6.330542469024659,
      "epoch": 0.12224322621298046,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999835696253206,
      "loss": 6.3117,
      "mean_token_accuracy": 0.11790118813514709,
      "num_tokens": 2679108.0,
      "step": 1455
    },
    {
      "entropy": 6.36133828163147,
      "epoch": 0.12266330602814535,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999832057344664,
      "loss": 6.2739,
      "mean_token_accuracy": 0.11782214790582657,
      "num_tokens": 2688126.0,
      "step": 1460
    },
    {
      "entropy": 6.110802221298218,
      "epoch": 0.12308338584331023,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999828378581504,
      "loss": 6.2493,
      "mean_token_accuracy": 0.1267126329243183,
      "num_tokens": 2697245.0,
      "step": 1465
    },
    {
      "entropy": 6.332847547531128,
      "epoch": 0.12350346565847511,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999824659963793,
      "loss": 6.3012,
      "mean_token_accuracy": 0.12391207665205002,
      "num_tokens": 2705934.0,
      "step": 1470
    },
    {
      "entropy": 6.2204491138458256,
      "epoch": 0.12392354547364,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999820901491598,
      "loss": 6.2299,
      "mean_token_accuracy": 0.12465188652276993,
      "num_tokens": 2714367.0,
      "step": 1475
    },
    {
      "entropy": 6.265383291244507,
      "epoch": 0.12434362528880487,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999817103164983,
      "loss": 6.2882,
      "mean_token_accuracy": 0.12172888666391372,
      "num_tokens": 2724366.0,
      "step": 1480
    },
    {
      "entropy": 6.282348680496216,
      "epoch": 0.12476370510396975,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999813264984017,
      "loss": 6.284,
      "mean_token_accuracy": 0.11969415470957756,
      "num_tokens": 2733980.0,
      "step": 1485
    },
    {
      "entropy": 6.340108251571655,
      "epoch": 0.12518378491913462,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999809386948767,
      "loss": 6.2714,
      "mean_token_accuracy": 0.12543393298983574,
      "num_tokens": 2744013.0,
      "step": 1490
    },
    {
      "entropy": 6.2037091732025145,
      "epoch": 0.12560386473429952,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999805469059302,
      "loss": 6.3445,
      "mean_token_accuracy": 0.11714137867093086,
      "num_tokens": 2753385.0,
      "step": 1495
    },
    {
      "entropy": 6.296666240692138,
      "epoch": 0.1260239445494644,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999801511315693,
      "loss": 6.1931,
      "mean_token_accuracy": 0.12192152738571167,
      "num_tokens": 2762875.0,
      "step": 1500
    },
    {
      "entropy": 6.284714651107788,
      "epoch": 0.1264440243646293,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999797513718007,
      "loss": 6.2602,
      "mean_token_accuracy": 0.12598440870642663,
      "num_tokens": 2772182.0,
      "step": 1505
    },
    {
      "entropy": 6.161528825759888,
      "epoch": 0.12686410417979416,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999793476266317,
      "loss": 6.2127,
      "mean_token_accuracy": 0.12566941007971763,
      "num_tokens": 2780814.0,
      "step": 1510
    },
    {
      "entropy": 6.568186330795288,
      "epoch": 0.12728418399495905,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999789398960695,
      "loss": 6.483,
      "mean_token_accuracy": 0.11876541525125503,
      "num_tokens": 2791104.0,
      "step": 1515
    },
    {
      "entropy": 6.092951726913452,
      "epoch": 0.12770426381012392,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999785281801212,
      "loss": 6.1965,
      "mean_token_accuracy": 0.12290481179952621,
      "num_tokens": 2800081.0,
      "step": 1520
    },
    {
      "entropy": 6.291093444824218,
      "epoch": 0.1281243436252888,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499978112478794,
      "loss": 6.3106,
      "mean_token_accuracy": 0.12357923462986946,
      "num_tokens": 2809096.0,
      "step": 1525
    },
    {
      "entropy": 6.3231532096862795,
      "epoch": 0.1285444234404537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999776927920955,
      "loss": 6.2832,
      "mean_token_accuracy": 0.11848211586475373,
      "num_tokens": 2818857.0,
      "step": 1530
    },
    {
      "entropy": 6.26645565032959,
      "epoch": 0.12896450325561856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499977269120033,
      "loss": 6.3691,
      "mean_token_accuracy": 0.11801103353500367,
      "num_tokens": 2829332.0,
      "step": 1535
    },
    {
      "entropy": 6.273476028442383,
      "epoch": 0.12938458307078346,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499976841462614,
      "loss": 6.2806,
      "mean_token_accuracy": 0.1197669893503189,
      "num_tokens": 2839193.0,
      "step": 1540
    },
    {
      "entropy": 6.308599948883057,
      "epoch": 0.12980466288594833,
      "grad_norm": 1.0,
      "learning_rate": 0.000499976409819846,
      "loss": 6.2735,
      "mean_token_accuracy": 0.11774821653962135,
      "num_tokens": 2848535.0,
      "step": 1545
    },
    {
      "entropy": 6.131243658065796,
      "epoch": 0.1302247427011132,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999759741917369,
      "loss": 6.1661,
      "mean_token_accuracy": 0.12612850666046144,
      "num_tokens": 2858090.0,
      "step": 1550
    },
    {
      "entropy": 6.301682853698731,
      "epoch": 0.1306448225162781,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999755345782941,
      "loss": 6.3181,
      "mean_token_accuracy": 0.1226440578699112,
      "num_tokens": 2866984.0,
      "step": 1555
    },
    {
      "entropy": 6.184937286376953,
      "epoch": 0.13106490233144297,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999750909795256,
      "loss": 6.1325,
      "mean_token_accuracy": 0.12440444529056549,
      "num_tokens": 2876550.0,
      "step": 1560
    },
    {
      "entropy": 6.258828830718994,
      "epoch": 0.13148498214660786,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999746433954394,
      "loss": 6.241,
      "mean_token_accuracy": 0.12188669666647911,
      "num_tokens": 2885782.0,
      "step": 1565
    },
    {
      "entropy": 6.217999792098999,
      "epoch": 0.13190506196177273,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499974191826043,
      "loss": 6.2134,
      "mean_token_accuracy": 0.1303790420293808,
      "num_tokens": 2894807.0,
      "step": 1570
    },
    {
      "entropy": 6.286883115768433,
      "epoch": 0.1323251417769376,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999737362713448,
      "loss": 6.2503,
      "mean_token_accuracy": 0.12286639586091042,
      "num_tokens": 2904076.0,
      "step": 1575
    },
    {
      "entropy": 6.20257887840271,
      "epoch": 0.1327452215921025,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999732767313527,
      "loss": 6.1442,
      "mean_token_accuracy": 0.12661461755633355,
      "num_tokens": 2913761.0,
      "step": 1580
    },
    {
      "entropy": 6.346931409835816,
      "epoch": 0.13316530140726737,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999728132060746,
      "loss": 6.3898,
      "mean_token_accuracy": 0.12459043636918068,
      "num_tokens": 2922848.0,
      "step": 1585
    },
    {
      "entropy": 6.276056003570557,
      "epoch": 0.13358538122243227,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999723456955192,
      "loss": 6.271,
      "mean_token_accuracy": 0.1242086909711361,
      "num_tokens": 2932718.0,
      "step": 1590
    },
    {
      "entropy": 6.151839303970337,
      "epoch": 0.13400546103759714,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999718741996945,
      "loss": 6.2133,
      "mean_token_accuracy": 0.12332009747624398,
      "num_tokens": 2942686.0,
      "step": 1595
    },
    {
      "entropy": 6.2299316883087155,
      "epoch": 0.13442554085276204,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499971398718609,
      "loss": 6.194,
      "mean_token_accuracy": 0.12265397682785988,
      "num_tokens": 2952096.0,
      "step": 1600
    },
    {
      "entropy": 6.27053918838501,
      "epoch": 0.1348456206679269,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999709192522708,
      "loss": 6.2496,
      "mean_token_accuracy": 0.12414331436157226,
      "num_tokens": 2960660.0,
      "step": 1605
    },
    {
      "entropy": 6.299257707595825,
      "epoch": 0.13526570048309178,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999704358006887,
      "loss": 6.2485,
      "mean_token_accuracy": 0.1208350658416748,
      "num_tokens": 2969834.0,
      "step": 1610
    },
    {
      "entropy": 6.205888414382935,
      "epoch": 0.13568578029825668,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999699483638712,
      "loss": 6.2531,
      "mean_token_accuracy": 0.12345886677503586,
      "num_tokens": 2979023.0,
      "step": 1615
    },
    {
      "entropy": 6.2445619106292725,
      "epoch": 0.13610586011342155,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999694569418269,
      "loss": 6.2532,
      "mean_token_accuracy": 0.12339803278446197,
      "num_tokens": 2988083.0,
      "step": 1620
    },
    {
      "entropy": 6.20722599029541,
      "epoch": 0.13652593992858644,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999689615345645,
      "loss": 6.1689,
      "mean_token_accuracy": 0.1274717427790165,
      "num_tokens": 2997240.0,
      "step": 1625
    },
    {
      "entropy": 6.2464118003845215,
      "epoch": 0.1369460197437513,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999684621420928,
      "loss": 6.2565,
      "mean_token_accuracy": 0.12297938466072082,
      "num_tokens": 3007077.0,
      "step": 1630
    },
    {
      "entropy": 6.293700885772705,
      "epoch": 0.13736609955891618,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999679587644205,
      "loss": 6.2787,
      "mean_token_accuracy": 0.1208807609975338,
      "num_tokens": 3015821.0,
      "step": 1635
    },
    {
      "entropy": 6.17975435256958,
      "epoch": 0.13778617937408108,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999674514015568,
      "loss": 6.2054,
      "mean_token_accuracy": 0.1252801388502121,
      "num_tokens": 3025858.0,
      "step": 1640
    },
    {
      "entropy": 6.2544519901275635,
      "epoch": 0.13820625918924595,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999669400535105,
      "loss": 6.1887,
      "mean_token_accuracy": 0.11709433272480965,
      "num_tokens": 3035537.0,
      "step": 1645
    },
    {
      "entropy": 6.045716142654419,
      "epoch": 0.13862633900441085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999664247202907,
      "loss": 6.1026,
      "mean_token_accuracy": 0.12171815410256386,
      "num_tokens": 3044204.0,
      "step": 1650
    },
    {
      "entropy": 6.264136171340942,
      "epoch": 0.13904641881957572,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999659054019066,
      "loss": 6.2747,
      "mean_token_accuracy": 0.1242525890469551,
      "num_tokens": 3053111.0,
      "step": 1655
    },
    {
      "entropy": 6.191974449157715,
      "epoch": 0.1394664986347406,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999653820983673,
      "loss": 6.1818,
      "mean_token_accuracy": 0.12419796586036683,
      "num_tokens": 3062456.0,
      "step": 1660
    },
    {
      "entropy": 6.188469123840332,
      "epoch": 0.13988657844990549,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499964854809682,
      "loss": 6.2114,
      "mean_token_accuracy": 0.12520652115345002,
      "num_tokens": 3071132.0,
      "step": 1665
    },
    {
      "entropy": 6.217535066604614,
      "epoch": 0.14030665826507036,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999643235358602,
      "loss": 6.1733,
      "mean_token_accuracy": 0.12733130380511284,
      "num_tokens": 3080892.0,
      "step": 1670
    },
    {
      "entropy": 6.150455570220947,
      "epoch": 0.14072673808023525,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999637882769112,
      "loss": 6.1088,
      "mean_token_accuracy": 0.13008806556463243,
      "num_tokens": 3089874.0,
      "step": 1675
    },
    {
      "entropy": 6.232478332519531,
      "epoch": 0.14114681789540012,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999632490328447,
      "loss": 6.2504,
      "mean_token_accuracy": 0.12480302304029464,
      "num_tokens": 3099535.0,
      "step": 1680
    },
    {
      "entropy": 6.220491170883179,
      "epoch": 0.14156689771056502,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999627058036699,
      "loss": 6.1932,
      "mean_token_accuracy": 0.12512605339288713,
      "num_tokens": 3108772.0,
      "step": 1685
    },
    {
      "entropy": 6.254866027832032,
      "epoch": 0.1419869775257299,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999621585893966,
      "loss": 6.2305,
      "mean_token_accuracy": 0.11818314492702484,
      "num_tokens": 3118333.0,
      "step": 1690
    },
    {
      "entropy": 6.258799934387207,
      "epoch": 0.14240705734089476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999616073900346,
      "loss": 6.2544,
      "mean_token_accuracy": 0.1175099603831768,
      "num_tokens": 3127356.0,
      "step": 1695
    },
    {
      "entropy": 6.158872365951538,
      "epoch": 0.14282713715605966,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999610522055935,
      "loss": 6.2288,
      "mean_token_accuracy": 0.12072905600070953,
      "num_tokens": 3136859.0,
      "step": 1700
    },
    {
      "entropy": 6.2665447235107425,
      "epoch": 0.14324721697122453,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999604930360832,
      "loss": 6.2553,
      "mean_token_accuracy": 0.11907806620001793,
      "num_tokens": 3146607.0,
      "step": 1705
    },
    {
      "entropy": 6.134842443466186,
      "epoch": 0.14366729678638943,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999599298815136,
      "loss": 6.1946,
      "mean_token_accuracy": 0.12945861145853996,
      "num_tokens": 3156327.0,
      "step": 1710
    },
    {
      "entropy": 6.1354063034057615,
      "epoch": 0.1440873766015543,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004999593627418947,
      "loss": 6.1466,
      "mean_token_accuracy": 0.13011169731616973,
      "num_tokens": 3165559.0,
      "step": 1715
    },
    {
      "entropy": 6.2760594367980955,
      "epoch": 0.14450745641671917,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999587916172365,
      "loss": 6.247,
      "mean_token_accuracy": 0.11565925851464272,
      "num_tokens": 3173850.0,
      "step": 1720
    },
    {
      "entropy": 6.219358253479004,
      "epoch": 0.14492753623188406,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999582165075492,
      "loss": 6.1819,
      "mean_token_accuracy": 0.12041235193610192,
      "num_tokens": 3182838.0,
      "step": 1725
    },
    {
      "entropy": 6.098177671432495,
      "epoch": 0.14534761604704893,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999576374128429,
      "loss": 6.1848,
      "mean_token_accuracy": 0.12102061733603478,
      "num_tokens": 3191692.0,
      "step": 1730
    },
    {
      "entropy": 6.281035900115967,
      "epoch": 0.14576769586221383,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999570543331279,
      "loss": 6.2096,
      "mean_token_accuracy": 0.12320142686367035,
      "num_tokens": 3200069.0,
      "step": 1735
    },
    {
      "entropy": 6.174470567703247,
      "epoch": 0.1461877756773787,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999564672684145,
      "loss": 6.2813,
      "mean_token_accuracy": 0.11924844831228257,
      "num_tokens": 3209653.0,
      "step": 1740
    },
    {
      "entropy": 6.267424774169922,
      "epoch": 0.14660785549254357,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999558762187131,
      "loss": 6.1641,
      "mean_token_accuracy": 0.1311913624405861,
      "num_tokens": 3218313.0,
      "step": 1745
    },
    {
      "entropy": 6.102459383010864,
      "epoch": 0.14702793530770847,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999552811840342,
      "loss": 6.0922,
      "mean_token_accuracy": 0.12970734164118766,
      "num_tokens": 3227525.0,
      "step": 1750
    },
    {
      "entropy": 6.164451694488525,
      "epoch": 0.14744801512287334,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999546821643884,
      "loss": 6.2098,
      "mean_token_accuracy": 0.12783958092331887,
      "num_tokens": 3237022.0,
      "step": 1755
    },
    {
      "entropy": 6.150622081756592,
      "epoch": 0.14786809493803824,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999540791597861,
      "loss": 6.1154,
      "mean_token_accuracy": 0.1278826355934143,
      "num_tokens": 3246605.0,
      "step": 1760
    },
    {
      "entropy": 6.106969261169434,
      "epoch": 0.1482881747532031,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999534721702383,
      "loss": 6.065,
      "mean_token_accuracy": 0.13145939782261848,
      "num_tokens": 3255587.0,
      "step": 1765
    },
    {
      "entropy": 6.141215419769287,
      "epoch": 0.148708254568368,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999528611957553,
      "loss": 6.1666,
      "mean_token_accuracy": 0.1264194056391716,
      "num_tokens": 3265669.0,
      "step": 1770
    },
    {
      "entropy": 6.255412673950195,
      "epoch": 0.14912833438353287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999522462363485,
      "loss": 6.1518,
      "mean_token_accuracy": 0.13116262778639792,
      "num_tokens": 3275013.0,
      "step": 1775
    },
    {
      "entropy": 6.1518439769744875,
      "epoch": 0.14954841419869774,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999516272920283,
      "loss": 6.255,
      "mean_token_accuracy": 0.1255984991788864,
      "num_tokens": 3284723.0,
      "step": 1780
    },
    {
      "entropy": 6.11306095123291,
      "epoch": 0.14996849401386264,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499951004362806,
      "loss": 6.0833,
      "mean_token_accuracy": 0.12718486189842224,
      "num_tokens": 3293860.0,
      "step": 1785
    },
    {
      "entropy": 6.046157026290894,
      "epoch": 0.1503885738290275,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999503774486924,
      "loss": 6.1405,
      "mean_token_accuracy": 0.1226385310292244,
      "num_tokens": 3303158.0,
      "step": 1790
    },
    {
      "entropy": 6.138220262527466,
      "epoch": 0.1508086536441924,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999497465496987,
      "loss": 6.0637,
      "mean_token_accuracy": 0.12298208549618721,
      "num_tokens": 3313068.0,
      "step": 1795
    },
    {
      "entropy": 6.1492797374725345,
      "epoch": 0.15122873345935728,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499949111665836,
      "loss": 6.1591,
      "mean_token_accuracy": 0.12638431563973426,
      "num_tokens": 3321885.0,
      "step": 1800
    },
    {
      "entropy": 6.2002543926239015,
      "epoch": 0.15164881327452215,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999484727971158,
      "loss": 6.1371,
      "mean_token_accuracy": 0.12693015187978746,
      "num_tokens": 3330924.0,
      "step": 1805
    },
    {
      "entropy": 6.14166145324707,
      "epoch": 0.15206889308968705,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499947829943549,
      "loss": 6.176,
      "mean_token_accuracy": 0.12311631590127944,
      "num_tokens": 3340070.0,
      "step": 1810
    },
    {
      "entropy": 6.207650995254516,
      "epoch": 0.15248897290485192,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999471831051474,
      "loss": 6.1689,
      "mean_token_accuracy": 0.13365908414125444,
      "num_tokens": 3349870.0,
      "step": 1815
    },
    {
      "entropy": 6.160492658615112,
      "epoch": 0.1529090527200168,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999465322819222,
      "loss": 6.2169,
      "mean_token_accuracy": 0.12209457084536553,
      "num_tokens": 3359573.0,
      "step": 1820
    },
    {
      "entropy": 6.222156381607055,
      "epoch": 0.15332913253518168,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999458774738851,
      "loss": 6.1684,
      "mean_token_accuracy": 0.13460491448640824,
      "num_tokens": 3368577.0,
      "step": 1825
    },
    {
      "entropy": 6.116939735412598,
      "epoch": 0.15374921235034655,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999452186810476,
      "loss": 6.1162,
      "mean_token_accuracy": 0.13111207485198975,
      "num_tokens": 3377801.0,
      "step": 1830
    },
    {
      "entropy": 6.230794095993042,
      "epoch": 0.15416929216551145,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999445559034214,
      "loss": 6.1624,
      "mean_token_accuracy": 0.12796897292137147,
      "num_tokens": 3386666.0,
      "step": 1835
    },
    {
      "entropy": 6.314156770706177,
      "epoch": 0.15458937198067632,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999438891410181,
      "loss": 6.3117,
      "mean_token_accuracy": 0.12008170932531356,
      "num_tokens": 3396086.0,
      "step": 1840
    },
    {
      "entropy": 6.129251384735108,
      "epoch": 0.15500945179584122,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999432183938496,
      "loss": 6.2244,
      "mean_token_accuracy": 0.13115857616066934,
      "num_tokens": 3404894.0,
      "step": 1845
    },
    {
      "entropy": 6.144184207916259,
      "epoch": 0.1554295316110061,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999425436619279,
      "loss": 6.2119,
      "mean_token_accuracy": 0.12016721740365029,
      "num_tokens": 3414172.0,
      "step": 1850
    },
    {
      "entropy": 6.27680606842041,
      "epoch": 0.15584961142617096,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499941864945265,
      "loss": 6.1711,
      "mean_token_accuracy": 0.12068818733096123,
      "num_tokens": 3423409.0,
      "step": 1855
    },
    {
      "entropy": 6.115350151062012,
      "epoch": 0.15626969124133586,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999411822438726,
      "loss": 6.1448,
      "mean_token_accuracy": 0.12664692029356955,
      "num_tokens": 3433047.0,
      "step": 1860
    },
    {
      "entropy": 6.1765364646911625,
      "epoch": 0.15668977105650073,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499940495557763,
      "loss": 6.1229,
      "mean_token_accuracy": 0.12646402716636657,
      "num_tokens": 3442490.0,
      "step": 1865
    },
    {
      "entropy": 6.212595748901367,
      "epoch": 0.15710985087166562,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999398048869485,
      "loss": 6.1955,
      "mean_token_accuracy": 0.12634197920560836,
      "num_tokens": 3451804.0,
      "step": 1870
    },
    {
      "entropy": 6.19734468460083,
      "epoch": 0.1575299306868305,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499939110231441,
      "loss": 6.1767,
      "mean_token_accuracy": 0.1279009036719799,
      "num_tokens": 3461481.0,
      "step": 1875
    },
    {
      "entropy": 6.233086156845093,
      "epoch": 0.1579500105019954,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999384115912531,
      "loss": 6.2344,
      "mean_token_accuracy": 0.12624539732933043,
      "num_tokens": 3471798.0,
      "step": 1880
    },
    {
      "entropy": 6.015159845352173,
      "epoch": 0.15837009031716026,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499937708966397,
      "loss": 6.1259,
      "mean_token_accuracy": 0.12552440091967582,
      "num_tokens": 3481386.0,
      "step": 1885
    },
    {
      "entropy": 6.193439149856568,
      "epoch": 0.15879017013232513,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999370023568853,
      "loss": 6.1184,
      "mean_token_accuracy": 0.12572901472449302,
      "num_tokens": 3489981.0,
      "step": 1890
    },
    {
      "entropy": 6.112298202514649,
      "epoch": 0.15921024994749003,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999362917627304,
      "loss": 6.0882,
      "mean_token_accuracy": 0.1290791854262352,
      "num_tokens": 3498551.0,
      "step": 1895
    },
    {
      "entropy": 6.154214668273926,
      "epoch": 0.1596303297626549,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999355771839448,
      "loss": 6.0678,
      "mean_token_accuracy": 0.13199156373739243,
      "num_tokens": 3507921.0,
      "step": 1900
    },
    {
      "entropy": 6.25932993888855,
      "epoch": 0.1600504095778198,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999348586205414,
      "loss": 6.2391,
      "mean_token_accuracy": 0.12772736102342605,
      "num_tokens": 3517570.0,
      "step": 1905
    },
    {
      "entropy": 6.199037790298462,
      "epoch": 0.16047048939298467,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999341360725327,
      "loss": 6.2333,
      "mean_token_accuracy": 0.12291170060634612,
      "num_tokens": 3526774.0,
      "step": 1910
    },
    {
      "entropy": 6.160675954818726,
      "epoch": 0.16089056920814954,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999334095399317,
      "loss": 6.1636,
      "mean_token_accuracy": 0.13550060987472534,
      "num_tokens": 3535319.0,
      "step": 1915
    },
    {
      "entropy": 6.045867156982422,
      "epoch": 0.16131064902331443,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999326790227512,
      "loss": 6.1319,
      "mean_token_accuracy": 0.13111512288451194,
      "num_tokens": 3544468.0,
      "step": 1920
    },
    {
      "entropy": 6.046793222427368,
      "epoch": 0.1617307288384793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999319445210041,
      "loss": 6.0122,
      "mean_token_accuracy": 0.13604443669319152,
      "num_tokens": 3553529.0,
      "step": 1925
    },
    {
      "entropy": 6.111843824386597,
      "epoch": 0.1621508086536442,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999312060347034,
      "loss": 6.0683,
      "mean_token_accuracy": 0.12906411960721015,
      "num_tokens": 3563053.0,
      "step": 1930
    },
    {
      "entropy": 6.050889587402343,
      "epoch": 0.16257088846880907,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999304635638621,
      "loss": 6.0231,
      "mean_token_accuracy": 0.13277052119374275,
      "num_tokens": 3571877.0,
      "step": 1935
    },
    {
      "entropy": 6.067962026596069,
      "epoch": 0.16299096828397394,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999297171084935,
      "loss": 6.0714,
      "mean_token_accuracy": 0.1285587027668953,
      "num_tokens": 3581496.0,
      "step": 1940
    },
    {
      "entropy": 6.136378002166748,
      "epoch": 0.16341104809913884,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999289666686109,
      "loss": 6.0886,
      "mean_token_accuracy": 0.12886847257614137,
      "num_tokens": 3590752.0,
      "step": 1945
    },
    {
      "entropy": 6.010327911376953,
      "epoch": 0.1638311279143037,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999282122442274,
      "loss": 6.0834,
      "mean_token_accuracy": 0.13124383464455605,
      "num_tokens": 3599885.0,
      "step": 1950
    },
    {
      "entropy": 6.264139842987061,
      "epoch": 0.1642512077294686,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999274538353564,
      "loss": 6.1701,
      "mean_token_accuracy": 0.12521142810583114,
      "num_tokens": 3610039.0,
      "step": 1955
    },
    {
      "entropy": 6.073106908798218,
      "epoch": 0.16467128754463348,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999266914420114,
      "loss": 6.1051,
      "mean_token_accuracy": 0.12478168606758118,
      "num_tokens": 3619954.0,
      "step": 1960
    },
    {
      "entropy": 6.187624120712281,
      "epoch": 0.16509136735979837,
      "grad_norm": 1.125,
      "learning_rate": 0.000499925925064206,
      "loss": 6.0438,
      "mean_token_accuracy": 0.13393656834959983,
      "num_tokens": 3628164.0,
      "step": 1965
    },
    {
      "entropy": 6.205664491653442,
      "epoch": 0.16551144717496324,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999251547019535,
      "loss": 6.1949,
      "mean_token_accuracy": 0.13081590309739113,
      "num_tokens": 3636778.0,
      "step": 1970
    },
    {
      "entropy": 6.189227771759033,
      "epoch": 0.16593152699012811,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999243803552678,
      "loss": 6.1562,
      "mean_token_accuracy": 0.1314916841685772,
      "num_tokens": 3647046.0,
      "step": 1975
    },
    {
      "entropy": 6.080912733078003,
      "epoch": 0.166351606805293,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999236020241625,
      "loss": 6.0778,
      "mean_token_accuracy": 0.12676682993769645,
      "num_tokens": 3656130.0,
      "step": 1980
    },
    {
      "entropy": 6.122231149673462,
      "epoch": 0.16677168662045788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999228197086514,
      "loss": 6.1524,
      "mean_token_accuracy": 0.12139208093285561,
      "num_tokens": 3666145.0,
      "step": 1985
    },
    {
      "entropy": 6.180514907836914,
      "epoch": 0.16719176643562278,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999220334087484,
      "loss": 6.1973,
      "mean_token_accuracy": 0.12297596782445908,
      "num_tokens": 3676722.0,
      "step": 1990
    },
    {
      "entropy": 6.160478258132935,
      "epoch": 0.16761184625078765,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999212431244673,
      "loss": 6.1951,
      "mean_token_accuracy": 0.12390567734837532,
      "num_tokens": 3685880.0,
      "step": 1995
    },
    {
      "entropy": 6.056979942321777,
      "epoch": 0.16803192606595252,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999204488558222,
      "loss": 6.0309,
      "mean_token_accuracy": 0.13338653817772866,
      "num_tokens": 3695167.0,
      "step": 2000
    },
    {
      "entropy": 6.132491636276245,
      "epoch": 0.16845200588111742,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999196506028273,
      "loss": 6.1254,
      "mean_token_accuracy": 0.12816390842199327,
      "num_tokens": 3703700.0,
      "step": 2005
    },
    {
      "entropy": 6.12127537727356,
      "epoch": 0.1688720856962823,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999188483654965,
      "loss": 6.0676,
      "mean_token_accuracy": 0.13127523884177209,
      "num_tokens": 3712825.0,
      "step": 2010
    },
    {
      "entropy": 6.0288161277771,
      "epoch": 0.16929216551144718,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999180421438442,
      "loss": 6.0433,
      "mean_token_accuracy": 0.13027536049485205,
      "num_tokens": 3721807.0,
      "step": 2015
    },
    {
      "entropy": 6.227120208740234,
      "epoch": 0.16971224532661205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999172319378846,
      "loss": 6.215,
      "mean_token_accuracy": 0.12012537494301796,
      "num_tokens": 3730502.0,
      "step": 2020
    },
    {
      "entropy": 6.142453145980835,
      "epoch": 0.17013232514177692,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999164177476319,
      "loss": 6.1007,
      "mean_token_accuracy": 0.13319698646664618,
      "num_tokens": 3739696.0,
      "step": 2025
    },
    {
      "entropy": 5.983246898651123,
      "epoch": 0.17055240495694182,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999155995731009,
      "loss": 6.0931,
      "mean_token_accuracy": 0.12937767952680587,
      "num_tokens": 3748675.0,
      "step": 2030
    },
    {
      "entropy": 6.279529285430908,
      "epoch": 0.1709724847721067,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999147774143057,
      "loss": 6.1818,
      "mean_token_accuracy": 0.12119651660323143,
      "num_tokens": 3757714.0,
      "step": 2035
    },
    {
      "entropy": 6.021195363998413,
      "epoch": 0.1713925645872716,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499913951271261,
      "loss": 5.9995,
      "mean_token_accuracy": 0.13538317754864693,
      "num_tokens": 3767589.0,
      "step": 2040
    },
    {
      "entropy": 6.069708347320557,
      "epoch": 0.17181264440243646,
      "grad_norm": 1.25,
      "learning_rate": 0.0004999131211439816,
      "loss": 6.1103,
      "mean_token_accuracy": 0.1307745970785618,
      "num_tokens": 3777261.0,
      "step": 2045
    },
    {
      "entropy": 6.199843311309815,
      "epoch": 0.17223272421760136,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499912287032482,
      "loss": 6.0616,
      "mean_token_accuracy": 0.138626891374588,
      "num_tokens": 3786658.0,
      "step": 2050
    },
    {
      "entropy": 5.944658851623535,
      "epoch": 0.17265280403276623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499911448936777,
      "loss": 6.0687,
      "mean_token_accuracy": 0.13718581050634385,
      "num_tokens": 3794977.0,
      "step": 2055
    },
    {
      "entropy": 6.036355209350586,
      "epoch": 0.1730728838479311,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999106068568816,
      "loss": 6.1519,
      "mean_token_accuracy": 0.12767373770475388,
      "num_tokens": 3805138.0,
      "step": 2060
    },
    {
      "entropy": 6.216224002838135,
      "epoch": 0.173492963663096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999097607928106,
      "loss": 6.0775,
      "mean_token_accuracy": 0.13682154789566994,
      "num_tokens": 3814444.0,
      "step": 2065
    },
    {
      "entropy": 6.071376657485962,
      "epoch": 0.17391304347826086,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999089107445788,
      "loss": 6.0367,
      "mean_token_accuracy": 0.12884577140212058,
      "num_tokens": 3822859.0,
      "step": 2070
    },
    {
      "entropy": 6.008500671386718,
      "epoch": 0.17433312329342576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999080567122016,
      "loss": 6.0618,
      "mean_token_accuracy": 0.13177087977528573,
      "num_tokens": 3833159.0,
      "step": 2075
    },
    {
      "entropy": 6.108531808853149,
      "epoch": 0.17475320310859063,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999071986956941,
      "loss": 6.0712,
      "mean_token_accuracy": 0.132802564650774,
      "num_tokens": 3842136.0,
      "step": 2080
    },
    {
      "entropy": 6.1049566745758055,
      "epoch": 0.1751732829237555,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999063366950713,
      "loss": 6.1334,
      "mean_token_accuracy": 0.12718930542469026,
      "num_tokens": 3851406.0,
      "step": 2085
    },
    {
      "entropy": 6.059695482254028,
      "epoch": 0.1755933627389204,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999054707103486,
      "loss": 6.0576,
      "mean_token_accuracy": 0.13101131469011307,
      "num_tokens": 3861061.0,
      "step": 2090
    },
    {
      "entropy": 6.151092004776001,
      "epoch": 0.17601344255408527,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999046007415412,
      "loss": 6.0408,
      "mean_token_accuracy": 0.12858548611402512,
      "num_tokens": 3870357.0,
      "step": 2095
    },
    {
      "entropy": 6.117391204833984,
      "epoch": 0.17643352236925017,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999037267886646,
      "loss": 6.0479,
      "mean_token_accuracy": 0.1312727876007557,
      "num_tokens": 3879393.0,
      "step": 2100
    },
    {
      "entropy": 6.020799350738526,
      "epoch": 0.17685360218441504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999028488517343,
      "loss": 6.0773,
      "mean_token_accuracy": 0.1310334399342537,
      "num_tokens": 3888030.0,
      "step": 2105
    },
    {
      "entropy": 6.100592184066772,
      "epoch": 0.1772736819995799,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999019669307659,
      "loss": 6.0848,
      "mean_token_accuracy": 0.13095394000411034,
      "num_tokens": 3897430.0,
      "step": 2110
    },
    {
      "entropy": 6.128114986419678,
      "epoch": 0.1776937618147448,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999010810257749,
      "loss": 6.1085,
      "mean_token_accuracy": 0.1252062700688839,
      "num_tokens": 3907711.0,
      "step": 2115
    },
    {
      "entropy": 6.000101137161255,
      "epoch": 0.17811384162990967,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999001911367771,
      "loss": 6.0329,
      "mean_token_accuracy": 0.13599146604537965,
      "num_tokens": 3915816.0,
      "step": 2120
    },
    {
      "entropy": 6.115037250518799,
      "epoch": 0.17853392144507457,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998992972637883,
      "loss": 6.1562,
      "mean_token_accuracy": 0.12152478694915772,
      "num_tokens": 3925162.0,
      "step": 2125
    },
    {
      "entropy": 6.161215543746948,
      "epoch": 0.17895400126023944,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998983994068242,
      "loss": 6.0325,
      "mean_token_accuracy": 0.13381896317005157,
      "num_tokens": 3934476.0,
      "step": 2130
    },
    {
      "entropy": 6.025839996337891,
      "epoch": 0.17937408107540434,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998974975659006,
      "loss": 6.0854,
      "mean_token_accuracy": 0.1257789060473442,
      "num_tokens": 3943501.0,
      "step": 2135
    },
    {
      "entropy": 6.134726524353027,
      "epoch": 0.1797941608905692,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998965917410338,
      "loss": 6.0793,
      "mean_token_accuracy": 0.12519558221101762,
      "num_tokens": 3953663.0,
      "step": 2140
    },
    {
      "entropy": 6.026215839385986,
      "epoch": 0.18021424070573408,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998956819322397,
      "loss": 6.0415,
      "mean_token_accuracy": 0.1311854176223278,
      "num_tokens": 3962634.0,
      "step": 2145
    },
    {
      "entropy": 6.095232534408569,
      "epoch": 0.18063432052089898,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998947681395343,
      "loss": 6.0476,
      "mean_token_accuracy": 0.13477237224578859,
      "num_tokens": 3972496.0,
      "step": 2150
    },
    {
      "entropy": 6.216877603530884,
      "epoch": 0.18105440033606385,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499893850362934,
      "loss": 6.294,
      "mean_token_accuracy": 0.12388142496347428,
      "num_tokens": 3980724.0,
      "step": 2155
    },
    {
      "entropy": 6.148508739471436,
      "epoch": 0.18147448015122875,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998929286024548,
      "loss": 6.1193,
      "mean_token_accuracy": 0.12675963416695596,
      "num_tokens": 3989842.0,
      "step": 2160
    },
    {
      "entropy": 6.089916467666626,
      "epoch": 0.18189455996639362,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998920028581133,
      "loss": 6.0396,
      "mean_token_accuracy": 0.13960086852312087,
      "num_tokens": 3998534.0,
      "step": 2165
    },
    {
      "entropy": 6.071965646743775,
      "epoch": 0.18231463978155849,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998910731299258,
      "loss": 6.066,
      "mean_token_accuracy": 0.12839170619845391,
      "num_tokens": 4007677.0,
      "step": 2170
    },
    {
      "entropy": 6.075281476974487,
      "epoch": 0.18273471959672338,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998901394179085,
      "loss": 6.1252,
      "mean_token_accuracy": 0.1271742030978203,
      "num_tokens": 4016347.0,
      "step": 2175
    },
    {
      "entropy": 6.099024391174316,
      "epoch": 0.18315479941188825,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998892017220784,
      "loss": 5.981,
      "mean_token_accuracy": 0.13410005643963813,
      "num_tokens": 4025199.0,
      "step": 2180
    },
    {
      "entropy": 6.039233541488647,
      "epoch": 0.18357487922705315,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998882600424519,
      "loss": 6.0482,
      "mean_token_accuracy": 0.12497256994247437,
      "num_tokens": 4033933.0,
      "step": 2185
    },
    {
      "entropy": 6.077206420898437,
      "epoch": 0.18399495904221802,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998873143790455,
      "loss": 5.9828,
      "mean_token_accuracy": 0.14159614518284797,
      "num_tokens": 4042891.0,
      "step": 2190
    },
    {
      "entropy": 6.073574495315552,
      "epoch": 0.1844150388573829,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998863647318763,
      "loss": 6.0991,
      "mean_token_accuracy": 0.12677306681871414,
      "num_tokens": 4051123.0,
      "step": 2195
    },
    {
      "entropy": 6.040490913391113,
      "epoch": 0.1848351186725478,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998854111009608,
      "loss": 6.0708,
      "mean_token_accuracy": 0.12666793614625932,
      "num_tokens": 4060025.0,
      "step": 2200
    },
    {
      "entropy": 6.00008749961853,
      "epoch": 0.18525519848771266,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998844534863161,
      "loss": 5.9771,
      "mean_token_accuracy": 0.12744748294353486,
      "num_tokens": 4069363.0,
      "step": 2205
    },
    {
      "entropy": 6.136739826202392,
      "epoch": 0.18567527830287756,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998834918879592,
      "loss": 6.1326,
      "mean_token_accuracy": 0.1333842933177948,
      "num_tokens": 4078855.0,
      "step": 2210
    },
    {
      "entropy": 6.1249613761901855,
      "epoch": 0.18609535811804243,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499882526305907,
      "loss": 6.0922,
      "mean_token_accuracy": 0.1307423233985901,
      "num_tokens": 4087801.0,
      "step": 2215
    },
    {
      "entropy": 6.088332986831665,
      "epoch": 0.18651543793320732,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998815567401765,
      "loss": 6.1042,
      "mean_token_accuracy": 0.1286585159599781,
      "num_tokens": 4096949.0,
      "step": 2220
    },
    {
      "entropy": 6.1469251155853275,
      "epoch": 0.1869355177483722,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998805831907851,
      "loss": 6.0481,
      "mean_token_accuracy": 0.13261425495147705,
      "num_tokens": 4105399.0,
      "step": 2225
    },
    {
      "entropy": 6.030412244796753,
      "epoch": 0.18735559756353706,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998796056577501,
      "loss": 6.0131,
      "mean_token_accuracy": 0.12757696062326432,
      "num_tokens": 4113873.0,
      "step": 2230
    },
    {
      "entropy": 6.031377696990967,
      "epoch": 0.18777567737870196,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998786241410886,
      "loss": 6.0712,
      "mean_token_accuracy": 0.13365804105997087,
      "num_tokens": 4123528.0,
      "step": 2235
    },
    {
      "entropy": 6.171572303771972,
      "epoch": 0.18819575719386683,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499877638640818,
      "loss": 6.0658,
      "mean_token_accuracy": 0.1285923771560192,
      "num_tokens": 4133370.0,
      "step": 2240
    },
    {
      "entropy": 5.986340188980103,
      "epoch": 0.18861583700903173,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499876649156956,
      "loss": 5.9815,
      "mean_token_accuracy": 0.13429070338606836,
      "num_tokens": 4142370.0,
      "step": 2245
    },
    {
      "entropy": 6.028431034088134,
      "epoch": 0.1890359168241966,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998756556895196,
      "loss": 6.0667,
      "mean_token_accuracy": 0.13125480934977532,
      "num_tokens": 4152367.0,
      "step": 2250
    },
    {
      "entropy": 6.113050174713135,
      "epoch": 0.18945599663936147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499874658238527,
      "loss": 6.0648,
      "mean_token_accuracy": 0.1346374697983265,
      "num_tokens": 4161126.0,
      "step": 2255
    },
    {
      "entropy": 6.028095388412476,
      "epoch": 0.18987607645452637,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998736568039957,
      "loss": 5.96,
      "mean_token_accuracy": 0.13207067623734475,
      "num_tokens": 4169910.0,
      "step": 2260
    },
    {
      "entropy": 6.0694433689117435,
      "epoch": 0.19029615626969124,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998726513859432,
      "loss": 6.1159,
      "mean_token_accuracy": 0.12696689143776893,
      "num_tokens": 4179893.0,
      "step": 2265
    },
    {
      "entropy": 6.183452177047729,
      "epoch": 0.19071623608485613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998716419843875,
      "loss": 6.1192,
      "mean_token_accuracy": 0.1379365175962448,
      "num_tokens": 4190065.0,
      "step": 2270
    },
    {
      "entropy": 5.935803985595703,
      "epoch": 0.191136315900021,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998706285993465,
      "loss": 6.0341,
      "mean_token_accuracy": 0.13346357494592667,
      "num_tokens": 4198395.0,
      "step": 2275
    },
    {
      "entropy": 6.13513503074646,
      "epoch": 0.19155639571518587,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998696112308381,
      "loss": 6.066,
      "mean_token_accuracy": 0.1285228006541729,
      "num_tokens": 4207555.0,
      "step": 2280
    },
    {
      "entropy": 5.988098859786987,
      "epoch": 0.19197647553035077,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998685898788803,
      "loss": 6.0031,
      "mean_token_accuracy": 0.13245714083313942,
      "num_tokens": 4216533.0,
      "step": 2285
    },
    {
      "entropy": 6.156089878082275,
      "epoch": 0.19239655534551564,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998675645434914,
      "loss": 6.1082,
      "mean_token_accuracy": 0.13477368876338006,
      "num_tokens": 4225575.0,
      "step": 2290
    },
    {
      "entropy": 5.991475677490234,
      "epoch": 0.19281663516068054,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998665352246891,
      "loss": 5.8834,
      "mean_token_accuracy": 0.1395990490913391,
      "num_tokens": 4234306.0,
      "step": 2295
    },
    {
      "entropy": 5.991662073135376,
      "epoch": 0.1932367149758454,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998655019224921,
      "loss": 6.0833,
      "mean_token_accuracy": 0.13282228037714958,
      "num_tokens": 4243998.0,
      "step": 2300
    },
    {
      "entropy": 6.109613370895386,
      "epoch": 0.19365679479101028,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998644646369185,
      "loss": 5.9798,
      "mean_token_accuracy": 0.1297495998442173,
      "num_tokens": 4253653.0,
      "step": 2305
    },
    {
      "entropy": 5.980562829971314,
      "epoch": 0.19407687460617518,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998634233679865,
      "loss": 6.0677,
      "mean_token_accuracy": 0.12498517185449601,
      "num_tokens": 4263305.0,
      "step": 2310
    },
    {
      "entropy": 6.036490631103516,
      "epoch": 0.19449695442134005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499862378115715,
      "loss": 5.942,
      "mean_token_accuracy": 0.13776571899652482,
      "num_tokens": 4272212.0,
      "step": 2315
    },
    {
      "entropy": 6.152922439575195,
      "epoch": 0.19491703423650494,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998613288801221,
      "loss": 6.1425,
      "mean_token_accuracy": 0.12971725761890412,
      "num_tokens": 4281445.0,
      "step": 2320
    },
    {
      "entropy": 6.123716592788696,
      "epoch": 0.1953371140516698,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998602756612267,
      "loss": 6.0573,
      "mean_token_accuracy": 0.13039504289627074,
      "num_tokens": 4290938.0,
      "step": 2325
    },
    {
      "entropy": 6.004360866546631,
      "epoch": 0.1957571938668347,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998592184590471,
      "loss": 6.0764,
      "mean_token_accuracy": 0.13114980682730676,
      "num_tokens": 4300022.0,
      "step": 2330
    },
    {
      "entropy": 5.995278835296631,
      "epoch": 0.19617727368199958,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998581572736024,
      "loss": 5.9693,
      "mean_token_accuracy": 0.1386754259467125,
      "num_tokens": 4308910.0,
      "step": 2335
    },
    {
      "entropy": 5.990830326080323,
      "epoch": 0.19659735349716445,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998570921049112,
      "loss": 5.964,
      "mean_token_accuracy": 0.13531014919281006,
      "num_tokens": 4317136.0,
      "step": 2340
    },
    {
      "entropy": 6.019982814788818,
      "epoch": 0.19701743331232935,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998560229529924,
      "loss": 6.0043,
      "mean_token_accuracy": 0.13840724304318427,
      "num_tokens": 4326163.0,
      "step": 2345
    },
    {
      "entropy": 6.181583261489868,
      "epoch": 0.19743751312749422,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998549498178649,
      "loss": 6.1515,
      "mean_token_accuracy": 0.13239141255617143,
      "num_tokens": 4335837.0,
      "step": 2350
    },
    {
      "entropy": 6.1028111457824705,
      "epoch": 0.19785759294265912,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998538726995477,
      "loss": 6.0502,
      "mean_token_accuracy": 0.13465801179409026,
      "num_tokens": 4345108.0,
      "step": 2355
    },
    {
      "entropy": 6.11204948425293,
      "epoch": 0.198277672757824,
      "grad_norm": 1.078125,
      "learning_rate": 0.00049985279159806,
      "loss": 6.0904,
      "mean_token_accuracy": 0.13041200041770934,
      "num_tokens": 4353761.0,
      "step": 2360
    },
    {
      "entropy": 6.025763607025146,
      "epoch": 0.19869775257298886,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998517065134208,
      "loss": 6.0492,
      "mean_token_accuracy": 0.1321948856115341,
      "num_tokens": 4363244.0,
      "step": 2365
    },
    {
      "entropy": 6.045213079452514,
      "epoch": 0.19911783238815375,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998506174456494,
      "loss": 6.0414,
      "mean_token_accuracy": 0.1313652828335762,
      "num_tokens": 4373034.0,
      "step": 2370
    },
    {
      "entropy": 6.018647909164429,
      "epoch": 0.19953791220331862,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998495243947653,
      "loss": 5.9888,
      "mean_token_accuracy": 0.12499892637133599,
      "num_tokens": 4382554.0,
      "step": 2375
    },
    {
      "entropy": 6.050378942489624,
      "epoch": 0.19995799201848352,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998484273607875,
      "loss": 6.0109,
      "mean_token_accuracy": 0.13378100991249084,
      "num_tokens": 4391001.0,
      "step": 2380
    },
    {
      "entropy": 5.911934089660645,
      "epoch": 0.2003780718336484,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998473263437356,
      "loss": 5.9179,
      "mean_token_accuracy": 0.13600271940231323,
      "num_tokens": 4400632.0,
      "step": 2385
    },
    {
      "entropy": 6.02356653213501,
      "epoch": 0.20079815164881326,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499846221343629,
      "loss": 6.0212,
      "mean_token_accuracy": 0.12660589516162873,
      "num_tokens": 4409565.0,
      "step": 2390
    },
    {
      "entropy": 6.01599760055542,
      "epoch": 0.20121823146397816,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998451123604875,
      "loss": 5.9683,
      "mean_token_accuracy": 0.13648212924599648,
      "num_tokens": 4418384.0,
      "step": 2395
    },
    {
      "entropy": 6.052209377288818,
      "epoch": 0.20163831127914303,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004998439993943306,
      "loss": 6.0768,
      "mean_token_accuracy": 0.13455061092972756,
      "num_tokens": 4427581.0,
      "step": 2400
    },
    {
      "entropy": 6.126885080337525,
      "epoch": 0.20205839109430793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998428824451779,
      "loss": 6.0655,
      "mean_token_accuracy": 0.12827305421233176,
      "num_tokens": 4436572.0,
      "step": 2405
    },
    {
      "entropy": 6.031488513946533,
      "epoch": 0.2024784709094728,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004998417615130495,
      "loss": 6.0686,
      "mean_token_accuracy": 0.13068403899669648,
      "num_tokens": 4445230.0,
      "step": 2410
    },
    {
      "entropy": 6.176716995239258,
      "epoch": 0.2028985507246377,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998406365979649,
      "loss": 6.1411,
      "mean_token_accuracy": 0.13211808502674102,
      "num_tokens": 4454251.0,
      "step": 2415
    },
    {
      "entropy": 6.0446735382080075,
      "epoch": 0.20331863053980256,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998395076999443,
      "loss": 5.9835,
      "mean_token_accuracy": 0.13375458046793937,
      "num_tokens": 4463949.0,
      "step": 2420
    },
    {
      "entropy": 6.100032329559326,
      "epoch": 0.20373871035496743,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998383748190076,
      "loss": 6.1638,
      "mean_token_accuracy": 0.12677136287093163,
      "num_tokens": 4473373.0,
      "step": 2425
    },
    {
      "entropy": 6.1406354904174805,
      "epoch": 0.20415879017013233,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004998372379551748,
      "loss": 6.0192,
      "mean_token_accuracy": 0.1297541990876198,
      "num_tokens": 4482303.0,
      "step": 2430
    },
    {
      "entropy": 5.960326719284057,
      "epoch": 0.2045788699852972,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998360971084663,
      "loss": 5.9691,
      "mean_token_accuracy": 0.13114270120859145,
      "num_tokens": 4491214.0,
      "step": 2435
    },
    {
      "entropy": 5.993530750274658,
      "epoch": 0.2049989498004621,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998349522789019,
      "loss": 5.8971,
      "mean_token_accuracy": 0.13634659722447395,
      "num_tokens": 4500099.0,
      "step": 2440
    },
    {
      "entropy": 5.949355411529541,
      "epoch": 0.20541902961562697,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004998338034665021,
      "loss": 5.9773,
      "mean_token_accuracy": 0.13802963197231294,
      "num_tokens": 4509893.0,
      "step": 2445
    },
    {
      "entropy": 6.010091781616211,
      "epoch": 0.20583910943079184,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998326506712872,
      "loss": 5.9481,
      "mean_token_accuracy": 0.1345847800374031,
      "num_tokens": 4518606.0,
      "step": 2450
    },
    {
      "entropy": 6.054938316345215,
      "epoch": 0.20625918924595674,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998314938932778,
      "loss": 6.0368,
      "mean_token_accuracy": 0.13336761966347693,
      "num_tokens": 4528392.0,
      "step": 2455
    },
    {
      "entropy": 6.06166090965271,
      "epoch": 0.2066792690611216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998303331324943,
      "loss": 6.0068,
      "mean_token_accuracy": 0.13653545677661896,
      "num_tokens": 4536983.0,
      "step": 2460
    },
    {
      "entropy": 5.900071907043457,
      "epoch": 0.2070993488762865,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998291683889571,
      "loss": 5.9105,
      "mean_token_accuracy": 0.14012753888964652,
      "num_tokens": 4544967.0,
      "step": 2465
    },
    {
      "entropy": 6.0055500030517575,
      "epoch": 0.20751942869145137,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499827999662687,
      "loss": 5.9915,
      "mean_token_accuracy": 0.1313713811337948,
      "num_tokens": 4554646.0,
      "step": 2470
    },
    {
      "entropy": 6.090430212020874,
      "epoch": 0.20793950850661624,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998268269537046,
      "loss": 6.0166,
      "mean_token_accuracy": 0.13576155975461007,
      "num_tokens": 4564040.0,
      "step": 2475
    },
    {
      "entropy": 6.014799499511719,
      "epoch": 0.20835958832178114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998256502620308,
      "loss": 6.0293,
      "mean_token_accuracy": 0.13742104843258857,
      "num_tokens": 4573758.0,
      "step": 2480
    },
    {
      "entropy": 6.133381319046021,
      "epoch": 0.208779668136946,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998244695876864,
      "loss": 6.0616,
      "mean_token_accuracy": 0.13331351354718207,
      "num_tokens": 4582097.0,
      "step": 2485
    },
    {
      "entropy": 5.951827907562256,
      "epoch": 0.2091997479521109,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998232849306921,
      "loss": 6.0184,
      "mean_token_accuracy": 0.13623687997460365,
      "num_tokens": 4590687.0,
      "step": 2490
    },
    {
      "entropy": 6.113849639892578,
      "epoch": 0.20961982776727578,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998220962910693,
      "loss": 6.0063,
      "mean_token_accuracy": 0.1295908585190773,
      "num_tokens": 4599497.0,
      "step": 2495
    },
    {
      "entropy": 6.004902267456055,
      "epoch": 0.21003990758244068,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998209036688386,
      "loss": 5.9761,
      "mean_token_accuracy": 0.13407944440841674,
      "num_tokens": 4607958.0,
      "step": 2500
    },
    {
      "entropy": 6.0813335418701175,
      "epoch": 0.21045998739760555,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998197070640216,
      "loss": 6.1175,
      "mean_token_accuracy": 0.1263462521135807,
      "num_tokens": 4617515.0,
      "step": 2505
    },
    {
      "entropy": 6.137440729141235,
      "epoch": 0.21088006721277042,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998185064766391,
      "loss": 5.9907,
      "mean_token_accuracy": 0.1353794366121292,
      "num_tokens": 4627037.0,
      "step": 2510
    },
    {
      "entropy": 5.9435793399810795,
      "epoch": 0.21130014702793531,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998173019067127,
      "loss": 5.9898,
      "mean_token_accuracy": 0.13374351039528848,
      "num_tokens": 4637393.0,
      "step": 2515
    },
    {
      "entropy": 5.990574741363526,
      "epoch": 0.21172022684310018,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998160933542633,
      "loss": 6.0354,
      "mean_token_accuracy": 0.1225339263677597,
      "num_tokens": 4646832.0,
      "step": 2520
    },
    {
      "entropy": 6.078363418579102,
      "epoch": 0.21214030665826508,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998148808193128,
      "loss": 6.0571,
      "mean_token_accuracy": 0.1361754283308983,
      "num_tokens": 4655719.0,
      "step": 2525
    },
    {
      "entropy": 6.027440595626831,
      "epoch": 0.21256038647342995,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998136643018823,
      "loss": 6.0247,
      "mean_token_accuracy": 0.13285491690039636,
      "num_tokens": 4665364.0,
      "step": 2530
    },
    {
      "entropy": 6.0071735858917235,
      "epoch": 0.21298046628859482,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998124438019935,
      "loss": 5.9795,
      "mean_token_accuracy": 0.13230021819472312,
      "num_tokens": 4674760.0,
      "step": 2535
    },
    {
      "entropy": 5.971972846984864,
      "epoch": 0.21340054610375972,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998112193196681,
      "loss": 5.9064,
      "mean_token_accuracy": 0.1363543339073658,
      "num_tokens": 4683900.0,
      "step": 2540
    },
    {
      "entropy": 5.9554856300354,
      "epoch": 0.2138206259189246,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998099908549277,
      "loss": 5.9628,
      "mean_token_accuracy": 0.12749610617756843,
      "num_tokens": 4693915.0,
      "step": 2545
    },
    {
      "entropy": 5.917505264282227,
      "epoch": 0.2142407057340895,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499808758407794,
      "loss": 5.8105,
      "mean_token_accuracy": 0.14476394206285476,
      "num_tokens": 4703102.0,
      "step": 2550
    },
    {
      "entropy": 6.020478820800781,
      "epoch": 0.21466078554925436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998075219782889,
      "loss": 6.0403,
      "mean_token_accuracy": 0.13109349682927132,
      "num_tokens": 4712925.0,
      "step": 2555
    },
    {
      "entropy": 6.050074434280395,
      "epoch": 0.21508086536441923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998062815664344,
      "loss": 5.9753,
      "mean_token_accuracy": 0.13155975714325904,
      "num_tokens": 4722641.0,
      "step": 2560
    },
    {
      "entropy": 5.953602123260498,
      "epoch": 0.21550094517958412,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998050371722524,
      "loss": 6.0504,
      "mean_token_accuracy": 0.12833617106080056,
      "num_tokens": 4732603.0,
      "step": 2565
    },
    {
      "entropy": 5.983808517456055,
      "epoch": 0.215921024994749,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998037887957649,
      "loss": 5.8814,
      "mean_token_accuracy": 0.1358911283314228,
      "num_tokens": 4742644.0,
      "step": 2570
    },
    {
      "entropy": 6.108869409561157,
      "epoch": 0.2163411048099139,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998025364369939,
      "loss": 6.2019,
      "mean_token_accuracy": 0.1304102584719658,
      "num_tokens": 4751482.0,
      "step": 2575
    },
    {
      "entropy": 6.196333599090576,
      "epoch": 0.21676118462507876,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998012800959619,
      "loss": 6.0606,
      "mean_token_accuracy": 0.13010098412632942,
      "num_tokens": 4760593.0,
      "step": 2580
    },
    {
      "entropy": 6.040443658828735,
      "epoch": 0.21718126444024366,
      "grad_norm": 1.25,
      "learning_rate": 0.0004998000197726909,
      "loss": 6.0456,
      "mean_token_accuracy": 0.13714693188667298,
      "num_tokens": 4769294.0,
      "step": 2585
    },
    {
      "entropy": 6.037139892578125,
      "epoch": 0.21760134425540853,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997987554672033,
      "loss": 5.9468,
      "mean_token_accuracy": 0.1388247735798359,
      "num_tokens": 4779239.0,
      "step": 2590
    },
    {
      "entropy": 6.028454113006592,
      "epoch": 0.2180214240705734,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997974871795215,
      "loss": 6.0394,
      "mean_token_accuracy": 0.1312641680240631,
      "num_tokens": 4788211.0,
      "step": 2595
    },
    {
      "entropy": 6.064166069030762,
      "epoch": 0.2184415038857383,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499796214909668,
      "loss": 6.0092,
      "mean_token_accuracy": 0.13773878663778305,
      "num_tokens": 4797921.0,
      "step": 2600
    },
    {
      "entropy": 6.0709045886993405,
      "epoch": 0.21886158370090317,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997949386576653,
      "loss": 6.0042,
      "mean_token_accuracy": 0.1314603127539158,
      "num_tokens": 4807772.0,
      "step": 2605
    },
    {
      "entropy": 5.99371829032898,
      "epoch": 0.21928166351606806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499793658423536,
      "loss": 6.0192,
      "mean_token_accuracy": 0.13455276489257811,
      "num_tokens": 4817999.0,
      "step": 2610
    },
    {
      "entropy": 6.059015130996704,
      "epoch": 0.21970174333123293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997923742073028,
      "loss": 5.9804,
      "mean_token_accuracy": 0.14437463730573655,
      "num_tokens": 4826679.0,
      "step": 2615
    },
    {
      "entropy": 5.943931245803833,
      "epoch": 0.2201218231463978,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997910860089884,
      "loss": 5.9832,
      "mean_token_accuracy": 0.13589627295732498,
      "num_tokens": 4834998.0,
      "step": 2620
    },
    {
      "entropy": 6.057482719421387,
      "epoch": 0.2205419029615627,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997897938286156,
      "loss": 5.9317,
      "mean_token_accuracy": 0.13643529042601585,
      "num_tokens": 4843635.0,
      "step": 2625
    },
    {
      "entropy": 6.018534517288208,
      "epoch": 0.22096198277672757,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997884976662075,
      "loss": 6.055,
      "mean_token_accuracy": 0.1327926956117153,
      "num_tokens": 4852027.0,
      "step": 2630
    },
    {
      "entropy": 6.1170580863952635,
      "epoch": 0.22138206259189247,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997871975217868,
      "loss": 5.9753,
      "mean_token_accuracy": 0.14027014896273612,
      "num_tokens": 4861244.0,
      "step": 2635
    },
    {
      "entropy": 5.881848526000977,
      "epoch": 0.22180214240705734,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997858933953768,
      "loss": 5.8911,
      "mean_token_accuracy": 0.13821944668889047,
      "num_tokens": 4869902.0,
      "step": 2640
    },
    {
      "entropy": 5.9110313892364506,
      "epoch": 0.2222222222222222,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997845852870004,
      "loss": 5.8706,
      "mean_token_accuracy": 0.1410742297768593,
      "num_tokens": 4878502.0,
      "step": 2645
    },
    {
      "entropy": 5.945323467254639,
      "epoch": 0.2226423020373871,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997832731966806,
      "loss": 5.9249,
      "mean_token_accuracy": 0.1411375291645527,
      "num_tokens": 4888348.0,
      "step": 2650
    },
    {
      "entropy": 5.97375717163086,
      "epoch": 0.22306238185255198,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997819571244411,
      "loss": 5.9955,
      "mean_token_accuracy": 0.13679953366518022,
      "num_tokens": 4897302.0,
      "step": 2655
    },
    {
      "entropy": 6.008918142318725,
      "epoch": 0.22348246166771688,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997806370703049,
      "loss": 6.0213,
      "mean_token_accuracy": 0.13776542693376542,
      "num_tokens": 4907078.0,
      "step": 2660
    },
    {
      "entropy": 5.947820472717285,
      "epoch": 0.22390254148288175,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997793130342954,
      "loss": 5.8494,
      "mean_token_accuracy": 0.13728704303503036,
      "num_tokens": 4917489.0,
      "step": 2665
    },
    {
      "entropy": 5.903385925292969,
      "epoch": 0.22432262129804661,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997779850164363,
      "loss": 5.9513,
      "mean_token_accuracy": 0.1372543305158615,
      "num_tokens": 4927073.0,
      "step": 2670
    },
    {
      "entropy": 6.074141645431519,
      "epoch": 0.2247427011132115,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997766530167508,
      "loss": 6.0449,
      "mean_token_accuracy": 0.13193764314055442,
      "num_tokens": 4935464.0,
      "step": 2675
    },
    {
      "entropy": 6.150123214721679,
      "epoch": 0.22516278092837638,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997753170352627,
      "loss": 6.1293,
      "mean_token_accuracy": 0.13081697300076484,
      "num_tokens": 4944718.0,
      "step": 2680
    },
    {
      "entropy": 6.070659351348877,
      "epoch": 0.22558286074354128,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997739770719955,
      "loss": 6.0077,
      "mean_token_accuracy": 0.13340400233864785,
      "num_tokens": 4954223.0,
      "step": 2685
    },
    {
      "entropy": 5.978194713592529,
      "epoch": 0.22600294055870615,
      "grad_norm": 1.125,
      "learning_rate": 0.000499772633126973,
      "loss": 6.0415,
      "mean_token_accuracy": 0.12924405336380004,
      "num_tokens": 4963371.0,
      "step": 2690
    },
    {
      "entropy": 5.975288677215576,
      "epoch": 0.22642302037387105,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997712852002192,
      "loss": 5.9086,
      "mean_token_accuracy": 0.1422348402440548,
      "num_tokens": 4972973.0,
      "step": 2695
    },
    {
      "entropy": 6.000183725357056,
      "epoch": 0.22684310018903592,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997699332917578,
      "loss": 6.141,
      "mean_token_accuracy": 0.12485837489366532,
      "num_tokens": 4982808.0,
      "step": 2700
    },
    {
      "entropy": 6.124313545227051,
      "epoch": 0.2272631800042008,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997685774016127,
      "loss": 6.0087,
      "mean_token_accuracy": 0.13304658830165864,
      "num_tokens": 4992427.0,
      "step": 2705
    },
    {
      "entropy": 6.108034229278564,
      "epoch": 0.22768325981936569,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499767217529808,
      "loss": 6.1924,
      "mean_token_accuracy": 0.12314376682043075,
      "num_tokens": 5003562.0,
      "step": 2710
    },
    {
      "entropy": 6.024952030181884,
      "epoch": 0.22810333963453056,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997658536763678,
      "loss": 5.8848,
      "mean_token_accuracy": 0.13965026810765266,
      "num_tokens": 5013429.0,
      "step": 2715
    },
    {
      "entropy": 6.031959342956543,
      "epoch": 0.22852341944969545,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997644858413163,
      "loss": 6.0173,
      "mean_token_accuracy": 0.1369493454694748,
      "num_tokens": 5022045.0,
      "step": 2720
    },
    {
      "entropy": 5.9200338363647464,
      "epoch": 0.22894349926486032,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997631140246775,
      "loss": 5.8591,
      "mean_token_accuracy": 0.14441144168376924,
      "num_tokens": 5032260.0,
      "step": 2725
    },
    {
      "entropy": 5.958108282089233,
      "epoch": 0.2293635790800252,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499761738226476,
      "loss": 5.9041,
      "mean_token_accuracy": 0.1375686287879944,
      "num_tokens": 5041688.0,
      "step": 2730
    },
    {
      "entropy": 5.965337133407592,
      "epoch": 0.2297836588951901,
      "grad_norm": 1.359375,
      "learning_rate": 0.000499760358446736,
      "loss": 6.0072,
      "mean_token_accuracy": 0.13241190686821938,
      "num_tokens": 5051005.0,
      "step": 2735
    },
    {
      "entropy": 6.077162218093872,
      "epoch": 0.23020373871035496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499758974685482,
      "loss": 5.9379,
      "mean_token_accuracy": 0.13544429317116738,
      "num_tokens": 5060084.0,
      "step": 2740
    },
    {
      "entropy": 5.981188869476318,
      "epoch": 0.23062381852551986,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997575869427385,
      "loss": 5.953,
      "mean_token_accuracy": 0.13910676240921022,
      "num_tokens": 5069081.0,
      "step": 2745
    },
    {
      "entropy": 5.978755378723145,
      "epoch": 0.23104389834068473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049975619521853,
      "loss": 5.9429,
      "mean_token_accuracy": 0.13454415425658225,
      "num_tokens": 5078597.0,
      "step": 2750
    },
    {
      "entropy": 5.911752319335937,
      "epoch": 0.2314639781558496,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997547995128814,
      "loss": 5.9829,
      "mean_token_accuracy": 0.13646793067455293,
      "num_tokens": 5087607.0,
      "step": 2755
    },
    {
      "entropy": 6.035622882843017,
      "epoch": 0.2318840579710145,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004997533998258171,
      "loss": 5.9832,
      "mean_token_accuracy": 0.13701630160212516,
      "num_tokens": 5097412.0,
      "step": 2760
    },
    {
      "entropy": 6.060332536697388,
      "epoch": 0.23230413778617937,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997519961573622,
      "loss": 6.0518,
      "mean_token_accuracy": 0.1287323147058487,
      "num_tokens": 5105817.0,
      "step": 2765
    },
    {
      "entropy": 6.132694864273072,
      "epoch": 0.23272421760134426,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004997505885075414,
      "loss": 6.0843,
      "mean_token_accuracy": 0.13087237104773522,
      "num_tokens": 5114958.0,
      "step": 2770
    },
    {
      "entropy": 6.002990245819092,
      "epoch": 0.23314429741650913,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997491768763795,
      "loss": 6.0022,
      "mean_token_accuracy": 0.13458002656698226,
      "num_tokens": 5123728.0,
      "step": 2775
    },
    {
      "entropy": 6.000336790084839,
      "epoch": 0.23356437723167403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997477612639018,
      "loss": 6.0532,
      "mean_token_accuracy": 0.12724062129855157,
      "num_tokens": 5134099.0,
      "step": 2780
    },
    {
      "entropy": 6.150645542144775,
      "epoch": 0.2339844570468389,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997463416701332,
      "loss": 6.0567,
      "mean_token_accuracy": 0.12823428884148597,
      "num_tokens": 5142934.0,
      "step": 2785
    },
    {
      "entropy": 5.973050594329834,
      "epoch": 0.23440453686200377,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997449180950989,
      "loss": 5.9005,
      "mean_token_accuracy": 0.15188876092433928,
      "num_tokens": 5151835.0,
      "step": 2790
    },
    {
      "entropy": 5.945583391189575,
      "epoch": 0.23482461667716867,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997434905388241,
      "loss": 5.9533,
      "mean_token_accuracy": 0.14123927503824235,
      "num_tokens": 5161136.0,
      "step": 2795
    },
    {
      "entropy": 5.966424179077149,
      "epoch": 0.23524469649233354,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499742059001334,
      "loss": 5.887,
      "mean_token_accuracy": 0.141261176019907,
      "num_tokens": 5170741.0,
      "step": 2800
    },
    {
      "entropy": 5.931414556503296,
      "epoch": 0.23566477630749844,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997406234826541,
      "loss": 5.9226,
      "mean_token_accuracy": 0.14311096221208572,
      "num_tokens": 5180549.0,
      "step": 2805
    },
    {
      "entropy": 5.932320833206177,
      "epoch": 0.2360848561226633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997391839828098,
      "loss": 5.8876,
      "mean_token_accuracy": 0.14145613387227057,
      "num_tokens": 5189486.0,
      "step": 2810
    },
    {
      "entropy": 5.9496715545654295,
      "epoch": 0.23650493593782818,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997377405018266,
      "loss": 5.9643,
      "mean_token_accuracy": 0.1311965249478817,
      "num_tokens": 5198525.0,
      "step": 2815
    },
    {
      "entropy": 6.014245939254761,
      "epoch": 0.23692501575299307,
      "grad_norm": 1.109375,
      "learning_rate": 0.00049973629303973,
      "loss": 6.0158,
      "mean_token_accuracy": 0.13539923653006553,
      "num_tokens": 5207124.0,
      "step": 2820
    },
    {
      "entropy": 5.944891500473022,
      "epoch": 0.23734509556815794,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997348415965457,
      "loss": 5.85,
      "mean_token_accuracy": 0.1407323271036148,
      "num_tokens": 5216529.0,
      "step": 2825
    },
    {
      "entropy": 5.993428897857666,
      "epoch": 0.23776517538332284,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997333861722995,
      "loss": 5.9831,
      "mean_token_accuracy": 0.138421493768692,
      "num_tokens": 5225796.0,
      "step": 2830
    },
    {
      "entropy": 6.012037515640259,
      "epoch": 0.2381852551984877,
      "grad_norm": 1.234375,
      "learning_rate": 0.000499731926767017,
      "loss": 6.0269,
      "mean_token_accuracy": 0.13556732088327408,
      "num_tokens": 5233876.0,
      "step": 2835
    },
    {
      "entropy": 5.9344642639160154,
      "epoch": 0.23860533501365258,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997304633807242,
      "loss": 6.0019,
      "mean_token_accuracy": 0.12836523801088334,
      "num_tokens": 5244782.0,
      "step": 2840
    },
    {
      "entropy": 5.987623119354248,
      "epoch": 0.23902541482881748,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997289960134468,
      "loss": 5.9579,
      "mean_token_accuracy": 0.1335374064743519,
      "num_tokens": 5253453.0,
      "step": 2845
    },
    {
      "entropy": 5.978561115264893,
      "epoch": 0.23944549464398235,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997275246652111,
      "loss": 5.9635,
      "mean_token_accuracy": 0.14111838340759278,
      "num_tokens": 5262355.0,
      "step": 2850
    },
    {
      "entropy": 5.952275371551513,
      "epoch": 0.23986557445914725,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499726049336043,
      "loss": 5.8897,
      "mean_token_accuracy": 0.1419723652303219,
      "num_tokens": 5271959.0,
      "step": 2855
    },
    {
      "entropy": 6.006475210189819,
      "epoch": 0.24028565427431212,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997245700259686,
      "loss": 5.9216,
      "mean_token_accuracy": 0.1432231843471527,
      "num_tokens": 5281393.0,
      "step": 2860
    },
    {
      "entropy": 5.981584358215332,
      "epoch": 0.240705734089477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997230867350141,
      "loss": 6.0599,
      "mean_token_accuracy": 0.13176842033863068,
      "num_tokens": 5290979.0,
      "step": 2865
    },
    {
      "entropy": 6.072908639907837,
      "epoch": 0.24112581390464188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997215994632059,
      "loss": 5.9983,
      "mean_token_accuracy": 0.13976021558046342,
      "num_tokens": 5300263.0,
      "step": 2870
    },
    {
      "entropy": 5.99395980834961,
      "epoch": 0.24154589371980675,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997201082105704,
      "loss": 6.0192,
      "mean_token_accuracy": 0.13117292299866676,
      "num_tokens": 5309522.0,
      "step": 2875
    },
    {
      "entropy": 6.010568284988404,
      "epoch": 0.24196597353497165,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997186129771338,
      "loss": 6.0248,
      "mean_token_accuracy": 0.13696857616305352,
      "num_tokens": 5319770.0,
      "step": 2880
    },
    {
      "entropy": 6.136290264129639,
      "epoch": 0.24238605335013652,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997171137629226,
      "loss": 6.0295,
      "mean_token_accuracy": 0.1379177153110504,
      "num_tokens": 5328400.0,
      "step": 2885
    },
    {
      "entropy": 5.855829429626465,
      "epoch": 0.24280613316530142,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997156105679636,
      "loss": 5.8334,
      "mean_token_accuracy": 0.14593008533120155,
      "num_tokens": 5336338.0,
      "step": 2890
    },
    {
      "entropy": 5.898982286453247,
      "epoch": 0.2432262129804663,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997141033922832,
      "loss": 5.9375,
      "mean_token_accuracy": 0.13418934270739555,
      "num_tokens": 5345391.0,
      "step": 2895
    },
    {
      "entropy": 6.035576057434082,
      "epoch": 0.24364629279563116,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997125922359081,
      "loss": 5.9508,
      "mean_token_accuracy": 0.13234915360808372,
      "num_tokens": 5354709.0,
      "step": 2900
    },
    {
      "entropy": 6.013994407653809,
      "epoch": 0.24406637261079606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997110770988652,
      "loss": 5.8796,
      "mean_token_accuracy": 0.1399741917848587,
      "num_tokens": 5363738.0,
      "step": 2905
    },
    {
      "entropy": 5.9528099536895756,
      "epoch": 0.24448645242596093,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997095579811813,
      "loss": 6.0023,
      "mean_token_accuracy": 0.13593828454613685,
      "num_tokens": 5373583.0,
      "step": 2910
    },
    {
      "entropy": 6.03057632446289,
      "epoch": 0.24490653224112582,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997080348828833,
      "loss": 6.0477,
      "mean_token_accuracy": 0.1340787522494793,
      "num_tokens": 5383486.0,
      "step": 2915
    },
    {
      "entropy": 5.969451522827148,
      "epoch": 0.2453266120562907,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997065078039981,
      "loss": 5.9591,
      "mean_token_accuracy": 0.1328844092786312,
      "num_tokens": 5391974.0,
      "step": 2920
    },
    {
      "entropy": 6.031870555877686,
      "epoch": 0.24574669187145556,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997049767445529,
      "loss": 5.9995,
      "mean_token_accuracy": 0.13087670058012008,
      "num_tokens": 5400882.0,
      "step": 2925
    },
    {
      "entropy": 6.1388874530792235,
      "epoch": 0.24616677168662046,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997034417045746,
      "loss": 5.958,
      "mean_token_accuracy": 0.13255189657211303,
      "num_tokens": 5410538.0,
      "step": 2930
    },
    {
      "entropy": 5.891916513442993,
      "epoch": 0.24658685150178533,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997019026840907,
      "loss": 5.8523,
      "mean_token_accuracy": 0.1406748116016388,
      "num_tokens": 5419406.0,
      "step": 2935
    },
    {
      "entropy": 5.81290192604065,
      "epoch": 0.24700693131695023,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997003596831282,
      "loss": 5.9661,
      "mean_token_accuracy": 0.13368260413408278,
      "num_tokens": 5428817.0,
      "step": 2940
    },
    {
      "entropy": 6.030734586715698,
      "epoch": 0.2474270111321151,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996988127017145,
      "loss": 5.9967,
      "mean_token_accuracy": 0.1356920287013054,
      "num_tokens": 5438277.0,
      "step": 2945
    },
    {
      "entropy": 5.991678762435913,
      "epoch": 0.24784709094728,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996972617398772,
      "loss": 6.0095,
      "mean_token_accuracy": 0.13253712952136992,
      "num_tokens": 5447440.0,
      "step": 2950
    },
    {
      "entropy": 5.990732574462891,
      "epoch": 0.24826717076244487,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996957067976435,
      "loss": 5.9368,
      "mean_token_accuracy": 0.13873122334480287,
      "num_tokens": 5455988.0,
      "step": 2955
    },
    {
      "entropy": 6.013759565353394,
      "epoch": 0.24868725057760974,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996941478750411,
      "loss": 5.9498,
      "mean_token_accuracy": 0.13479771465063095,
      "num_tokens": 5464996.0,
      "step": 2960
    },
    {
      "entropy": 6.04653902053833,
      "epoch": 0.24910733039277463,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996925849720975,
      "loss": 6.0789,
      "mean_token_accuracy": 0.12909941822290422,
      "num_tokens": 5474174.0,
      "step": 2965
    },
    {
      "entropy": 6.1094592094421385,
      "epoch": 0.2495274102079395,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996910180888405,
      "loss": 5.9515,
      "mean_token_accuracy": 0.14010420814156532,
      "num_tokens": 5482838.0,
      "step": 2970
    },
    {
      "entropy": 5.933987855911255,
      "epoch": 0.2499474900231044,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996894472252977,
      "loss": 5.9796,
      "mean_token_accuracy": 0.13611237108707427,
      "num_tokens": 5491616.0,
      "step": 2975
    },
    {
      "entropy": 5.946248006820679,
      "epoch": 0.25036756983826924,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996878723814973,
      "loss": 5.9758,
      "mean_token_accuracy": 0.13201134279370308,
      "num_tokens": 5500942.0,
      "step": 2980
    },
    {
      "entropy": 6.011964797973633,
      "epoch": 0.25078764965343414,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996862935574667,
      "loss": 5.9171,
      "mean_token_accuracy": 0.13369757011532785,
      "num_tokens": 5510078.0,
      "step": 2985
    },
    {
      "entropy": 5.91240873336792,
      "epoch": 0.25120772946859904,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996847107532342,
      "loss": 5.9402,
      "mean_token_accuracy": 0.13632848113775253,
      "num_tokens": 5518924.0,
      "step": 2990
    },
    {
      "entropy": 6.007929849624634,
      "epoch": 0.25162780928376394,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996831239688277,
      "loss": 5.9543,
      "mean_token_accuracy": 0.13016238808631897,
      "num_tokens": 5527385.0,
      "step": 2995
    },
    {
      "entropy": 5.906063604354858,
      "epoch": 0.2520478890989288,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996815332042754,
      "loss": 5.8144,
      "mean_token_accuracy": 0.14219059348106383,
      "num_tokens": 5536781.0,
      "step": 3000
    },
    {
      "epoch": 0.2520478890989288,
      "eval_entropy": 5.731865753634434,
      "eval_loss": 5.98077917098999,
      "eval_mean_token_accuracy": 0.14166069063261308,
      "eval_num_tokens": 5536781.0,
      "eval_runtime": 27.289,
      "eval_samples_per_second": 1369.272,
      "eval_steps_per_second": 171.168,
      "step": 3000
    },
    {
      "entropy": 5.928550434112549,
      "epoch": 0.2524679689140937,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996799384596054,
      "loss": 6.0018,
      "mean_token_accuracy": 0.13845922499895097,
      "num_tokens": 5545893.0,
      "step": 3005
    },
    {
      "entropy": 6.047553873062133,
      "epoch": 0.2528880487292586,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996783397348461,
      "loss": 5.9557,
      "mean_token_accuracy": 0.13133809193968773,
      "num_tokens": 5555818.0,
      "step": 3010
    },
    {
      "entropy": 5.98364634513855,
      "epoch": 0.2533081285444234,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996767370300256,
      "loss": 5.9338,
      "mean_token_accuracy": 0.13684593588113786,
      "num_tokens": 5565331.0,
      "step": 3015
    },
    {
      "entropy": 6.016663599014282,
      "epoch": 0.2537282083595883,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996751303451724,
      "loss": 5.9132,
      "mean_token_accuracy": 0.1414400041103363,
      "num_tokens": 5574003.0,
      "step": 3020
    },
    {
      "entropy": 5.934581279754639,
      "epoch": 0.2541482881747532,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996735196803149,
      "loss": 5.8367,
      "mean_token_accuracy": 0.14427052065730095,
      "num_tokens": 5582517.0,
      "step": 3025
    },
    {
      "entropy": 5.937156009674072,
      "epoch": 0.2545683679899181,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996719050354818,
      "loss": 6.0272,
      "mean_token_accuracy": 0.1350693427026272,
      "num_tokens": 5591952.0,
      "step": 3030
    },
    {
      "entropy": 5.999459314346313,
      "epoch": 0.25498844780508295,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996702864107015,
      "loss": 5.9271,
      "mean_token_accuracy": 0.1392418310046196,
      "num_tokens": 5601460.0,
      "step": 3035
    },
    {
      "entropy": 6.104486131668091,
      "epoch": 0.25540852762024785,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996686638060028,
      "loss": 6.05,
      "mean_token_accuracy": 0.13465244546532631,
      "num_tokens": 5610776.0,
      "step": 3040
    },
    {
      "entropy": 5.98651351928711,
      "epoch": 0.25582860743541275,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004996670372214144,
      "loss": 5.9593,
      "mean_token_accuracy": 0.1381534829735756,
      "num_tokens": 5619627.0,
      "step": 3045
    },
    {
      "entropy": 5.846577882766724,
      "epoch": 0.2562486872505776,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996654066569651,
      "loss": 5.8254,
      "mean_token_accuracy": 0.1441572315990925,
      "num_tokens": 5628969.0,
      "step": 3050
    },
    {
      "entropy": 5.9285993576049805,
      "epoch": 0.2566687670657425,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996637721126839,
      "loss": 5.8895,
      "mean_token_accuracy": 0.13785991445183754,
      "num_tokens": 5638629.0,
      "step": 3055
    },
    {
      "entropy": 6.003174924850464,
      "epoch": 0.2570888468809074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996621335885996,
      "loss": 5.9755,
      "mean_token_accuracy": 0.13731449097394943,
      "num_tokens": 5647571.0,
      "step": 3060
    },
    {
      "entropy": 5.987164306640625,
      "epoch": 0.2575089266960722,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004996604910847413,
      "loss": 5.8754,
      "mean_token_accuracy": 0.14926859214901925,
      "num_tokens": 5656709.0,
      "step": 3065
    },
    {
      "entropy": 5.953637361526489,
      "epoch": 0.2579290065112371,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499658844601138,
      "loss": 6.0688,
      "mean_token_accuracy": 0.1354634039103985,
      "num_tokens": 5665714.0,
      "step": 3070
    },
    {
      "entropy": 6.097519016265869,
      "epoch": 0.258349086326402,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499657194137819,
      "loss": 6.0234,
      "mean_token_accuracy": 0.13649424612522126,
      "num_tokens": 5675854.0,
      "step": 3075
    },
    {
      "entropy": 5.9704999923706055,
      "epoch": 0.2587691661415669,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996555396948136,
      "loss": 5.8448,
      "mean_token_accuracy": 0.13540665656328202,
      "num_tokens": 5685690.0,
      "step": 3080
    },
    {
      "entropy": 5.92389030456543,
      "epoch": 0.25918924595673176,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996538812721509,
      "loss": 5.8958,
      "mean_token_accuracy": 0.1428774431347847,
      "num_tokens": 5695766.0,
      "step": 3085
    },
    {
      "entropy": 5.918001127243042,
      "epoch": 0.25960932577189666,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004996522188698603,
      "loss": 5.9586,
      "mean_token_accuracy": 0.13920465260744094,
      "num_tokens": 5704365.0,
      "step": 3090
    },
    {
      "entropy": 6.103138256072998,
      "epoch": 0.26002940558706156,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996505524879714,
      "loss": 6.0694,
      "mean_token_accuracy": 0.13200636729598045,
      "num_tokens": 5713345.0,
      "step": 3095
    },
    {
      "entropy": 6.0067219734191895,
      "epoch": 0.2604494854022264,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996488821265137,
      "loss": 5.8544,
      "mean_token_accuracy": 0.1430236168205738,
      "num_tokens": 5722907.0,
      "step": 3100
    },
    {
      "entropy": 5.831529951095581,
      "epoch": 0.2608695652173913,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996472077855166,
      "loss": 5.898,
      "mean_token_accuracy": 0.14116744548082352,
      "num_tokens": 5731589.0,
      "step": 3105
    },
    {
      "entropy": 5.962000036239624,
      "epoch": 0.2612896450325562,
      "grad_norm": 1.125,
      "learning_rate": 0.00049964552946501,
      "loss": 5.9005,
      "mean_token_accuracy": 0.13415754735469818,
      "num_tokens": 5739922.0,
      "step": 3110
    },
    {
      "entropy": 5.892115211486816,
      "epoch": 0.2617097248477211,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996438471650235,
      "loss": 5.8122,
      "mean_token_accuracy": 0.14354836270213128,
      "num_tokens": 5749206.0,
      "step": 3115
    },
    {
      "entropy": 5.935215997695923,
      "epoch": 0.26212980466288593,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996421608855869,
      "loss": 5.8703,
      "mean_token_accuracy": 0.1430413119494915,
      "num_tokens": 5758803.0,
      "step": 3120
    },
    {
      "entropy": 5.921274280548095,
      "epoch": 0.26254988447805083,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996404706267301,
      "loss": 5.9525,
      "mean_token_accuracy": 0.1340932957828045,
      "num_tokens": 5768368.0,
      "step": 3125
    },
    {
      "entropy": 5.895799207687378,
      "epoch": 0.26296996429321573,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499638776388483,
      "loss": 5.8028,
      "mean_token_accuracy": 0.14530150592327118,
      "num_tokens": 5776707.0,
      "step": 3130
    },
    {
      "entropy": 5.956259107589721,
      "epoch": 0.26339004410838057,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996370781708757,
      "loss": 5.9872,
      "mean_token_accuracy": 0.13445577397942543,
      "num_tokens": 5787037.0,
      "step": 3135
    },
    {
      "entropy": 5.948802375793457,
      "epoch": 0.26381012392354547,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996353759739382,
      "loss": 5.9353,
      "mean_token_accuracy": 0.14012779742479325,
      "num_tokens": 5796630.0,
      "step": 3140
    },
    {
      "entropy": 5.918614721298217,
      "epoch": 0.26423020373871037,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004996336697977007,
      "loss": 5.9569,
      "mean_token_accuracy": 0.13389407992362976,
      "num_tokens": 5806402.0,
      "step": 3145
    },
    {
      "entropy": 5.899946784973144,
      "epoch": 0.2646502835538752,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996319596421933,
      "loss": 5.8948,
      "mean_token_accuracy": 0.14252272099256516,
      "num_tokens": 5815742.0,
      "step": 3150
    },
    {
      "entropy": 5.881059455871582,
      "epoch": 0.2650703633690401,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996302455074466,
      "loss": 5.9116,
      "mean_token_accuracy": 0.13981909155845643,
      "num_tokens": 5824915.0,
      "step": 3155
    },
    {
      "entropy": 6.034554386138916,
      "epoch": 0.265490443184205,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996285273934906,
      "loss": 5.9346,
      "mean_token_accuracy": 0.13784030005335807,
      "num_tokens": 5834978.0,
      "step": 3160
    },
    {
      "entropy": 6.000072288513183,
      "epoch": 0.2659105229993699,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499626805300356,
      "loss": 6.0896,
      "mean_token_accuracy": 0.13608634248375892,
      "num_tokens": 5845684.0,
      "step": 3165
    },
    {
      "entropy": 6.076979541778565,
      "epoch": 0.26633060281453474,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996250792280732,
      "loss": 5.9723,
      "mean_token_accuracy": 0.131855096668005,
      "num_tokens": 5854905.0,
      "step": 3170
    },
    {
      "entropy": 6.017704057693481,
      "epoch": 0.26675068262969964,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996233491766727,
      "loss": 5.995,
      "mean_token_accuracy": 0.1348758891224861,
      "num_tokens": 5863654.0,
      "step": 3175
    },
    {
      "entropy": 5.958924627304077,
      "epoch": 0.26717076244486454,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996216151461854,
      "loss": 5.9782,
      "mean_token_accuracy": 0.14263538494706154,
      "num_tokens": 5872442.0,
      "step": 3180
    },
    {
      "entropy": 6.007285785675049,
      "epoch": 0.2675908422600294,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996198771366417,
      "loss": 5.9003,
      "mean_token_accuracy": 0.14180114939808847,
      "num_tokens": 5882372.0,
      "step": 3185
    },
    {
      "entropy": 5.762058162689209,
      "epoch": 0.2680109220751943,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996181351480726,
      "loss": 5.7096,
      "mean_token_accuracy": 0.1481250509619713,
      "num_tokens": 5891113.0,
      "step": 3190
    },
    {
      "entropy": 5.856069707870484,
      "epoch": 0.2684310018903592,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996163891805089,
      "loss": 5.9546,
      "mean_token_accuracy": 0.14241180717945098,
      "num_tokens": 5899582.0,
      "step": 3195
    },
    {
      "entropy": 6.006877613067627,
      "epoch": 0.2688510817055241,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996146392339815,
      "loss": 5.903,
      "mean_token_accuracy": 0.13792204037308692,
      "num_tokens": 5908938.0,
      "step": 3200
    },
    {
      "entropy": 5.92903504371643,
      "epoch": 0.2692711615206889,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996128853085215,
      "loss": 5.873,
      "mean_token_accuracy": 0.14175159782171248,
      "num_tokens": 5918055.0,
      "step": 3205
    },
    {
      "entropy": 5.962112808227539,
      "epoch": 0.2696912413358538,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996111274041598,
      "loss": 5.8745,
      "mean_token_accuracy": 0.1343413420021534,
      "num_tokens": 5926744.0,
      "step": 3210
    },
    {
      "entropy": 5.949939441680908,
      "epoch": 0.2701113211510187,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996093655209277,
      "loss": 5.9643,
      "mean_token_accuracy": 0.13674649894237517,
      "num_tokens": 5936521.0,
      "step": 3215
    },
    {
      "entropy": 6.075492525100708,
      "epoch": 0.27053140096618356,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004996075996588563,
      "loss": 6.0296,
      "mean_token_accuracy": 0.1298151694238186,
      "num_tokens": 5945010.0,
      "step": 3220
    },
    {
      "entropy": 5.941525030136108,
      "epoch": 0.27095148078134845,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499605829817977,
      "loss": 5.9212,
      "mean_token_accuracy": 0.13907130137085916,
      "num_tokens": 5953766.0,
      "step": 3225
    },
    {
      "entropy": 5.937248849868775,
      "epoch": 0.27137156059651335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499604055998321,
      "loss": 5.8382,
      "mean_token_accuracy": 0.14228671863675119,
      "num_tokens": 5962168.0,
      "step": 3230
    },
    {
      "entropy": 5.887500286102295,
      "epoch": 0.2717916404116782,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996022781999198,
      "loss": 5.8689,
      "mean_token_accuracy": 0.14494396820664407,
      "num_tokens": 5971627.0,
      "step": 3235
    },
    {
      "entropy": 5.927862691879272,
      "epoch": 0.2722117202268431,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499600496422805,
      "loss": 5.949,
      "mean_token_accuracy": 0.13576936945319176,
      "num_tokens": 5981775.0,
      "step": 3240
    },
    {
      "entropy": 5.942725610733032,
      "epoch": 0.272631800042008,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000499598710667008,
      "loss": 5.8857,
      "mean_token_accuracy": 0.1394686594605446,
      "num_tokens": 5991097.0,
      "step": 3245
    },
    {
      "entropy": 5.9221861362457275,
      "epoch": 0.2730518798571729,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004995969209325604,
      "loss": 5.9369,
      "mean_token_accuracy": 0.1354317285120487,
      "num_tokens": 5999517.0,
      "step": 3250
    },
    {
      "entropy": 5.942779064178467,
      "epoch": 0.2734719596723377,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995951272194941,
      "loss": 5.9079,
      "mean_token_accuracy": 0.13176682814955712,
      "num_tokens": 6008545.0,
      "step": 3255
    },
    {
      "entropy": 6.00344705581665,
      "epoch": 0.2738920394875026,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004995933295278407,
      "loss": 5.8989,
      "mean_token_accuracy": 0.13847036063671112,
      "num_tokens": 6017366.0,
      "step": 3260
    },
    {
      "entropy": 5.900600910186768,
      "epoch": 0.2743121193026675,
      "grad_norm": 1.375,
      "learning_rate": 0.0004995915278576321,
      "loss": 5.843,
      "mean_token_accuracy": 0.14386921525001525,
      "num_tokens": 6025597.0,
      "step": 3265
    },
    {
      "entropy": 6.0085962295532225,
      "epoch": 0.27473219911783237,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004995897222089004,
      "loss": 5.9437,
      "mean_token_accuracy": 0.1424303874373436,
      "num_tokens": 6034239.0,
      "step": 3270
    },
    {
      "entropy": 6.0732769012451175,
      "epoch": 0.27515227893299726,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995879125816772,
      "loss": 5.9769,
      "mean_token_accuracy": 0.13496886044740677,
      "num_tokens": 6043837.0,
      "step": 3275
    },
    {
      "entropy": 5.846703004837036,
      "epoch": 0.27557235874816216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995860989759949,
      "loss": 5.9195,
      "mean_token_accuracy": 0.14464289993047713,
      "num_tokens": 6053217.0,
      "step": 3280
    },
    {
      "entropy": 6.013448190689087,
      "epoch": 0.27599243856332706,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995842813918855,
      "loss": 5.9292,
      "mean_token_accuracy": 0.141995108127594,
      "num_tokens": 6061553.0,
      "step": 3285
    },
    {
      "entropy": 5.900888109207154,
      "epoch": 0.2764125183784919,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004995824598293812,
      "loss": 5.8195,
      "mean_token_accuracy": 0.14140584841370582,
      "num_tokens": 6070080.0,
      "step": 3290
    },
    {
      "entropy": 5.960742044448852,
      "epoch": 0.2768325981936568,
      "grad_norm": 1.25,
      "learning_rate": 0.0004995806342885142,
      "loss": 5.9653,
      "mean_token_accuracy": 0.14112535640597343,
      "num_tokens": 6078438.0,
      "step": 3295
    },
    {
      "entropy": 5.993828868865966,
      "epoch": 0.2772526780088217,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499578804769317,
      "loss": 5.9591,
      "mean_token_accuracy": 0.13578373640775682,
      "num_tokens": 6087794.0,
      "step": 3300
    },
    {
      "entropy": 6.002421045303345,
      "epoch": 0.27767275782398654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995769712718218,
      "loss": 5.9684,
      "mean_token_accuracy": 0.13969296216964722,
      "num_tokens": 6096709.0,
      "step": 3305
    },
    {
      "entropy": 5.897220087051392,
      "epoch": 0.27809283763915144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995751337960613,
      "loss": 5.9029,
      "mean_token_accuracy": 0.14022547677159308,
      "num_tokens": 6105866.0,
      "step": 3310
    },
    {
      "entropy": 6.0129883766174315,
      "epoch": 0.27851291745431633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995732923420679,
      "loss": 5.8481,
      "mean_token_accuracy": 0.1446751207113266,
      "num_tokens": 6114882.0,
      "step": 3315
    },
    {
      "entropy": 5.85595121383667,
      "epoch": 0.2789329972694812,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995714469098743,
      "loss": 5.8116,
      "mean_token_accuracy": 0.1394299313426018,
      "num_tokens": 6123978.0,
      "step": 3320
    },
    {
      "entropy": 5.841559362411499,
      "epoch": 0.2793530770846461,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499569597499513,
      "loss": 5.9636,
      "mean_token_accuracy": 0.1420348674058914,
      "num_tokens": 6133246.0,
      "step": 3325
    },
    {
      "entropy": 5.949489116668701,
      "epoch": 0.27977315689981097,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995677441110172,
      "loss": 5.8295,
      "mean_token_accuracy": 0.1390916422009468,
      "num_tokens": 6142865.0,
      "step": 3330
    },
    {
      "entropy": 5.958270597457886,
      "epoch": 0.28019323671497587,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995658867444192,
      "loss": 5.9185,
      "mean_token_accuracy": 0.13403844311833382,
      "num_tokens": 6152492.0,
      "step": 3335
    },
    {
      "entropy": 5.9247795104980465,
      "epoch": 0.2806133165301407,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995640253997523,
      "loss": 5.9182,
      "mean_token_accuracy": 0.13453099131584167,
      "num_tokens": 6161953.0,
      "step": 3340
    },
    {
      "entropy": 5.869791507720947,
      "epoch": 0.2810333963453056,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995621600770492,
      "loss": 5.7688,
      "mean_token_accuracy": 0.14543831422924997,
      "num_tokens": 6171467.0,
      "step": 3345
    },
    {
      "entropy": 5.874020433425903,
      "epoch": 0.2814534761604705,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995602907763431,
      "loss": 5.8552,
      "mean_token_accuracy": 0.1376187428832054,
      "num_tokens": 6180646.0,
      "step": 3350
    },
    {
      "entropy": 5.929383373260498,
      "epoch": 0.28187355597563535,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995584174976672,
      "loss": 5.8713,
      "mean_token_accuracy": 0.1346499465405941,
      "num_tokens": 6189832.0,
      "step": 3355
    },
    {
      "entropy": 5.927767086029053,
      "epoch": 0.28229363579080025,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995565402410544,
      "loss": 5.7654,
      "mean_token_accuracy": 0.14761848300695418,
      "num_tokens": 6198339.0,
      "step": 3360
    },
    {
      "entropy": 5.892761945724487,
      "epoch": 0.28271371560596514,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004995546590065383,
      "loss": 5.8571,
      "mean_token_accuracy": 0.1422846756875515,
      "num_tokens": 6207564.0,
      "step": 3365
    },
    {
      "entropy": 5.9255454540252686,
      "epoch": 0.28313379542113004,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995527737941518,
      "loss": 5.9347,
      "mean_token_accuracy": 0.13853738307952881,
      "num_tokens": 6216056.0,
      "step": 3370
    },
    {
      "entropy": 5.896576976776123,
      "epoch": 0.2835538752362949,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995508846039287,
      "loss": 5.8965,
      "mean_token_accuracy": 0.13626314997673034,
      "num_tokens": 6225573.0,
      "step": 3375
    },
    {
      "entropy": 6.006501722335815,
      "epoch": 0.2839739550514598,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004995489914359023,
      "loss": 6.0096,
      "mean_token_accuracy": 0.13380790427327155,
      "num_tokens": 6235057.0,
      "step": 3380
    },
    {
      "entropy": 6.008948469161988,
      "epoch": 0.2843940348666247,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004995470942901061,
      "loss": 5.9217,
      "mean_token_accuracy": 0.13976462185382843,
      "num_tokens": 6244164.0,
      "step": 3385
    },
    {
      "entropy": 6.019155550003052,
      "epoch": 0.2848141146817895,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004995451931665738,
      "loss": 5.9344,
      "mean_token_accuracy": 0.13991687223315238,
      "num_tokens": 6253095.0,
      "step": 3390
    },
    {
      "entropy": 5.957593202590942,
      "epoch": 0.2852341944969544,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499543288065339,
      "loss": 5.8639,
      "mean_token_accuracy": 0.13952580690383912,
      "num_tokens": 6261134.0,
      "step": 3395
    },
    {
      "entropy": 5.833978176116943,
      "epoch": 0.2856542743121193,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004995413789864354,
      "loss": 5.879,
      "mean_token_accuracy": 0.14211501479148864,
      "num_tokens": 6270384.0,
      "step": 3400
    },
    {
      "entropy": 5.952889156341553,
      "epoch": 0.28607435412728416,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995394659298971,
      "loss": 5.8153,
      "mean_token_accuracy": 0.14977512061595916,
      "num_tokens": 6279702.0,
      "step": 3405
    },
    {
      "entropy": 5.889900207519531,
      "epoch": 0.28649443394244906,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995375488957576,
      "loss": 5.8612,
      "mean_token_accuracy": 0.13872572034597397,
      "num_tokens": 6288297.0,
      "step": 3410
    },
    {
      "entropy": 5.9331536293029785,
      "epoch": 0.28691451375761395,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499535627884051,
      "loss": 5.9516,
      "mean_token_accuracy": 0.13464640453457832,
      "num_tokens": 6297288.0,
      "step": 3415
    },
    {
      "entropy": 6.120673799514771,
      "epoch": 0.28733459357277885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995337028948115,
      "loss": 5.974,
      "mean_token_accuracy": 0.13250542506575586,
      "num_tokens": 6306719.0,
      "step": 3420
    },
    {
      "entropy": 5.837868595123291,
      "epoch": 0.2877546733879437,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995317739280731,
      "loss": 5.8018,
      "mean_token_accuracy": 0.14693671017885207,
      "num_tokens": 6316639.0,
      "step": 3425
    },
    {
      "entropy": 5.928999614715576,
      "epoch": 0.2881747532031086,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995298409838699,
      "loss": 5.9251,
      "mean_token_accuracy": 0.1393180750310421,
      "num_tokens": 6326879.0,
      "step": 3430
    },
    {
      "entropy": 5.877130842208862,
      "epoch": 0.2885948330182735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499527904062236,
      "loss": 5.8293,
      "mean_token_accuracy": 0.14481945633888244,
      "num_tokens": 6335729.0,
      "step": 3435
    },
    {
      "entropy": 5.915560340881347,
      "epoch": 0.28901491283343833,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995259631632061,
      "loss": 5.8973,
      "mean_token_accuracy": 0.13230996280908586,
      "num_tokens": 6345154.0,
      "step": 3440
    },
    {
      "entropy": 5.962369394302368,
      "epoch": 0.28943499264860323,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995240182868143,
      "loss": 5.8479,
      "mean_token_accuracy": 0.14117665588855743,
      "num_tokens": 6354309.0,
      "step": 3445
    },
    {
      "entropy": 5.844361209869385,
      "epoch": 0.2898550724637681,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995220694330951,
      "loss": 5.8255,
      "mean_token_accuracy": 0.14228973686695098,
      "num_tokens": 6363389.0,
      "step": 3450
    },
    {
      "entropy": 5.88813967704773,
      "epoch": 0.290275152278933,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995201166020832,
      "loss": 5.8844,
      "mean_token_accuracy": 0.13614363521337508,
      "num_tokens": 6372475.0,
      "step": 3455
    },
    {
      "entropy": 5.972552013397217,
      "epoch": 0.29069523209409787,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499518159793813,
      "loss": 5.8329,
      "mean_token_accuracy": 0.14460824504494668,
      "num_tokens": 6380906.0,
      "step": 3460
    },
    {
      "entropy": 5.866478300094604,
      "epoch": 0.29111531190926276,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499516199008319,
      "loss": 5.8563,
      "mean_token_accuracy": 0.14013876989483834,
      "num_tokens": 6390085.0,
      "step": 3465
    },
    {
      "entropy": 5.941966962814331,
      "epoch": 0.29153539172442766,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004995142342456364,
      "loss": 5.9125,
      "mean_token_accuracy": 0.13554606810212136,
      "num_tokens": 6399441.0,
      "step": 3470
    },
    {
      "entropy": 6.016114854812622,
      "epoch": 0.2919554715395925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995122655057997,
      "loss": 5.9876,
      "mean_token_accuracy": 0.13846278935670853,
      "num_tokens": 6408995.0,
      "step": 3475
    },
    {
      "entropy": 5.820221567153931,
      "epoch": 0.2923755513547574,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004995102927888437,
      "loss": 5.7443,
      "mean_token_accuracy": 0.1459552101790905,
      "num_tokens": 6418080.0,
      "step": 3480
    },
    {
      "entropy": 5.911068105697632,
      "epoch": 0.2927956311699223,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004995083160948036,
      "loss": 5.9075,
      "mean_token_accuracy": 0.13615152686834336,
      "num_tokens": 6426732.0,
      "step": 3485
    },
    {
      "entropy": 5.973341417312622,
      "epoch": 0.29321571098508714,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995063354237141,
      "loss": 5.9199,
      "mean_token_accuracy": 0.14315774142742158,
      "num_tokens": 6435957.0,
      "step": 3490
    },
    {
      "entropy": 5.914236402511596,
      "epoch": 0.29363579080025204,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004995043507756107,
      "loss": 5.864,
      "mean_token_accuracy": 0.13712269440293312,
      "num_tokens": 6445642.0,
      "step": 3495
    },
    {
      "entropy": 5.931887197494507,
      "epoch": 0.29405587061541694,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004995023621505282,
      "loss": 5.906,
      "mean_token_accuracy": 0.1387566529214382,
      "num_tokens": 6454664.0,
      "step": 3500
    },
    {
      "entropy": 5.8483740329742435,
      "epoch": 0.29447595043058183,
      "grad_norm": 1.234375,
      "learning_rate": 0.000499500369548502,
      "loss": 5.8204,
      "mean_token_accuracy": 0.13878127932548523,
      "num_tokens": 6463224.0,
      "step": 3505
    },
    {
      "entropy": 6.101959562301635,
      "epoch": 0.2948960302457467,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994983729695674,
      "loss": 6.0552,
      "mean_token_accuracy": 0.13270595893263817,
      "num_tokens": 6473112.0,
      "step": 3510
    },
    {
      "entropy": 5.9638348579406735,
      "epoch": 0.2953161100609116,
      "grad_norm": 1.375,
      "learning_rate": 0.0004994963724137595,
      "loss": 5.8923,
      "mean_token_accuracy": 0.14195917025208474,
      "num_tokens": 6482062.0,
      "step": 3515
    },
    {
      "entropy": 5.8792516708374025,
      "epoch": 0.29573618987607647,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004994943678811142,
      "loss": 5.8699,
      "mean_token_accuracy": 0.13623269721865655,
      "num_tokens": 6490568.0,
      "step": 3520
    },
    {
      "entropy": 5.920031452178955,
      "epoch": 0.2961562696912413,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994923593716667,
      "loss": 5.9294,
      "mean_token_accuracy": 0.1400933049619198,
      "num_tokens": 6500815.0,
      "step": 3525
    },
    {
      "entropy": 5.929981470108032,
      "epoch": 0.2965763495064062,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994903468854527,
      "loss": 5.8119,
      "mean_token_accuracy": 0.1481688842177391,
      "num_tokens": 6509529.0,
      "step": 3530
    },
    {
      "entropy": 5.864474868774414,
      "epoch": 0.2969964293215711,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004994883304225077,
      "loss": 5.8729,
      "mean_token_accuracy": 0.13643766716122627,
      "num_tokens": 6517934.0,
      "step": 3535
    },
    {
      "entropy": 5.941485595703125,
      "epoch": 0.297416509136736,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994863099828675,
      "loss": 5.8357,
      "mean_token_accuracy": 0.13737112134695054,
      "num_tokens": 6526098.0,
      "step": 3540
    },
    {
      "entropy": 5.857544040679931,
      "epoch": 0.29783658895190085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499484285566568,
      "loss": 5.8676,
      "mean_token_accuracy": 0.13825918808579446,
      "num_tokens": 6535831.0,
      "step": 3545
    },
    {
      "entropy": 5.862282800674438,
      "epoch": 0.29825666876706575,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994822571736449,
      "loss": 5.7848,
      "mean_token_accuracy": 0.138202403485775,
      "num_tokens": 6545704.0,
      "step": 3550
    },
    {
      "entropy": 5.923550367355347,
      "epoch": 0.29867674858223064,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994802248041342,
      "loss": 5.8178,
      "mean_token_accuracy": 0.14043487086892129,
      "num_tokens": 6554423.0,
      "step": 3555
    },
    {
      "entropy": 5.915227842330933,
      "epoch": 0.2990968283973955,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000499478188458072,
      "loss": 5.8699,
      "mean_token_accuracy": 0.14151394963264466,
      "num_tokens": 6563989.0,
      "step": 3560
    },
    {
      "entropy": 5.978702878952026,
      "epoch": 0.2995169082125604,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004994761481354943,
      "loss": 6.0168,
      "mean_token_accuracy": 0.1372433789074421,
      "num_tokens": 6572745.0,
      "step": 3565
    },
    {
      "entropy": 6.085881900787354,
      "epoch": 0.2999369880277253,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994741038364371,
      "loss": 6.0023,
      "mean_token_accuracy": 0.1347330242395401,
      "num_tokens": 6581723.0,
      "step": 3570
    },
    {
      "entropy": 5.837055253982544,
      "epoch": 0.3003570678428901,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004994720555609369,
      "loss": 5.7255,
      "mean_token_accuracy": 0.14691844433546067,
      "num_tokens": 6590342.0,
      "step": 3575
    },
    {
      "entropy": 5.793679332733154,
      "epoch": 0.300777147658055,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004994700033090297,
      "loss": 5.7828,
      "mean_token_accuracy": 0.14932364374399185,
      "num_tokens": 6599206.0,
      "step": 3580
    },
    {
      "entropy": 6.014719247817993,
      "epoch": 0.3011972274732199,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499467947080752,
      "loss": 6.1117,
      "mean_token_accuracy": 0.1273707590997219,
      "num_tokens": 6608947.0,
      "step": 3585
    },
    {
      "entropy": 5.987322616577148,
      "epoch": 0.3016173072883848,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004994658868761402,
      "loss": 5.8883,
      "mean_token_accuracy": 0.14657592847943307,
      "num_tokens": 6618378.0,
      "step": 3590
    },
    {
      "entropy": 5.946252870559692,
      "epoch": 0.30203738710354966,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994638226952307,
      "loss": 5.9383,
      "mean_token_accuracy": 0.1343943029642105,
      "num_tokens": 6627527.0,
      "step": 3595
    },
    {
      "entropy": 5.973345470428467,
      "epoch": 0.30245746691871456,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994617545380604,
      "loss": 5.8799,
      "mean_token_accuracy": 0.1394343391060829,
      "num_tokens": 6636964.0,
      "step": 3600
    },
    {
      "entropy": 5.846103811264038,
      "epoch": 0.30287754673387945,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994596824046656,
      "loss": 5.8373,
      "mean_token_accuracy": 0.14053009524941446,
      "num_tokens": 6646074.0,
      "step": 3605
    },
    {
      "entropy": 5.920796251296997,
      "epoch": 0.3032976265490443,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499457606295083,
      "loss": 5.897,
      "mean_token_accuracy": 0.14094351455569268,
      "num_tokens": 6655027.0,
      "step": 3610
    },
    {
      "entropy": 5.804931497573852,
      "epoch": 0.3037177063642092,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004994555262093495,
      "loss": 5.6923,
      "mean_token_accuracy": 0.1544790118932724,
      "num_tokens": 6663747.0,
      "step": 3615
    },
    {
      "entropy": 6.0041478157043455,
      "epoch": 0.3041377861793741,
      "grad_norm": 1.28125,
      "learning_rate": 0.000499453442147502,
      "loss": 6.0113,
      "mean_token_accuracy": 0.13213684484362603,
      "num_tokens": 6672922.0,
      "step": 3620
    },
    {
      "entropy": 5.917838859558105,
      "epoch": 0.304557865994539,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994513541095773,
      "loss": 5.8314,
      "mean_token_accuracy": 0.14857635647058487,
      "num_tokens": 6682233.0,
      "step": 3625
    },
    {
      "entropy": 5.912606573104858,
      "epoch": 0.30497794580970383,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004994492620956126,
      "loss": 5.8901,
      "mean_token_accuracy": 0.1396655946969986,
      "num_tokens": 6691593.0,
      "step": 3630
    },
    {
      "entropy": 5.919918155670166,
      "epoch": 0.30539802562486873,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994471661056445,
      "loss": 5.8861,
      "mean_token_accuracy": 0.14072583466768265,
      "num_tokens": 6701318.0,
      "step": 3635
    },
    {
      "entropy": 6.009689807891846,
      "epoch": 0.3058181054400336,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994450661397106,
      "loss": 5.892,
      "mean_token_accuracy": 0.1426208183169365,
      "num_tokens": 6710059.0,
      "step": 3640
    },
    {
      "entropy": 6.047525215148926,
      "epoch": 0.30623818525519847,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499442962197848,
      "loss": 5.975,
      "mean_token_accuracy": 0.13458139076828957,
      "num_tokens": 6719811.0,
      "step": 3645
    },
    {
      "entropy": 5.868422555923462,
      "epoch": 0.30665826507036337,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004994408542800937,
      "loss": 5.8541,
      "mean_token_accuracy": 0.14217756688594818,
      "num_tokens": 6728789.0,
      "step": 3650
    },
    {
      "entropy": 5.868635082244873,
      "epoch": 0.30707834488552826,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994387423864855,
      "loss": 5.8459,
      "mean_token_accuracy": 0.14152047485113145,
      "num_tokens": 6737706.0,
      "step": 3655
    },
    {
      "entropy": 5.855863761901856,
      "epoch": 0.3074984247006931,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994366265170603,
      "loss": 5.7885,
      "mean_token_accuracy": 0.15381008386611938,
      "num_tokens": 6746861.0,
      "step": 3660
    },
    {
      "entropy": 5.958421134948731,
      "epoch": 0.307918504515858,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004994345066718558,
      "loss": 5.9914,
      "mean_token_accuracy": 0.13055020123720168,
      "num_tokens": 6755242.0,
      "step": 3665
    },
    {
      "entropy": 6.003214979171753,
      "epoch": 0.3083385843310229,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994323828509098,
      "loss": 5.9324,
      "mean_token_accuracy": 0.13040165677666665,
      "num_tokens": 6764549.0,
      "step": 3670
    },
    {
      "entropy": 5.916761636734009,
      "epoch": 0.3087586641461878,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004994302550542596,
      "loss": 5.9094,
      "mean_token_accuracy": 0.145879103243351,
      "num_tokens": 6774123.0,
      "step": 3675
    },
    {
      "entropy": 5.818877267837524,
      "epoch": 0.30917874396135264,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499428123281943,
      "loss": 5.6883,
      "mean_token_accuracy": 0.14885310381650924,
      "num_tokens": 6782922.0,
      "step": 3680
    },
    {
      "entropy": 5.8812672138214115,
      "epoch": 0.30959882377651754,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994259875339978,
      "loss": 5.9311,
      "mean_token_accuracy": 0.13911449760198594,
      "num_tokens": 6792042.0,
      "step": 3685
    },
    {
      "entropy": 6.0224377632141115,
      "epoch": 0.31001890359168244,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994238478104617,
      "loss": 5.9298,
      "mean_token_accuracy": 0.13841390311717988,
      "num_tokens": 6800994.0,
      "step": 3690
    },
    {
      "entropy": 5.922906446456909,
      "epoch": 0.3104389834068473,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994217041113727,
      "loss": 5.8716,
      "mean_token_accuracy": 0.14270309880375862,
      "num_tokens": 6809938.0,
      "step": 3695
    },
    {
      "entropy": 5.951845121383667,
      "epoch": 0.3108590632220122,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994195564367688,
      "loss": 5.9849,
      "mean_token_accuracy": 0.1360231176018715,
      "num_tokens": 6820289.0,
      "step": 3700
    },
    {
      "entropy": 5.994351577758789,
      "epoch": 0.3112791430371771,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004994174047866882,
      "loss": 5.8235,
      "mean_token_accuracy": 0.14149386510252954,
      "num_tokens": 6830068.0,
      "step": 3705
    },
    {
      "entropy": 5.771749830245971,
      "epoch": 0.3116992228523419,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004994152491611686,
      "loss": 5.8521,
      "mean_token_accuracy": 0.1437979094684124,
      "num_tokens": 6838591.0,
      "step": 3710
    },
    {
      "entropy": 5.865754890441894,
      "epoch": 0.3121193026675068,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994130895602485,
      "loss": 5.8204,
      "mean_token_accuracy": 0.13915161341428756,
      "num_tokens": 6847796.0,
      "step": 3715
    },
    {
      "entropy": 6.016102695465088,
      "epoch": 0.3125393824826717,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000499410925983966,
      "loss": 5.9097,
      "mean_token_accuracy": 0.14341016113758087,
      "num_tokens": 6856585.0,
      "step": 3720
    },
    {
      "entropy": 5.82035460472107,
      "epoch": 0.3129594622978366,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004994087584323596,
      "loss": 5.8224,
      "mean_token_accuracy": 0.1492151916027069,
      "num_tokens": 6865757.0,
      "step": 3725
    },
    {
      "entropy": 5.874684762954712,
      "epoch": 0.31337954211300145,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994065869054676,
      "loss": 5.8703,
      "mean_token_accuracy": 0.13963879272341728,
      "num_tokens": 6875371.0,
      "step": 3730
    },
    {
      "entropy": 5.973430156707764,
      "epoch": 0.31379962192816635,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004994044114033283,
      "loss": 5.9223,
      "mean_token_accuracy": 0.1317138932645321,
      "num_tokens": 6884050.0,
      "step": 3735
    },
    {
      "entropy": 5.9677238941192625,
      "epoch": 0.31421970174333125,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004994022319259806,
      "loss": 5.8772,
      "mean_token_accuracy": 0.14338692352175714,
      "num_tokens": 6893079.0,
      "step": 3740
    },
    {
      "entropy": 5.878354215621949,
      "epoch": 0.3146397815584961,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994000484734629,
      "loss": 5.9909,
      "mean_token_accuracy": 0.14075467139482498,
      "num_tokens": 6903100.0,
      "step": 3745
    },
    {
      "entropy": 5.928855514526367,
      "epoch": 0.315059861373661,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993978610458137,
      "loss": 5.837,
      "mean_token_accuracy": 0.14158818423748015,
      "num_tokens": 6912164.0,
      "step": 3750
    },
    {
      "entropy": 5.849039506912232,
      "epoch": 0.3154799411888259,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993956696430721,
      "loss": 5.8489,
      "mean_token_accuracy": 0.13852613866329194,
      "num_tokens": 6921183.0,
      "step": 3755
    },
    {
      "entropy": 5.947899580001831,
      "epoch": 0.3159000210039908,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993934742652768,
      "loss": 5.918,
      "mean_token_accuracy": 0.13974586948752404,
      "num_tokens": 6931325.0,
      "step": 3760
    },
    {
      "entropy": 5.930049276351928,
      "epoch": 0.3163201008191556,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993912749124665,
      "loss": 5.8111,
      "mean_token_accuracy": 0.14738105684518815,
      "num_tokens": 6940234.0,
      "step": 3765
    },
    {
      "entropy": 5.850810146331787,
      "epoch": 0.3167401806343205,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004993890715846804,
      "loss": 5.9136,
      "mean_token_accuracy": 0.1437673717737198,
      "num_tokens": 6949067.0,
      "step": 3770
    },
    {
      "entropy": 5.959416913986206,
      "epoch": 0.3171602604494854,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993868642819574,
      "loss": 5.8851,
      "mean_token_accuracy": 0.13802511468529702,
      "num_tokens": 6959085.0,
      "step": 3775
    },
    {
      "entropy": 5.905188941955567,
      "epoch": 0.31758034026465026,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004993846530043367,
      "loss": 5.9143,
      "mean_token_accuracy": 0.1347965881228447,
      "num_tokens": 6967392.0,
      "step": 3780
    },
    {
      "entropy": 5.903332805633545,
      "epoch": 0.31800042007981516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993824377518574,
      "loss": 5.8461,
      "mean_token_accuracy": 0.1488291099667549,
      "num_tokens": 6976369.0,
      "step": 3785
    },
    {
      "entropy": 5.983600234985351,
      "epoch": 0.31842049989498006,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993802185245587,
      "loss": 5.8623,
      "mean_token_accuracy": 0.14806569889187812,
      "num_tokens": 6985889.0,
      "step": 3790
    },
    {
      "entropy": 5.85070834159851,
      "epoch": 0.3188405797101449,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049937799532248,
      "loss": 5.8915,
      "mean_token_accuracy": 0.1322341412305832,
      "num_tokens": 6995396.0,
      "step": 3795
    },
    {
      "entropy": 6.083252477645874,
      "epoch": 0.3192606595253098,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993757681456607,
      "loss": 5.94,
      "mean_token_accuracy": 0.13651170060038567,
      "num_tokens": 7004666.0,
      "step": 3800
    },
    {
      "entropy": 5.98352370262146,
      "epoch": 0.3196807393404747,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993735369941401,
      "loss": 5.9741,
      "mean_token_accuracy": 0.13378495275974273,
      "num_tokens": 7014608.0,
      "step": 3805
    },
    {
      "entropy": 5.872314596176148,
      "epoch": 0.3201008191556396,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993713018679579,
      "loss": 5.8367,
      "mean_token_accuracy": 0.13734248280525208,
      "num_tokens": 7023671.0,
      "step": 3810
    },
    {
      "entropy": 5.887682008743286,
      "epoch": 0.32052089897080444,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993690627671536,
      "loss": 5.8912,
      "mean_token_accuracy": 0.13631067648530007,
      "num_tokens": 7033786.0,
      "step": 3815
    },
    {
      "entropy": 5.925025987625122,
      "epoch": 0.32094097878596933,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004993668196917669,
      "loss": 5.8026,
      "mean_token_accuracy": 0.14422772228717803,
      "num_tokens": 7042162.0,
      "step": 3820
    },
    {
      "entropy": 5.993115663528442,
      "epoch": 0.32136105860113423,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993645726418375,
      "loss": 5.946,
      "mean_token_accuracy": 0.13835487216711045,
      "num_tokens": 7051903.0,
      "step": 3825
    },
    {
      "entropy": 5.828875875473022,
      "epoch": 0.3217811384162991,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993623216174053,
      "loss": 5.7755,
      "mean_token_accuracy": 0.15191829651594163,
      "num_tokens": 7060229.0,
      "step": 3830
    },
    {
      "entropy": 5.8507789134979244,
      "epoch": 0.32220121823146397,
      "grad_norm": 1.21875,
      "learning_rate": 0.00049936006661851,
      "loss": 5.8708,
      "mean_token_accuracy": 0.13942239433526993,
      "num_tokens": 7069040.0,
      "step": 3835
    },
    {
      "entropy": 5.9059672355651855,
      "epoch": 0.32262129804662887,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004993578076451917,
      "loss": 5.755,
      "mean_token_accuracy": 0.1418701082468033,
      "num_tokens": 7078409.0,
      "step": 3840
    },
    {
      "entropy": 5.73867621421814,
      "epoch": 0.32304137786179377,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993555446974903,
      "loss": 5.841,
      "mean_token_accuracy": 0.13865280598402024,
      "num_tokens": 7087983.0,
      "step": 3845
    },
    {
      "entropy": 5.833849382400513,
      "epoch": 0.3234614576769586,
      "grad_norm": 1.515625,
      "learning_rate": 0.000499353277775446,
      "loss": 5.7975,
      "mean_token_accuracy": 0.14864109456539154,
      "num_tokens": 7097277.0,
      "step": 3850
    },
    {
      "entropy": 5.906425857543946,
      "epoch": 0.3238815374921235,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004993510068790989,
      "loss": 5.6902,
      "mean_token_accuracy": 0.15845565646886825,
      "num_tokens": 7105918.0,
      "step": 3855
    },
    {
      "entropy": 5.729353284835815,
      "epoch": 0.3243016173072884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993487320084892,
      "loss": 5.7663,
      "mean_token_accuracy": 0.15123388469219207,
      "num_tokens": 7115049.0,
      "step": 3860
    },
    {
      "entropy": 5.881364727020264,
      "epoch": 0.32472169712245325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993464531636573,
      "loss": 5.848,
      "mean_token_accuracy": 0.1397578552365303,
      "num_tokens": 7124862.0,
      "step": 3865
    },
    {
      "entropy": 5.853383731842041,
      "epoch": 0.32514177693761814,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993441703446435,
      "loss": 5.7449,
      "mean_token_accuracy": 0.14876592308282852,
      "num_tokens": 7133280.0,
      "step": 3870
    },
    {
      "entropy": 5.926945161819458,
      "epoch": 0.32556185675278304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993418835514882,
      "loss": 5.9468,
      "mean_token_accuracy": 0.13614385947585106,
      "num_tokens": 7142446.0,
      "step": 3875
    },
    {
      "entropy": 5.948226881027222,
      "epoch": 0.3259819365679479,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993395927842321,
      "loss": 5.8437,
      "mean_token_accuracy": 0.1408906787633896,
      "num_tokens": 7152143.0,
      "step": 3880
    },
    {
      "entropy": 5.944639015197754,
      "epoch": 0.3264020163831128,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004993372980429155,
      "loss": 5.9314,
      "mean_token_accuracy": 0.13912539780139924,
      "num_tokens": 7162046.0,
      "step": 3885
    },
    {
      "entropy": 5.925892972946167,
      "epoch": 0.3268220961982777,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004993349993275792,
      "loss": 5.8046,
      "mean_token_accuracy": 0.141475647687912,
      "num_tokens": 7171557.0,
      "step": 3890
    },
    {
      "entropy": 5.751782655715942,
      "epoch": 0.3272421760134426,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004993326966382639,
      "loss": 5.7152,
      "mean_token_accuracy": 0.14672790616750717,
      "num_tokens": 7180927.0,
      "step": 3895
    },
    {
      "entropy": 5.833934020996094,
      "epoch": 0.3276622558286074,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004993303899750104,
      "loss": 5.784,
      "mean_token_accuracy": 0.14524296969175338,
      "num_tokens": 7189552.0,
      "step": 3900
    },
    {
      "entropy": 5.929954242706299,
      "epoch": 0.3280823356437723,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004993280793378595,
      "loss": 5.8197,
      "mean_token_accuracy": 0.14072833955287933,
      "num_tokens": 7197857.0,
      "step": 3905
    },
    {
      "entropy": 5.886449480056763,
      "epoch": 0.3285024154589372,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993257647268522,
      "loss": 5.7906,
      "mean_token_accuracy": 0.15043871477246284,
      "num_tokens": 7206785.0,
      "step": 3910
    },
    {
      "entropy": 5.909309577941895,
      "epoch": 0.32892249527410206,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993234461420295,
      "loss": 5.8845,
      "mean_token_accuracy": 0.13852151483297348,
      "num_tokens": 7216360.0,
      "step": 3915
    },
    {
      "entropy": 5.842478799819946,
      "epoch": 0.32934257508926695,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004993211235834326,
      "loss": 5.6839,
      "mean_token_accuracy": 0.16384934931993483,
      "num_tokens": 7224890.0,
      "step": 3920
    },
    {
      "entropy": 5.758263874053955,
      "epoch": 0.32976265490443185,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004993187970511023,
      "loss": 5.735,
      "mean_token_accuracy": 0.16684153228998183,
      "num_tokens": 7234442.0,
      "step": 3925
    },
    {
      "entropy": 5.873496627807617,
      "epoch": 0.33018273471959675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004993164665450801,
      "loss": 5.8937,
      "mean_token_accuracy": 0.14423306286334991,
      "num_tokens": 7244023.0,
      "step": 3930
    },
    {
      "entropy": 5.877293682098388,
      "epoch": 0.3306028145347616,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993141320654072,
      "loss": 5.7546,
      "mean_token_accuracy": 0.14760554879903792,
      "num_tokens": 7253548.0,
      "step": 3935
    },
    {
      "entropy": 5.862598562240601,
      "epoch": 0.3310228943499265,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499311793612125,
      "loss": 5.8187,
      "mean_token_accuracy": 0.14365085512399672,
      "num_tokens": 7262962.0,
      "step": 3940
    },
    {
      "entropy": 5.9090534210205075,
      "epoch": 0.3314429741650914,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993094511852748,
      "loss": 5.8412,
      "mean_token_accuracy": 0.1450173959136009,
      "num_tokens": 7272234.0,
      "step": 3945
    },
    {
      "entropy": 5.899152231216431,
      "epoch": 0.33186305398025623,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993071047848983,
      "loss": 5.8162,
      "mean_token_accuracy": 0.1440458543598652,
      "num_tokens": 7281524.0,
      "step": 3950
    },
    {
      "entropy": 5.8174926280975345,
      "epoch": 0.3322831337954211,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004993047544110368,
      "loss": 5.7104,
      "mean_token_accuracy": 0.1493962250649929,
      "num_tokens": 7289601.0,
      "step": 3955
    },
    {
      "entropy": 5.727307987213135,
      "epoch": 0.332703213610586,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004993024000637321,
      "loss": 5.6805,
      "mean_token_accuracy": 0.15279524326324462,
      "num_tokens": 7298508.0,
      "step": 3960
    },
    {
      "entropy": 5.863152980804443,
      "epoch": 0.33312329342575087,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993000417430259,
      "loss": 5.899,
      "mean_token_accuracy": 0.14116615280508996,
      "num_tokens": 7309065.0,
      "step": 3965
    },
    {
      "entropy": 5.987276601791382,
      "epoch": 0.33354337324091576,
      "grad_norm": 1.140625,
      "learning_rate": 0.00049929767944896,
      "loss": 5.9279,
      "mean_token_accuracy": 0.14180676117539406,
      "num_tokens": 7319669.0,
      "step": 3970
    },
    {
      "entropy": 5.944264316558838,
      "epoch": 0.33396345305608066,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004992953131815761,
      "loss": 5.8675,
      "mean_token_accuracy": 0.14308214634656907,
      "num_tokens": 7328425.0,
      "step": 3975
    },
    {
      "entropy": 5.837700128555298,
      "epoch": 0.33438353287124556,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004992929429409164,
      "loss": 5.7527,
      "mean_token_accuracy": 0.1491951271891594,
      "num_tokens": 7337369.0,
      "step": 3980
    },
    {
      "entropy": 5.8214014053344725,
      "epoch": 0.3348036126864104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992905687270225,
      "loss": 5.8183,
      "mean_token_accuracy": 0.1459653303027153,
      "num_tokens": 7346829.0,
      "step": 3985
    },
    {
      "entropy": 5.947172021865844,
      "epoch": 0.3352236925015753,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004992881905399368,
      "loss": 5.8631,
      "mean_token_accuracy": 0.14260231107473373,
      "num_tokens": 7355976.0,
      "step": 3990
    },
    {
      "entropy": 5.880902576446533,
      "epoch": 0.3356437723167402,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004992858083797013,
      "loss": 5.8357,
      "mean_token_accuracy": 0.13988892063498498,
      "num_tokens": 7365210.0,
      "step": 3995
    },
    {
      "entropy": 5.887827444076538,
      "epoch": 0.33606385213190504,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004992834222463581,
      "loss": 5.8916,
      "mean_token_accuracy": 0.1366453118622303,
      "num_tokens": 7374175.0,
      "step": 4000
    },
    {
      "entropy": 5.9429937362670895,
      "epoch": 0.33648393194706994,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004992810321399496,
      "loss": 5.9143,
      "mean_token_accuracy": 0.13915260806679725,
      "num_tokens": 7383302.0,
      "step": 4005
    },
    {
      "entropy": 5.919287061691284,
      "epoch": 0.33690401176223483,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992786380605182,
      "loss": 5.8924,
      "mean_token_accuracy": 0.1387575164437294,
      "num_tokens": 7392746.0,
      "step": 4010
    },
    {
      "entropy": 5.856390810012817,
      "epoch": 0.33732409157739973,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004992762400081062,
      "loss": 5.7403,
      "mean_token_accuracy": 0.1431375488638878,
      "num_tokens": 7401604.0,
      "step": 4015
    },
    {
      "entropy": 5.85037088394165,
      "epoch": 0.3377441713925646,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004992738379827559,
      "loss": 5.8333,
      "mean_token_accuracy": 0.14066973105072975,
      "num_tokens": 7410594.0,
      "step": 4020
    },
    {
      "entropy": 5.8780660152435305,
      "epoch": 0.33816425120772947,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004992714319845101,
      "loss": 5.7395,
      "mean_token_accuracy": 0.15344746708869933,
      "num_tokens": 7418831.0,
      "step": 4025
    },
    {
      "entropy": 5.772813749313355,
      "epoch": 0.33858433102289437,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992690220134116,
      "loss": 5.7892,
      "mean_token_accuracy": 0.1462782494723797,
      "num_tokens": 7427731.0,
      "step": 4030
    },
    {
      "entropy": 5.971381282806396,
      "epoch": 0.3390044108380592,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004992666080695027,
      "loss": 5.9117,
      "mean_token_accuracy": 0.13779002502560617,
      "num_tokens": 7436447.0,
      "step": 4035
    },
    {
      "entropy": 5.921543741226197,
      "epoch": 0.3394244906532241,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004992641901528262,
      "loss": 5.811,
      "mean_token_accuracy": 0.14552520364522933,
      "num_tokens": 7445352.0,
      "step": 4040
    },
    {
      "entropy": 5.886599731445313,
      "epoch": 0.339844570468389,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992617682634252,
      "loss": 5.8513,
      "mean_token_accuracy": 0.14520843252539634,
      "num_tokens": 7454298.0,
      "step": 4045
    },
    {
      "entropy": 5.877497959136963,
      "epoch": 0.34026465028355385,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992593424013424,
      "loss": 5.8719,
      "mean_token_accuracy": 0.14371060207486153,
      "num_tokens": 7463543.0,
      "step": 4050
    },
    {
      "entropy": 5.891790342330933,
      "epoch": 0.34068473009871875,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004992569125666209,
      "loss": 5.8933,
      "mean_token_accuracy": 0.1396215297281742,
      "num_tokens": 7472701.0,
      "step": 4055
    },
    {
      "entropy": 6.004440689086914,
      "epoch": 0.34110480991388364,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992544787593037,
      "loss": 5.8639,
      "mean_token_accuracy": 0.13958193585276604,
      "num_tokens": 7481123.0,
      "step": 4060
    },
    {
      "entropy": 5.980106163024902,
      "epoch": 0.34152488972904854,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992520409794338,
      "loss": 5.9379,
      "mean_token_accuracy": 0.1414543256163597,
      "num_tokens": 7490439.0,
      "step": 4065
    },
    {
      "entropy": 5.900824117660522,
      "epoch": 0.3419449695442134,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992495992270544,
      "loss": 5.7958,
      "mean_token_accuracy": 0.14631813019514084,
      "num_tokens": 7499326.0,
      "step": 4070
    },
    {
      "entropy": 5.891779184341431,
      "epoch": 0.3423650493593783,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992471535022089,
      "loss": 5.8633,
      "mean_token_accuracy": 0.14064768627285956,
      "num_tokens": 7509407.0,
      "step": 4075
    },
    {
      "entropy": 5.933660221099854,
      "epoch": 0.3427851291745432,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992447038049405,
      "loss": 5.9161,
      "mean_token_accuracy": 0.14266983717679976,
      "num_tokens": 7518443.0,
      "step": 4080
    },
    {
      "entropy": 5.83581509590149,
      "epoch": 0.343205208989708,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004992422501352927,
      "loss": 5.7669,
      "mean_token_accuracy": 0.14985841438174247,
      "num_tokens": 7527609.0,
      "step": 4085
    },
    {
      "entropy": 5.924132442474365,
      "epoch": 0.3436252888048729,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004992397924933089,
      "loss": 5.8592,
      "mean_token_accuracy": 0.13954362720251084,
      "num_tokens": 7536890.0,
      "step": 4090
    },
    {
      "entropy": 5.917565202713012,
      "epoch": 0.3440453686200378,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992373308790325,
      "loss": 5.811,
      "mean_token_accuracy": 0.14937977269291877,
      "num_tokens": 7546509.0,
      "step": 4095
    },
    {
      "entropy": 5.8026703834533695,
      "epoch": 0.3444654484352027,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004992348652925074,
      "loss": 5.8363,
      "mean_token_accuracy": 0.14076031297445296,
      "num_tokens": 7555336.0,
      "step": 4100
    },
    {
      "entropy": 5.921528148651123,
      "epoch": 0.34488552825036756,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004992323957337771,
      "loss": 5.7996,
      "mean_token_accuracy": 0.14651144444942474,
      "num_tokens": 7565210.0,
      "step": 4105
    },
    {
      "entropy": 5.882741546630859,
      "epoch": 0.34530560806553245,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992299222028855,
      "loss": 5.892,
      "mean_token_accuracy": 0.1487639456987381,
      "num_tokens": 7574516.0,
      "step": 4110
    },
    {
      "entropy": 5.829581546783447,
      "epoch": 0.34572568788069735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992274446998761,
      "loss": 5.7261,
      "mean_token_accuracy": 0.15058319717645646,
      "num_tokens": 7583219.0,
      "step": 4115
    },
    {
      "entropy": 5.973264932632446,
      "epoch": 0.3461457676958622,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992249632247929,
      "loss": 5.9592,
      "mean_token_accuracy": 0.13624709472060204,
      "num_tokens": 7592050.0,
      "step": 4120
    },
    {
      "entropy": 5.926892328262329,
      "epoch": 0.3465658475110271,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992224777776802,
      "loss": 5.7982,
      "mean_token_accuracy": 0.14514639526605605,
      "num_tokens": 7600718.0,
      "step": 4125
    },
    {
      "entropy": 5.865804290771484,
      "epoch": 0.346985927326192,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004992199883585816,
      "loss": 5.8233,
      "mean_token_accuracy": 0.14894086718559266,
      "num_tokens": 7609191.0,
      "step": 4130
    },
    {
      "entropy": 5.919741106033325,
      "epoch": 0.34740600714135683,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004992174949675413,
      "loss": 5.8602,
      "mean_token_accuracy": 0.14337126538157463,
      "num_tokens": 7618509.0,
      "step": 4135
    },
    {
      "entropy": 5.859026050567627,
      "epoch": 0.34782608695652173,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004992149976046037,
      "loss": 5.7922,
      "mean_token_accuracy": 0.1422334760427475,
      "num_tokens": 7627851.0,
      "step": 4140
    },
    {
      "entropy": 5.82733941078186,
      "epoch": 0.3482461667716866,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004992124962698128,
      "loss": 5.8424,
      "mean_token_accuracy": 0.14138613119721413,
      "num_tokens": 7636748.0,
      "step": 4145
    },
    {
      "entropy": 5.9045960903167725,
      "epoch": 0.3486662465868515,
      "grad_norm": 1.296875,
      "learning_rate": 0.000499209990963213,
      "loss": 5.7782,
      "mean_token_accuracy": 0.14405173510313035,
      "num_tokens": 7645436.0,
      "step": 4150
    },
    {
      "entropy": 5.942763233184815,
      "epoch": 0.34908632640201637,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004992074816848487,
      "loss": 5.912,
      "mean_token_accuracy": 0.1420220673084259,
      "num_tokens": 7655414.0,
      "step": 4155
    },
    {
      "entropy": 5.806369590759277,
      "epoch": 0.34950640621718126,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004992049684347642,
      "loss": 5.6828,
      "mean_token_accuracy": 0.1479445531964302,
      "num_tokens": 7664295.0,
      "step": 4160
    },
    {
      "entropy": 5.8392486572265625,
      "epoch": 0.34992648603234616,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004992024512130042,
      "loss": 5.8256,
      "mean_token_accuracy": 0.14557857811450958,
      "num_tokens": 7673295.0,
      "step": 4165
    },
    {
      "entropy": 5.829631280899048,
      "epoch": 0.350346565847511,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991999300196132,
      "loss": 5.8282,
      "mean_token_accuracy": 0.14183629676699638,
      "num_tokens": 7682932.0,
      "step": 4170
    },
    {
      "entropy": 5.967007064819336,
      "epoch": 0.3507666456626759,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004991974048546359,
      "loss": 5.8389,
      "mean_token_accuracy": 0.1410795919597149,
      "num_tokens": 7692105.0,
      "step": 4175
    },
    {
      "entropy": 5.839569425582885,
      "epoch": 0.3511867254778408,
      "grad_norm": 1.625,
      "learning_rate": 0.000499194875718117,
      "loss": 5.8278,
      "mean_token_accuracy": 0.14473577067255974,
      "num_tokens": 7701294.0,
      "step": 4180
    },
    {
      "entropy": 5.905254983901978,
      "epoch": 0.3516068052930057,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004991923426101013,
      "loss": 5.8161,
      "mean_token_accuracy": 0.13896411135792733,
      "num_tokens": 7710964.0,
      "step": 4185
    },
    {
      "entropy": 5.987090635299682,
      "epoch": 0.35202688510817054,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004991898055306337,
      "loss": 5.9588,
      "mean_token_accuracy": 0.13461354821920396,
      "num_tokens": 7719938.0,
      "step": 4190
    },
    {
      "entropy": 5.921365547180176,
      "epoch": 0.35244696492333544,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991872644797591,
      "loss": 5.8516,
      "mean_token_accuracy": 0.13835739120841026,
      "num_tokens": 7729129.0,
      "step": 4195
    },
    {
      "entropy": 5.872485494613647,
      "epoch": 0.35286704473850034,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004991847194575226,
      "loss": 5.8744,
      "mean_token_accuracy": 0.135695618391037,
      "num_tokens": 7738506.0,
      "step": 4200
    },
    {
      "entropy": 5.9815671920776365,
      "epoch": 0.3532871245536652,
      "grad_norm": 1.375,
      "learning_rate": 0.0004991821704639693,
      "loss": 5.9756,
      "mean_token_accuracy": 0.1382329933345318,
      "num_tokens": 7749320.0,
      "step": 4205
    },
    {
      "entropy": 6.020629215240478,
      "epoch": 0.3537072043688301,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004991796174991443,
      "loss": 5.8318,
      "mean_token_accuracy": 0.14441338777542115,
      "num_tokens": 7758735.0,
      "step": 4210
    },
    {
      "entropy": 5.784712409973144,
      "epoch": 0.354127284183995,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991770605630927,
      "loss": 5.7838,
      "mean_token_accuracy": 0.14566663056612014,
      "num_tokens": 7767556.0,
      "step": 4215
    },
    {
      "entropy": 5.859303379058838,
      "epoch": 0.3545473639991598,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004991744996558599,
      "loss": 5.8127,
      "mean_token_accuracy": 0.14865762144327163,
      "num_tokens": 7776615.0,
      "step": 4220
    },
    {
      "entropy": 5.8661195755004885,
      "epoch": 0.3549674438143247,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004991719347774913,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1501207634806633,
      "num_tokens": 7785288.0,
      "step": 4225
    },
    {
      "entropy": 5.876074028015137,
      "epoch": 0.3553875236294896,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004991693659280324,
      "loss": 5.768,
      "mean_token_accuracy": 0.14568567126989365,
      "num_tokens": 7794381.0,
      "step": 4230
    },
    {
      "entropy": 5.80495433807373,
      "epoch": 0.3558076034446545,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004991667931075284,
      "loss": 5.7286,
      "mean_token_accuracy": 0.14424108862876892,
      "num_tokens": 7803265.0,
      "step": 4235
    },
    {
      "entropy": 5.866269779205322,
      "epoch": 0.35622768325981935,
      "grad_norm": 1.25,
      "learning_rate": 0.0004991642163160252,
      "loss": 5.8494,
      "mean_token_accuracy": 0.14268068224191666,
      "num_tokens": 7812445.0,
      "step": 4240
    },
    {
      "entropy": 5.940937805175781,
      "epoch": 0.35664776307498425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991616355535684,
      "loss": 5.8433,
      "mean_token_accuracy": 0.14912385791540145,
      "num_tokens": 7822073.0,
      "step": 4245
    },
    {
      "entropy": 5.912532901763916,
      "epoch": 0.35706784289014915,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004991590508202036,
      "loss": 5.8223,
      "mean_token_accuracy": 0.14509293287992478,
      "num_tokens": 7831193.0,
      "step": 4250
    },
    {
      "entropy": 5.9145135402679445,
      "epoch": 0.357487922705314,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004991564621159766,
      "loss": 5.8496,
      "mean_token_accuracy": 0.1429229497909546,
      "num_tokens": 7840311.0,
      "step": 4255
    },
    {
      "entropy": 5.863569116592407,
      "epoch": 0.3579080025204789,
      "grad_norm": 1.375,
      "learning_rate": 0.0004991538694409334,
      "loss": 5.8807,
      "mean_token_accuracy": 0.13731356635689734,
      "num_tokens": 7849622.0,
      "step": 4260
    },
    {
      "entropy": 5.866101455688477,
      "epoch": 0.3583280823356438,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004991512727951198,
      "loss": 5.8355,
      "mean_token_accuracy": 0.14157909527420998,
      "num_tokens": 7859494.0,
      "step": 4265
    },
    {
      "entropy": 6.04155158996582,
      "epoch": 0.3587481621508087,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991486721785818,
      "loss": 5.9463,
      "mean_token_accuracy": 0.13783199936151505,
      "num_tokens": 7868526.0,
      "step": 4270
    },
    {
      "entropy": 5.926672124862671,
      "epoch": 0.3591682419659735,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004991460675913655,
      "loss": 5.7679,
      "mean_token_accuracy": 0.145780511200428,
      "num_tokens": 7877631.0,
      "step": 4275
    },
    {
      "entropy": 5.883526086807251,
      "epoch": 0.3595883217811384,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499143459033517,
      "loss": 5.8215,
      "mean_token_accuracy": 0.14927242919802666,
      "num_tokens": 7886814.0,
      "step": 4280
    },
    {
      "entropy": 5.745771408081055,
      "epoch": 0.3600084015963033,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004991408465050825,
      "loss": 5.6432,
      "mean_token_accuracy": 0.1510879337787628,
      "num_tokens": 7896337.0,
      "step": 4285
    },
    {
      "entropy": 5.832517528533936,
      "epoch": 0.36042848141146816,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991382300061084,
      "loss": 5.9099,
      "mean_token_accuracy": 0.13439226746559144,
      "num_tokens": 7906071.0,
      "step": 4290
    },
    {
      "entropy": 5.962070560455322,
      "epoch": 0.36084856122663306,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004991356095366409,
      "loss": 5.8959,
      "mean_token_accuracy": 0.14356234297156334,
      "num_tokens": 7915003.0,
      "step": 4295
    },
    {
      "entropy": 5.901134014129639,
      "epoch": 0.36126864104179796,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004991329850967266,
      "loss": 5.7553,
      "mean_token_accuracy": 0.14801599234342575,
      "num_tokens": 7924408.0,
      "step": 4300
    },
    {
      "entropy": 5.79054274559021,
      "epoch": 0.3616887208569628,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004991303566864118,
      "loss": 5.7258,
      "mean_token_accuracy": 0.14961198195815087,
      "num_tokens": 7934717.0,
      "step": 4305
    },
    {
      "entropy": 5.805612707138062,
      "epoch": 0.3621088006721277,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004991277243057431,
      "loss": 5.7802,
      "mean_token_accuracy": 0.14127594009041786,
      "num_tokens": 7944278.0,
      "step": 4310
    },
    {
      "entropy": 5.852519369125366,
      "epoch": 0.3625288804872926,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004991250879547673,
      "loss": 5.8079,
      "mean_token_accuracy": 0.147132358700037,
      "num_tokens": 7953344.0,
      "step": 4315
    },
    {
      "entropy": 5.814604616165161,
      "epoch": 0.3629489603024575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991224476335309,
      "loss": 5.8228,
      "mean_token_accuracy": 0.14245626628398894,
      "num_tokens": 7962869.0,
      "step": 4320
    },
    {
      "entropy": 5.943966865539551,
      "epoch": 0.36336904011762233,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004991198033420807,
      "loss": 5.8144,
      "mean_token_accuracy": 0.14347591027617454,
      "num_tokens": 7971981.0,
      "step": 4325
    },
    {
      "entropy": 5.808311653137207,
      "epoch": 0.36378911993278723,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004991171550804636,
      "loss": 5.7932,
      "mean_token_accuracy": 0.14206481873989105,
      "num_tokens": 7980979.0,
      "step": 4330
    },
    {
      "entropy": 5.905320501327514,
      "epoch": 0.36420919974795213,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004991145028487266,
      "loss": 5.8545,
      "mean_token_accuracy": 0.14277459532022477,
      "num_tokens": 7989607.0,
      "step": 4335
    },
    {
      "entropy": 5.837810134887695,
      "epoch": 0.36462927956311697,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991118466469165,
      "loss": 5.6845,
      "mean_token_accuracy": 0.1493909515440464,
      "num_tokens": 7998356.0,
      "step": 4340
    },
    {
      "entropy": 5.851336050033569,
      "epoch": 0.36504935937828187,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004991091864750805,
      "loss": 5.7847,
      "mean_token_accuracy": 0.15024245530366898,
      "num_tokens": 8007596.0,
      "step": 4345
    },
    {
      "entropy": 5.901401472091675,
      "epoch": 0.36546943919344677,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004991065223332655,
      "loss": 5.8505,
      "mean_token_accuracy": 0.14087127447128295,
      "num_tokens": 8016493.0,
      "step": 4350
    },
    {
      "entropy": 5.870211553573609,
      "epoch": 0.36588951900861166,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004991038542215191,
      "loss": 5.8227,
      "mean_token_accuracy": 0.13674163743853568,
      "num_tokens": 8025867.0,
      "step": 4355
    },
    {
      "entropy": 5.875178384780884,
      "epoch": 0.3663095988237765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004991011821398882,
      "loss": 5.8524,
      "mean_token_accuracy": 0.14893437176942825,
      "num_tokens": 8036251.0,
      "step": 4360
    },
    {
      "entropy": 5.93429970741272,
      "epoch": 0.3667296786389414,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990985060884202,
      "loss": 5.7934,
      "mean_token_accuracy": 0.14601994007825853,
      "num_tokens": 8045647.0,
      "step": 4365
    },
    {
      "entropy": 5.862146234512329,
      "epoch": 0.3671497584541063,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004990958260671627,
      "loss": 5.8699,
      "mean_token_accuracy": 0.13680800497531892,
      "num_tokens": 8056025.0,
      "step": 4370
    },
    {
      "entropy": 5.831518173217773,
      "epoch": 0.36756983826927114,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004990931420761629,
      "loss": 5.801,
      "mean_token_accuracy": 0.14898931458592415,
      "num_tokens": 8065029.0,
      "step": 4375
    },
    {
      "entropy": 5.986507749557495,
      "epoch": 0.36798991808443604,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004990904541154685,
      "loss": 5.7557,
      "mean_token_accuracy": 0.1577958881855011,
      "num_tokens": 8073249.0,
      "step": 4380
    },
    {
      "entropy": 5.926363086700439,
      "epoch": 0.36840999789960094,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004990877621851271,
      "loss": 5.8958,
      "mean_token_accuracy": 0.14099107980728148,
      "num_tokens": 8082039.0,
      "step": 4385
    },
    {
      "entropy": 5.77527756690979,
      "epoch": 0.3688300777147658,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004990850662851863,
      "loss": 5.7314,
      "mean_token_accuracy": 0.14939506947994233,
      "num_tokens": 8090011.0,
      "step": 4390
    },
    {
      "entropy": 5.903356552124023,
      "epoch": 0.3692501575299307,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990823664156941,
      "loss": 5.8578,
      "mean_token_accuracy": 0.1503001019358635,
      "num_tokens": 8099934.0,
      "step": 4395
    },
    {
      "entropy": 5.969454669952393,
      "epoch": 0.3696702373450956,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004990796625766981,
      "loss": 5.8544,
      "mean_token_accuracy": 0.14106032997369766,
      "num_tokens": 8108969.0,
      "step": 4400
    },
    {
      "entropy": 5.791495323181152,
      "epoch": 0.3700903171602605,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990769547682462,
      "loss": 5.7606,
      "mean_token_accuracy": 0.14634729623794557,
      "num_tokens": 8117372.0,
      "step": 4405
    },
    {
      "entropy": 6.0099778175354,
      "epoch": 0.3705103969754253,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990742429903866,
      "loss": 5.9586,
      "mean_token_accuracy": 0.13980280980467796,
      "num_tokens": 8127108.0,
      "step": 4410
    },
    {
      "entropy": 5.983245754241944,
      "epoch": 0.3709304767905902,
      "grad_norm": 1.234375,
      "learning_rate": 0.000499071527243167,
      "loss": 5.9341,
      "mean_token_accuracy": 0.13901495784521103,
      "num_tokens": 8137392.0,
      "step": 4415
    },
    {
      "entropy": 5.866679716110229,
      "epoch": 0.3713505566057551,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004990688075266357,
      "loss": 5.7869,
      "mean_token_accuracy": 0.15010641515254974,
      "num_tokens": 8146257.0,
      "step": 4420
    },
    {
      "entropy": 5.818060064315796,
      "epoch": 0.37177063642091995,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004990660838408409,
      "loss": 5.7406,
      "mean_token_accuracy": 0.14406471997499465,
      "num_tokens": 8154952.0,
      "step": 4425
    },
    {
      "entropy": 5.896796941757202,
      "epoch": 0.37219071623608485,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004990633561858308,
      "loss": 5.8106,
      "mean_token_accuracy": 0.14199803844094278,
      "num_tokens": 8164365.0,
      "step": 4430
    },
    {
      "entropy": 5.897768831253051,
      "epoch": 0.37261079605124975,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004990606245616537,
      "loss": 5.806,
      "mean_token_accuracy": 0.14310891702771186,
      "num_tokens": 8172614.0,
      "step": 4435
    },
    {
      "entropy": 5.942156267166138,
      "epoch": 0.37303087586641465,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990578889683579,
      "loss": 5.8804,
      "mean_token_accuracy": 0.1399701401591301,
      "num_tokens": 8182445.0,
      "step": 4440
    },
    {
      "entropy": 5.876403188705444,
      "epoch": 0.3734509556815795,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990551494059921,
      "loss": 5.759,
      "mean_token_accuracy": 0.14562231674790382,
      "num_tokens": 8191871.0,
      "step": 4445
    },
    {
      "entropy": 5.910711050033569,
      "epoch": 0.3738710354967444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004990524058746047,
      "loss": 5.9163,
      "mean_token_accuracy": 0.14722812101244925,
      "num_tokens": 8200658.0,
      "step": 4450
    },
    {
      "entropy": 5.870158910751343,
      "epoch": 0.3742911153119093,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004990496583742443,
      "loss": 5.828,
      "mean_token_accuracy": 0.13948734179139138,
      "num_tokens": 8209776.0,
      "step": 4455
    },
    {
      "entropy": 5.8510631084442135,
      "epoch": 0.3747111951270741,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004990469069049596,
      "loss": 5.7655,
      "mean_token_accuracy": 0.14834618717432022,
      "num_tokens": 8219401.0,
      "step": 4460
    },
    {
      "entropy": 5.841431474685669,
      "epoch": 0.375131274942239,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004990441514667993,
      "loss": 5.8164,
      "mean_token_accuracy": 0.14826851561665536,
      "num_tokens": 8228762.0,
      "step": 4465
    },
    {
      "entropy": 5.943975448608398,
      "epoch": 0.3755513547574039,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004990413920598121,
      "loss": 5.8117,
      "mean_token_accuracy": 0.1469542607665062,
      "num_tokens": 8236612.0,
      "step": 4470
    },
    {
      "entropy": 5.879862689971924,
      "epoch": 0.37597143457256876,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004990386286840471,
      "loss": 5.831,
      "mean_token_accuracy": 0.14322375506162643,
      "num_tokens": 8245043.0,
      "step": 4475
    },
    {
      "entropy": 6.000154876708985,
      "epoch": 0.37639151438773366,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004990358613395532,
      "loss": 5.9178,
      "mean_token_accuracy": 0.14013071805238725,
      "num_tokens": 8255270.0,
      "step": 4480
    },
    {
      "entropy": 5.975326061248779,
      "epoch": 0.37681159420289856,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004990330900263792,
      "loss": 5.8795,
      "mean_token_accuracy": 0.13827874809503554,
      "num_tokens": 8264761.0,
      "step": 4485
    },
    {
      "entropy": 5.914210987091065,
      "epoch": 0.37723167401806346,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004990303147445745,
      "loss": 5.8254,
      "mean_token_accuracy": 0.14292607977986335,
      "num_tokens": 8274308.0,
      "step": 4490
    },
    {
      "entropy": 5.824473428726196,
      "epoch": 0.3776517538332283,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004990275354941881,
      "loss": 5.7135,
      "mean_token_accuracy": 0.15530410706996917,
      "num_tokens": 8283323.0,
      "step": 4495
    },
    {
      "entropy": 5.8834668636322025,
      "epoch": 0.3780718336483932,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990247522752694,
      "loss": 6.0456,
      "mean_token_accuracy": 0.12988803088665007,
      "num_tokens": 8293452.0,
      "step": 4500
    },
    {
      "entropy": 5.955636644363404,
      "epoch": 0.3784919134635581,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990219650878674,
      "loss": 5.7218,
      "mean_token_accuracy": 0.15242000669240952,
      "num_tokens": 8302941.0,
      "step": 4505
    },
    {
      "entropy": 5.847525358200073,
      "epoch": 0.37891199327872294,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004990191739320318,
      "loss": 5.7466,
      "mean_token_accuracy": 0.1486208386719227,
      "num_tokens": 8311811.0,
      "step": 4510
    },
    {
      "entropy": 5.740224170684814,
      "epoch": 0.37933207309388783,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990163788078117,
      "loss": 5.6461,
      "mean_token_accuracy": 0.15186312943696975,
      "num_tokens": 8321130.0,
      "step": 4515
    },
    {
      "entropy": 5.8464127540588375,
      "epoch": 0.37975215290905273,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990135797152569,
      "loss": 5.7849,
      "mean_token_accuracy": 0.1421021416783333,
      "num_tokens": 8330233.0,
      "step": 4520
    },
    {
      "entropy": 5.841711711883545,
      "epoch": 0.3801722327242176,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004990107766544169,
      "loss": 5.7542,
      "mean_token_accuracy": 0.14685214608907698,
      "num_tokens": 8338585.0,
      "step": 4525
    },
    {
      "entropy": 5.820804738998413,
      "epoch": 0.38059231253938247,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004990079696253413,
      "loss": 5.7772,
      "mean_token_accuracy": 0.15359208285808562,
      "num_tokens": 8346618.0,
      "step": 4530
    },
    {
      "entropy": 5.856412267684936,
      "epoch": 0.38101239235454737,
      "grad_norm": 1.625,
      "learning_rate": 0.0004990051586280799,
      "loss": 5.7538,
      "mean_token_accuracy": 0.14895081669092178,
      "num_tokens": 8356273.0,
      "step": 4535
    },
    {
      "entropy": 5.84465913772583,
      "epoch": 0.38143247216971227,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004990023436626824,
      "loss": 5.7524,
      "mean_token_accuracy": 0.15021177157759666,
      "num_tokens": 8366668.0,
      "step": 4540
    },
    {
      "entropy": 5.9355755805969235,
      "epoch": 0.3818525519848771,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004989995247291988,
      "loss": 5.8869,
      "mean_token_accuracy": 0.14657219648361205,
      "num_tokens": 8375610.0,
      "step": 4545
    },
    {
      "entropy": 5.892274522781372,
      "epoch": 0.382272631800042,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004989967018276789,
      "loss": 5.7508,
      "mean_token_accuracy": 0.14968783855438234,
      "num_tokens": 8384455.0,
      "step": 4550
    },
    {
      "entropy": 5.7833092212677,
      "epoch": 0.3826927116152069,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004989938749581727,
      "loss": 5.7971,
      "mean_token_accuracy": 0.14261699542403222,
      "num_tokens": 8393868.0,
      "step": 4555
    },
    {
      "entropy": 5.87280683517456,
      "epoch": 0.38311279143037175,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004989910441207305,
      "loss": 5.8159,
      "mean_token_accuracy": 0.142460335791111,
      "num_tokens": 8402916.0,
      "step": 4560
    },
    {
      "entropy": 5.866139030456543,
      "epoch": 0.38353287124553664,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004989882093154023,
      "loss": 5.7386,
      "mean_token_accuracy": 0.14703422486782075,
      "num_tokens": 8411649.0,
      "step": 4565
    },
    {
      "entropy": 5.867181587219238,
      "epoch": 0.38395295106070154,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004989853705422381,
      "loss": 5.8685,
      "mean_token_accuracy": 0.14120551571249962,
      "num_tokens": 8420393.0,
      "step": 4570
    },
    {
      "entropy": 5.828695583343506,
      "epoch": 0.38437303087586644,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004989825278012886,
      "loss": 5.7515,
      "mean_token_accuracy": 0.14682712703943251,
      "num_tokens": 8429404.0,
      "step": 4575
    },
    {
      "entropy": 5.838935279846192,
      "epoch": 0.3847931106910313,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000498979681092604,
      "loss": 5.7956,
      "mean_token_accuracy": 0.1419872298836708,
      "num_tokens": 8438299.0,
      "step": 4580
    },
    {
      "entropy": 5.8123194694519045,
      "epoch": 0.3852131905061962,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004989768304162345,
      "loss": 5.7252,
      "mean_token_accuracy": 0.14794419556856156,
      "num_tokens": 8447392.0,
      "step": 4585
    },
    {
      "entropy": 5.904961681365966,
      "epoch": 0.3856332703213611,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004989739757722308,
      "loss": 5.8303,
      "mean_token_accuracy": 0.14383739084005356,
      "num_tokens": 8456361.0,
      "step": 4590
    },
    {
      "entropy": 5.8575108528137205,
      "epoch": 0.3860533501365259,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004989711171606436,
      "loss": 5.7685,
      "mean_token_accuracy": 0.14821998178958892,
      "num_tokens": 8465548.0,
      "step": 4595
    },
    {
      "entropy": 5.868388605117798,
      "epoch": 0.3864734299516908,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004989682545815232,
      "loss": 5.7918,
      "mean_token_accuracy": 0.1475853756070137,
      "num_tokens": 8474454.0,
      "step": 4600
    },
    {
      "entropy": 5.851971101760864,
      "epoch": 0.3868935097668557,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004989653880349207,
      "loss": 5.702,
      "mean_token_accuracy": 0.14926233142614365,
      "num_tokens": 8482694.0,
      "step": 4605
    },
    {
      "entropy": 5.900271463394165,
      "epoch": 0.38731358958202056,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004989625175208864,
      "loss": 5.8168,
      "mean_token_accuracy": 0.1423931635916233,
      "num_tokens": 8491162.0,
      "step": 4610
    },
    {
      "entropy": 5.7257490158081055,
      "epoch": 0.38773366939718545,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004989596430394717,
      "loss": 5.6821,
      "mean_token_accuracy": 0.15937959849834443,
      "num_tokens": 8500716.0,
      "step": 4615
    },
    {
      "entropy": 5.808234357833863,
      "epoch": 0.38815374921235035,
      "grad_norm": 1.34375,
      "learning_rate": 0.000498956764590727,
      "loss": 5.708,
      "mean_token_accuracy": 0.14701972305774688,
      "num_tokens": 8508871.0,
      "step": 4620
    },
    {
      "entropy": 5.95609483718872,
      "epoch": 0.38857382902751525,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004989538821747037,
      "loss": 5.9183,
      "mean_token_accuracy": 0.14093246757984162,
      "num_tokens": 8518450.0,
      "step": 4625
    },
    {
      "entropy": 5.905833387374878,
      "epoch": 0.3889939088426801,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004989509957914527,
      "loss": 5.8226,
      "mean_token_accuracy": 0.13851853534579278,
      "num_tokens": 8528238.0,
      "step": 4630
    },
    {
      "entropy": 5.809105253219604,
      "epoch": 0.389413988657845,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004989481054410251,
      "loss": 5.7101,
      "mean_token_accuracy": 0.14374718815088272,
      "num_tokens": 8537587.0,
      "step": 4635
    },
    {
      "entropy": 5.856853580474853,
      "epoch": 0.3898340684730099,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004989452111234721,
      "loss": 5.832,
      "mean_token_accuracy": 0.14598365053534507,
      "num_tokens": 8547703.0,
      "step": 4640
    },
    {
      "entropy": 5.896316909790039,
      "epoch": 0.39025414828817473,
      "grad_norm": 1.421875,
      "learning_rate": 0.000498942312838845,
      "loss": 5.7758,
      "mean_token_accuracy": 0.14673489332199097,
      "num_tokens": 8557001.0,
      "step": 4645
    },
    {
      "entropy": 5.753481054306031,
      "epoch": 0.3906742281033396,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004989394105871952,
      "loss": 5.6574,
      "mean_token_accuracy": 0.15356985479593277,
      "num_tokens": 8565638.0,
      "step": 4650
    },
    {
      "entropy": 5.9358145236969,
      "epoch": 0.3910943079185045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000498936504368574,
      "loss": 5.824,
      "mean_token_accuracy": 0.14542939141392708,
      "num_tokens": 8574428.0,
      "step": 4655
    },
    {
      "entropy": 5.850586032867431,
      "epoch": 0.3915143877336694,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004989335941830329,
      "loss": 5.7948,
      "mean_token_accuracy": 0.14559997841715813,
      "num_tokens": 8583157.0,
      "step": 4660
    },
    {
      "entropy": 5.788503217697143,
      "epoch": 0.39193446754883426,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004989306800306236,
      "loss": 5.7506,
      "mean_token_accuracy": 0.14149378091096879,
      "num_tokens": 8592382.0,
      "step": 4665
    },
    {
      "entropy": 5.7819007396697994,
      "epoch": 0.39235454736399916,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004989277619113975,
      "loss": 5.7277,
      "mean_token_accuracy": 0.15354567617177964,
      "num_tokens": 8601058.0,
      "step": 4670
    },
    {
      "entropy": 5.888355016708374,
      "epoch": 0.39277462717916406,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989248398254065,
      "loss": 5.809,
      "mean_token_accuracy": 0.1449069932103157,
      "num_tokens": 8609479.0,
      "step": 4675
    },
    {
      "entropy": 5.876518249511719,
      "epoch": 0.3931947069943289,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989219137727021,
      "loss": 5.7915,
      "mean_token_accuracy": 0.14826752841472626,
      "num_tokens": 8618860.0,
      "step": 4680
    },
    {
      "entropy": 5.84470911026001,
      "epoch": 0.3936147868094938,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004989189837533365,
      "loss": 5.735,
      "mean_token_accuracy": 0.14821926057338713,
      "num_tokens": 8627462.0,
      "step": 4685
    },
    {
      "entropy": 5.9267487049102785,
      "epoch": 0.3940348666246587,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004989160497673613,
      "loss": 5.9044,
      "mean_token_accuracy": 0.14269359409809113,
      "num_tokens": 8637569.0,
      "step": 4690
    },
    {
      "entropy": 5.871499490737915,
      "epoch": 0.39445494643982354,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004989131118148286,
      "loss": 5.717,
      "mean_token_accuracy": 0.14633206725120546,
      "num_tokens": 8645440.0,
      "step": 4695
    },
    {
      "entropy": 5.8017088890075685,
      "epoch": 0.39487502625498844,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004989101698957904,
      "loss": 5.8682,
      "mean_token_accuracy": 0.14381342753767967,
      "num_tokens": 8655077.0,
      "step": 4700
    },
    {
      "entropy": 5.897370004653931,
      "epoch": 0.39529510607015333,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004989072240102988,
      "loss": 5.802,
      "mean_token_accuracy": 0.14655678868293762,
      "num_tokens": 8663126.0,
      "step": 4705
    },
    {
      "entropy": 5.9488218307495115,
      "epoch": 0.39571518588531823,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004989042741584061,
      "loss": 5.7704,
      "mean_token_accuracy": 0.14519642665982246,
      "num_tokens": 8672386.0,
      "step": 4710
    },
    {
      "entropy": 5.720374536514282,
      "epoch": 0.3961352657004831,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004989013203401645,
      "loss": 5.7158,
      "mean_token_accuracy": 0.14469049870967865,
      "num_tokens": 8681930.0,
      "step": 4715
    },
    {
      "entropy": 5.841011619567871,
      "epoch": 0.396555345515648,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004988983625556264,
      "loss": 5.7771,
      "mean_token_accuracy": 0.14254847168922424,
      "num_tokens": 8690993.0,
      "step": 4720
    },
    {
      "entropy": 5.822189235687256,
      "epoch": 0.39697542533081287,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004988954008048438,
      "loss": 5.7542,
      "mean_token_accuracy": 0.1459605447947979,
      "num_tokens": 8699497.0,
      "step": 4725
    },
    {
      "entropy": 5.995753383636474,
      "epoch": 0.3973955051459777,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004988924350878697,
      "loss": 5.9601,
      "mean_token_accuracy": 0.13361823558807373,
      "num_tokens": 8709274.0,
      "step": 4730
    },
    {
      "entropy": 5.916924142837525,
      "epoch": 0.3978155849611426,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004988894654047563,
      "loss": 5.8045,
      "mean_token_accuracy": 0.14044342935085297,
      "num_tokens": 8718158.0,
      "step": 4735
    },
    {
      "entropy": 5.780064105987549,
      "epoch": 0.3982356647763075,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004988864917555562,
      "loss": 5.6905,
      "mean_token_accuracy": 0.147587950527668,
      "num_tokens": 8727459.0,
      "step": 4740
    },
    {
      "entropy": 5.872677993774414,
      "epoch": 0.3986557445914724,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004988835141403224,
      "loss": 5.838,
      "mean_token_accuracy": 0.15111552625894548,
      "num_tokens": 8737614.0,
      "step": 4745
    },
    {
      "entropy": 5.793733787536621,
      "epoch": 0.39907582440663725,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004988805325591073,
      "loss": 5.6586,
      "mean_token_accuracy": 0.14995396584272386,
      "num_tokens": 8746799.0,
      "step": 4750
    },
    {
      "entropy": 5.854796743392944,
      "epoch": 0.39949590422180214,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004988775470119639,
      "loss": 5.8437,
      "mean_token_accuracy": 0.14196141958236694,
      "num_tokens": 8756555.0,
      "step": 4755
    },
    {
      "entropy": 5.811031866073608,
      "epoch": 0.39991598403696704,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004988745574989451,
      "loss": 5.8678,
      "mean_token_accuracy": 0.14688000455498695,
      "num_tokens": 8765849.0,
      "step": 4760
    },
    {
      "entropy": 5.97343111038208,
      "epoch": 0.4003360638521319,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004988715640201036,
      "loss": 5.9254,
      "mean_token_accuracy": 0.14007072225213052,
      "num_tokens": 8775713.0,
      "step": 4765
    },
    {
      "entropy": 5.833617496490478,
      "epoch": 0.4007561436672968,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004988685665754928,
      "loss": 5.757,
      "mean_token_accuracy": 0.14587045535445214,
      "num_tokens": 8784717.0,
      "step": 4770
    },
    {
      "entropy": 5.850550556182862,
      "epoch": 0.4011762234824617,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004988655651651656,
      "loss": 5.7758,
      "mean_token_accuracy": 0.14603159129619597,
      "num_tokens": 8794388.0,
      "step": 4775
    },
    {
      "entropy": 5.799873685836792,
      "epoch": 0.4015963032976265,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004988625597891751,
      "loss": 5.7832,
      "mean_token_accuracy": 0.1439102217555046,
      "num_tokens": 8802436.0,
      "step": 4780
    },
    {
      "entropy": 5.901528596878052,
      "epoch": 0.4020163831127914,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004988595504475746,
      "loss": 5.7387,
      "mean_token_accuracy": 0.14495839625597,
      "num_tokens": 8811184.0,
      "step": 4785
    },
    {
      "entropy": 5.891771650314331,
      "epoch": 0.4024364629279563,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004988565371404175,
      "loss": 5.8182,
      "mean_token_accuracy": 0.14259228706359864,
      "num_tokens": 8820525.0,
      "step": 4790
    },
    {
      "entropy": 5.840786600112915,
      "epoch": 0.4028565427431212,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004988535198677571,
      "loss": 5.6717,
      "mean_token_accuracy": 0.15582364946603774,
      "num_tokens": 8828928.0,
      "step": 4795
    },
    {
      "entropy": 5.857646703720093,
      "epoch": 0.40327662255828606,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004988504986296469,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1377402052283287,
      "num_tokens": 8838615.0,
      "step": 4800
    },
    {
      "entropy": 5.874919462203979,
      "epoch": 0.40369670237345096,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004988474734261404,
      "loss": 5.8723,
      "mean_token_accuracy": 0.1379916787147522,
      "num_tokens": 8848709.0,
      "step": 4805
    },
    {
      "entropy": 5.955801677703858,
      "epoch": 0.40411678218861585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004988444442572911,
      "loss": 5.8116,
      "mean_token_accuracy": 0.1377601645886898,
      "num_tokens": 8858277.0,
      "step": 4810
    },
    {
      "entropy": 5.8524405002594,
      "epoch": 0.4045368620037807,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004988414111231528,
      "loss": 5.7865,
      "mean_token_accuracy": 0.146932952105999,
      "num_tokens": 8868436.0,
      "step": 4815
    },
    {
      "entropy": 5.81365385055542,
      "epoch": 0.4049569418189456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000498838374023779,
      "loss": 5.7632,
      "mean_token_accuracy": 0.1458624616265297,
      "num_tokens": 8877740.0,
      "step": 4820
    },
    {
      "entropy": 5.897781801223755,
      "epoch": 0.4053770216341105,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004988353329592239,
      "loss": 5.7534,
      "mean_token_accuracy": 0.1458469048142433,
      "num_tokens": 8887408.0,
      "step": 4825
    },
    {
      "entropy": 5.865758180618286,
      "epoch": 0.4057971014492754,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004988322879295409,
      "loss": 5.9214,
      "mean_token_accuracy": 0.13947931975126265,
      "num_tokens": 8897141.0,
      "step": 4830
    },
    {
      "entropy": 5.81132230758667,
      "epoch": 0.40621718126444023,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004988292389347844,
      "loss": 5.6894,
      "mean_token_accuracy": 0.153226038813591,
      "num_tokens": 8905747.0,
      "step": 4835
    },
    {
      "entropy": 5.9674177169799805,
      "epoch": 0.40663726107960513,
      "grad_norm": 1.28125,
      "learning_rate": 0.000498826185975008,
      "loss": 5.8416,
      "mean_token_accuracy": 0.14231978505849838,
      "num_tokens": 8914926.0,
      "step": 4840
    },
    {
      "entropy": 5.819942045211792,
      "epoch": 0.40705734089477,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004988231290502662,
      "loss": 5.8576,
      "mean_token_accuracy": 0.14279644340276718,
      "num_tokens": 8923956.0,
      "step": 4845
    },
    {
      "entropy": 5.920290994644165,
      "epoch": 0.40747742070993487,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004988200681606127,
      "loss": 5.724,
      "mean_token_accuracy": 0.1409930519759655,
      "num_tokens": 8932654.0,
      "step": 4850
    },
    {
      "entropy": 5.8871392726898195,
      "epoch": 0.40789750052509977,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000498817003306102,
      "loss": 5.7162,
      "mean_token_accuracy": 0.15066221207380295,
      "num_tokens": 8941716.0,
      "step": 4855
    },
    {
      "entropy": 5.792912864685059,
      "epoch": 0.40831758034026466,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004988139344867884,
      "loss": 5.7907,
      "mean_token_accuracy": 0.14440437257289887,
      "num_tokens": 8950377.0,
      "step": 4860
    },
    {
      "entropy": 5.823495244979858,
      "epoch": 0.4087376601554295,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988108617027261,
      "loss": 5.7438,
      "mean_token_accuracy": 0.1429793991148472,
      "num_tokens": 8959857.0,
      "step": 4865
    },
    {
      "entropy": 5.811827516555786,
      "epoch": 0.4091577399705944,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004988077849539698,
      "loss": 5.6945,
      "mean_token_accuracy": 0.14948356971144677,
      "num_tokens": 8968272.0,
      "step": 4870
    },
    {
      "entropy": 5.826247787475586,
      "epoch": 0.4095778197857593,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004988047042405736,
      "loss": 5.765,
      "mean_token_accuracy": 0.1494896613061428,
      "num_tokens": 8977445.0,
      "step": 4875
    },
    {
      "entropy": 5.932396030426025,
      "epoch": 0.4099978996009242,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004988016195625924,
      "loss": 5.8399,
      "mean_token_accuracy": 0.13975587710738183,
      "num_tokens": 8987315.0,
      "step": 4880
    },
    {
      "entropy": 5.874420547485352,
      "epoch": 0.41041797941608904,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004987985309200807,
      "loss": 5.8545,
      "mean_token_accuracy": 0.14247876554727554,
      "num_tokens": 8998119.0,
      "step": 4885
    },
    {
      "entropy": 5.734928989410401,
      "epoch": 0.41083805923125394,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004987954383130934,
      "loss": 5.7211,
      "mean_token_accuracy": 0.15389348119497298,
      "num_tokens": 9007167.0,
      "step": 4890
    },
    {
      "entropy": 5.831616592407227,
      "epoch": 0.41125813904641884,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000498792341741685,
      "loss": 5.7751,
      "mean_token_accuracy": 0.14183046370744706,
      "num_tokens": 9016690.0,
      "step": 4895
    },
    {
      "entropy": 5.9142228126525875,
      "epoch": 0.4116782188615837,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004987892412059106,
      "loss": 5.8608,
      "mean_token_accuracy": 0.14799081161618233,
      "num_tokens": 9026117.0,
      "step": 4900
    },
    {
      "entropy": 5.786413145065308,
      "epoch": 0.4120982986767486,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004987861367058251,
      "loss": 5.7333,
      "mean_token_accuracy": 0.14700580313801764,
      "num_tokens": 9035754.0,
      "step": 4905
    },
    {
      "entropy": 5.831767606735229,
      "epoch": 0.4125183784919135,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004987830282414833,
      "loss": 5.7383,
      "mean_token_accuracy": 0.1511758364737034,
      "num_tokens": 9045453.0,
      "step": 4910
    },
    {
      "entropy": 5.920518159866333,
      "epoch": 0.41293845830707837,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004987799158129404,
      "loss": 5.866,
      "mean_token_accuracy": 0.14217546358704566,
      "num_tokens": 9056045.0,
      "step": 4915
    },
    {
      "entropy": 5.811709785461426,
      "epoch": 0.4133585381222432,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004987767994202516,
      "loss": 5.7367,
      "mean_token_accuracy": 0.14320328831672668,
      "num_tokens": 9065728.0,
      "step": 4920
    },
    {
      "entropy": 5.832678318023682,
      "epoch": 0.4137786179374081,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004987736790634719,
      "loss": 5.7682,
      "mean_token_accuracy": 0.14474476575851442,
      "num_tokens": 9075522.0,
      "step": 4925
    },
    {
      "entropy": 5.789442873001098,
      "epoch": 0.414198697752573,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004987705547426568,
      "loss": 5.7403,
      "mean_token_accuracy": 0.141367207467556,
      "num_tokens": 9084412.0,
      "step": 4930
    },
    {
      "entropy": 5.866390943527222,
      "epoch": 0.41461877756773785,
      "grad_norm": 1.375,
      "learning_rate": 0.0004987674264578615,
      "loss": 5.8202,
      "mean_token_accuracy": 0.13946182429790496,
      "num_tokens": 9094289.0,
      "step": 4935
    },
    {
      "entropy": 5.853621578216552,
      "epoch": 0.41503885738290275,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004987642942091414,
      "loss": 5.7305,
      "mean_token_accuracy": 0.1456735163927078,
      "num_tokens": 9103124.0,
      "step": 4940
    },
    {
      "entropy": 5.831802606582642,
      "epoch": 0.41545893719806765,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004987611579965523,
      "loss": 5.6742,
      "mean_token_accuracy": 0.14086953178048134,
      "num_tokens": 9112794.0,
      "step": 4945
    },
    {
      "entropy": 5.873828983306884,
      "epoch": 0.4158790170132325,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004987580178201492,
      "loss": 5.8342,
      "mean_token_accuracy": 0.1499299481511116,
      "num_tokens": 9122718.0,
      "step": 4950
    },
    {
      "entropy": 5.850664281845093,
      "epoch": 0.4162990968283974,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004987548736799882,
      "loss": 5.8516,
      "mean_token_accuracy": 0.14340481981635095,
      "num_tokens": 9131855.0,
      "step": 4955
    },
    {
      "entropy": 5.820421504974365,
      "epoch": 0.4167191766435623,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004987517255761248,
      "loss": 5.6959,
      "mean_token_accuracy": 0.15145303905010224,
      "num_tokens": 9141102.0,
      "step": 4960
    },
    {
      "entropy": 5.806706857681275,
      "epoch": 0.4171392564587272,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004987485735086148,
      "loss": 5.7767,
      "mean_token_accuracy": 0.14581410586833954,
      "num_tokens": 9150552.0,
      "step": 4965
    },
    {
      "entropy": 5.901687812805176,
      "epoch": 0.417559336273892,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000498745417477514,
      "loss": 5.7678,
      "mean_token_accuracy": 0.1465201199054718,
      "num_tokens": 9160105.0,
      "step": 4970
    },
    {
      "entropy": 5.833481121063232,
      "epoch": 0.4179794160890569,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004987422574828784,
      "loss": 5.7442,
      "mean_token_accuracy": 0.14697531685233117,
      "num_tokens": 9169367.0,
      "step": 4975
    },
    {
      "entropy": 5.801551008224488,
      "epoch": 0.4183994959042218,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004987390935247639,
      "loss": 5.6473,
      "mean_token_accuracy": 0.1524960733950138,
      "num_tokens": 9177872.0,
      "step": 4980
    },
    {
      "entropy": 5.88420295715332,
      "epoch": 0.41881957571938666,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004987359256032265,
      "loss": 5.842,
      "mean_token_accuracy": 0.13641551584005357,
      "num_tokens": 9187879.0,
      "step": 4985
    },
    {
      "entropy": 5.839552402496338,
      "epoch": 0.41923965553455156,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004987327537183225,
      "loss": 5.7627,
      "mean_token_accuracy": 0.14583497866988182,
      "num_tokens": 9198281.0,
      "step": 4990
    },
    {
      "entropy": 5.830025053024292,
      "epoch": 0.41965973534971646,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004987295778701078,
      "loss": 5.7616,
      "mean_token_accuracy": 0.1472972884774208,
      "num_tokens": 9207670.0,
      "step": 4995
    },
    {
      "entropy": 5.890192222595215,
      "epoch": 0.42007981516488135,
      "grad_norm": 1.3125,
      "learning_rate": 0.000498726398058639,
      "loss": 5.7624,
      "mean_token_accuracy": 0.1478295773267746,
      "num_tokens": 9216995.0,
      "step": 5000
    },
    {
      "entropy": 5.86348524093628,
      "epoch": 0.4204998949800462,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004987232142839723,
      "loss": 5.84,
      "mean_token_accuracy": 0.14336878657341004,
      "num_tokens": 9227330.0,
      "step": 5005
    },
    {
      "entropy": 5.881588125228882,
      "epoch": 0.4209199747952111,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004987200265461638,
      "loss": 5.765,
      "mean_token_accuracy": 0.15374772921204566,
      "num_tokens": 9236666.0,
      "step": 5010
    },
    {
      "entropy": 5.874930953979492,
      "epoch": 0.421340054610376,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004987168348452705,
      "loss": 5.7753,
      "mean_token_accuracy": 0.14532062411308289,
      "num_tokens": 9246388.0,
      "step": 5015
    },
    {
      "entropy": 5.783464574813843,
      "epoch": 0.42176013442554083,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004987136391813485,
      "loss": 5.7285,
      "mean_token_accuracy": 0.15355198979377746,
      "num_tokens": 9255239.0,
      "step": 5020
    },
    {
      "entropy": 5.740076160430908,
      "epoch": 0.42218021424070573,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004987104395544547,
      "loss": 5.7036,
      "mean_token_accuracy": 0.14613911658525466,
      "num_tokens": 9264468.0,
      "step": 5025
    },
    {
      "entropy": 5.791272926330566,
      "epoch": 0.42260029405587063,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004987072359646455,
      "loss": 5.767,
      "mean_token_accuracy": 0.15074100941419602,
      "num_tokens": 9274140.0,
      "step": 5030
    },
    {
      "entropy": 5.844473838806152,
      "epoch": 0.42302037387103547,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004987040284119778,
      "loss": 5.7527,
      "mean_token_accuracy": 0.14496962279081343,
      "num_tokens": 9283539.0,
      "step": 5035
    },
    {
      "entropy": 5.793737411499023,
      "epoch": 0.42344045368620037,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004987008168965087,
      "loss": 5.7492,
      "mean_token_accuracy": 0.1446337193250656,
      "num_tokens": 9292664.0,
      "step": 5040
    },
    {
      "entropy": 5.882410097122192,
      "epoch": 0.42386053350136527,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004986976014182946,
      "loss": 5.8632,
      "mean_token_accuracy": 0.14475177749991416,
      "num_tokens": 9302814.0,
      "step": 5045
    },
    {
      "entropy": 5.91447172164917,
      "epoch": 0.42428061331653016,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004986943819773927,
      "loss": 5.8446,
      "mean_token_accuracy": 0.13993989303708076,
      "num_tokens": 9312654.0,
      "step": 5050
    },
    {
      "entropy": 5.944200277328491,
      "epoch": 0.424700693131695,
      "grad_norm": 1.71875,
      "learning_rate": 0.00049869115857386,
      "loss": 5.8629,
      "mean_token_accuracy": 0.13871566727757453,
      "num_tokens": 9322271.0,
      "step": 5055
    },
    {
      "entropy": 5.914797401428222,
      "epoch": 0.4251207729468599,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004986879312077536,
      "loss": 5.7991,
      "mean_token_accuracy": 0.14315683469176294,
      "num_tokens": 9331341.0,
      "step": 5060
    },
    {
      "entropy": 5.854491138458252,
      "epoch": 0.4255408527620248,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004986846998791308,
      "loss": 5.7292,
      "mean_token_accuracy": 0.1450161539018154,
      "num_tokens": 9339863.0,
      "step": 5065
    },
    {
      "entropy": 5.760708379745483,
      "epoch": 0.42596093257718964,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004986814645880485,
      "loss": 5.7122,
      "mean_token_accuracy": 0.14831122979521752,
      "num_tokens": 9349488.0,
      "step": 5070
    },
    {
      "entropy": 5.763440895080566,
      "epoch": 0.42638101239235454,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986782253345645,
      "loss": 5.7155,
      "mean_token_accuracy": 0.1446495160460472,
      "num_tokens": 9357977.0,
      "step": 5075
    },
    {
      "entropy": 5.873974847793579,
      "epoch": 0.42680109220751944,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004986749821187358,
      "loss": 5.8291,
      "mean_token_accuracy": 0.14651158899068834,
      "num_tokens": 9367449.0,
      "step": 5080
    },
    {
      "entropy": 5.939826440811157,
      "epoch": 0.42722117202268434,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00049867173494062,
      "loss": 5.8443,
      "mean_token_accuracy": 0.14635560363531114,
      "num_tokens": 9377070.0,
      "step": 5085
    },
    {
      "entropy": 5.814030504226684,
      "epoch": 0.4276412518378492,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004986684838002744,
      "loss": 5.6402,
      "mean_token_accuracy": 0.14356765672564506,
      "num_tokens": 9385881.0,
      "step": 5090
    },
    {
      "entropy": 5.812148237228394,
      "epoch": 0.4280613316530141,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004986652286977569,
      "loss": 5.7935,
      "mean_token_accuracy": 0.14297776967287062,
      "num_tokens": 9395159.0,
      "step": 5095
    },
    {
      "entropy": 5.8577290058135985,
      "epoch": 0.428481411468179,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004986619696331252,
      "loss": 5.7352,
      "mean_token_accuracy": 0.14707281142473222,
      "num_tokens": 9404590.0,
      "step": 5100
    },
    {
      "entropy": 5.849812984466553,
      "epoch": 0.4289014912833438,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004986587066064367,
      "loss": 5.7485,
      "mean_token_accuracy": 0.15095358043909074,
      "num_tokens": 9414452.0,
      "step": 5105
    },
    {
      "entropy": 5.895452976226807,
      "epoch": 0.4293215710985087,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004986554396177494,
      "loss": 5.8792,
      "mean_token_accuracy": 0.136289519071579,
      "num_tokens": 9424004.0,
      "step": 5110
    },
    {
      "entropy": 5.892696142196655,
      "epoch": 0.4297416509136736,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004986521686671212,
      "loss": 5.7555,
      "mean_token_accuracy": 0.1541634440422058,
      "num_tokens": 9433487.0,
      "step": 5115
    },
    {
      "entropy": 5.878170013427734,
      "epoch": 0.43016173072883845,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00049864889375461,
      "loss": 5.8153,
      "mean_token_accuracy": 0.14046706035733222,
      "num_tokens": 9442742.0,
      "step": 5120
    },
    {
      "entropy": 5.880551862716675,
      "epoch": 0.43058181054400335,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004986456148802738,
      "loss": 5.8803,
      "mean_token_accuracy": 0.1430477738380432,
      "num_tokens": 9452550.0,
      "step": 5125
    },
    {
      "entropy": 5.947691774368286,
      "epoch": 0.43100189035916825,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004986423320441707,
      "loss": 5.8325,
      "mean_token_accuracy": 0.13825444877147675,
      "num_tokens": 9461920.0,
      "step": 5130
    },
    {
      "entropy": 5.91607780456543,
      "epoch": 0.43142197017433315,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004986390452463588,
      "loss": 5.7649,
      "mean_token_accuracy": 0.1430374413728714,
      "num_tokens": 9470817.0,
      "step": 5135
    },
    {
      "entropy": 5.730179119110107,
      "epoch": 0.431842049989498,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004986357544868964,
      "loss": 5.7124,
      "mean_token_accuracy": 0.15054248571395873,
      "num_tokens": 9479936.0,
      "step": 5140
    },
    {
      "entropy": 5.873006772994995,
      "epoch": 0.4322621298046629,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004986324597658418,
      "loss": 5.7327,
      "mean_token_accuracy": 0.15253930985927583,
      "num_tokens": 9489818.0,
      "step": 5145
    },
    {
      "entropy": 5.753958749771118,
      "epoch": 0.4326822096198278,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004986291610832533,
      "loss": 5.7373,
      "mean_token_accuracy": 0.1437373712658882,
      "num_tokens": 9499688.0,
      "step": 5150
    },
    {
      "entropy": 5.927468347549438,
      "epoch": 0.4331022894349926,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004986258584391892,
      "loss": 5.8034,
      "mean_token_accuracy": 0.14375736117362975,
      "num_tokens": 9509581.0,
      "step": 5155
    },
    {
      "entropy": 5.965673732757568,
      "epoch": 0.4335223692501575,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004986225518337084,
      "loss": 5.8699,
      "mean_token_accuracy": 0.14116424545645714,
      "num_tokens": 9518556.0,
      "step": 5160
    },
    {
      "entropy": 5.785204839706421,
      "epoch": 0.4339424490653224,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004986192412668692,
      "loss": 5.7774,
      "mean_token_accuracy": 0.14554062783718108,
      "num_tokens": 9527612.0,
      "step": 5165
    },
    {
      "entropy": 5.780537843704224,
      "epoch": 0.4343625288804873,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004986159267387302,
      "loss": 5.6635,
      "mean_token_accuracy": 0.15324972867965697,
      "num_tokens": 9535882.0,
      "step": 5170
    },
    {
      "entropy": 5.812596940994263,
      "epoch": 0.43478260869565216,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004986126082493502,
      "loss": 5.7687,
      "mean_token_accuracy": 0.15061265081167222,
      "num_tokens": 9544799.0,
      "step": 5175
    },
    {
      "entropy": 5.831722640991211,
      "epoch": 0.43520268851081706,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004986092857987881,
      "loss": 5.6862,
      "mean_token_accuracy": 0.14822531789541243,
      "num_tokens": 9553805.0,
      "step": 5180
    },
    {
      "entropy": 5.813159799575805,
      "epoch": 0.43562276832598196,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004986059593871026,
      "loss": 5.717,
      "mean_token_accuracy": 0.14472756162285805,
      "num_tokens": 9563493.0,
      "step": 5185
    },
    {
      "entropy": 5.843486309051514,
      "epoch": 0.4360428481411468,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004986026290143527,
      "loss": 5.8036,
      "mean_token_accuracy": 0.1416473552584648,
      "num_tokens": 9572297.0,
      "step": 5190
    },
    {
      "entropy": 5.981090354919433,
      "epoch": 0.4364629279563117,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004985992946805973,
      "loss": 5.944,
      "mean_token_accuracy": 0.13955255076289177,
      "num_tokens": 9581967.0,
      "step": 5195
    },
    {
      "entropy": 5.916019868850708,
      "epoch": 0.4368830077714766,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004985959563858955,
      "loss": 5.8377,
      "mean_token_accuracy": 0.14462848305702208,
      "num_tokens": 9590885.0,
      "step": 5200
    },
    {
      "entropy": 5.895185899734497,
      "epoch": 0.43730308758664144,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004985926141303066,
      "loss": 5.7664,
      "mean_token_accuracy": 0.14494283646345138,
      "num_tokens": 9599247.0,
      "step": 5205
    },
    {
      "entropy": 5.799692010879516,
      "epoch": 0.43772316740180633,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004985892679138896,
      "loss": 5.6813,
      "mean_token_accuracy": 0.15228856280446051,
      "num_tokens": 9608296.0,
      "step": 5210
    },
    {
      "entropy": 5.7856512546539305,
      "epoch": 0.43814324721697123,
      "grad_norm": 1.625,
      "learning_rate": 0.0004985859177367038,
      "loss": 5.743,
      "mean_token_accuracy": 0.14503268152475357,
      "num_tokens": 9616734.0,
      "step": 5215
    },
    {
      "entropy": 5.923814201354981,
      "epoch": 0.43856332703213613,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004985825635988087,
      "loss": 5.8284,
      "mean_token_accuracy": 0.14173559248447418,
      "num_tokens": 9626246.0,
      "step": 5220
    },
    {
      "entropy": 5.93772325515747,
      "epoch": 0.43898340684730097,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004985792055002635,
      "loss": 5.6928,
      "mean_token_accuracy": 0.14719630777835846,
      "num_tokens": 9634963.0,
      "step": 5225
    },
    {
      "entropy": 5.874487066268921,
      "epoch": 0.43940348666246587,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004985758434411278,
      "loss": 5.7792,
      "mean_token_accuracy": 0.1490402415394783,
      "num_tokens": 9643615.0,
      "step": 5230
    },
    {
      "entropy": 5.830899858474732,
      "epoch": 0.43982356647763077,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004985724774214613,
      "loss": 5.7618,
      "mean_token_accuracy": 0.14308954402804375,
      "num_tokens": 9653306.0,
      "step": 5235
    },
    {
      "entropy": 5.849113464355469,
      "epoch": 0.4402436462927956,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004985691074413233,
      "loss": 5.7907,
      "mean_token_accuracy": 0.14236897826194764,
      "num_tokens": 9662389.0,
      "step": 5240
    },
    {
      "entropy": 5.755652236938476,
      "epoch": 0.4406637261079605,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004985657335007739,
      "loss": 5.7572,
      "mean_token_accuracy": 0.1460557647049427,
      "num_tokens": 9671183.0,
      "step": 5245
    },
    {
      "entropy": 5.91265082359314,
      "epoch": 0.4410838059231254,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004985623555998725,
      "loss": 5.7558,
      "mean_token_accuracy": 0.15076574087142944,
      "num_tokens": 9680544.0,
      "step": 5250
    },
    {
      "entropy": 5.915414428710937,
      "epoch": 0.4415038857382903,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004985589737386791,
      "loss": 5.7849,
      "mean_token_accuracy": 0.14628923386335374,
      "num_tokens": 9690137.0,
      "step": 5255
    },
    {
      "entropy": 5.76580057144165,
      "epoch": 0.44192396555345514,
      "grad_norm": 1.375,
      "learning_rate": 0.0004985555879172535,
      "loss": 5.718,
      "mean_token_accuracy": 0.14928821474313736,
      "num_tokens": 9699149.0,
      "step": 5260
    },
    {
      "entropy": 5.85457215309143,
      "epoch": 0.44234404536862004,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000498552198135656,
      "loss": 5.8061,
      "mean_token_accuracy": 0.1479725457727909,
      "num_tokens": 9709308.0,
      "step": 5265
    },
    {
      "entropy": 5.86218228340149,
      "epoch": 0.44276412518378494,
      "grad_norm": 1.5,
      "learning_rate": 0.0004985488043939462,
      "loss": 5.7433,
      "mean_token_accuracy": 0.14763879179954528,
      "num_tokens": 9718462.0,
      "step": 5270
    },
    {
      "entropy": 5.848517847061157,
      "epoch": 0.4431842049989498,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004985454066921846,
      "loss": 5.6657,
      "mean_token_accuracy": 0.15294953137636186,
      "num_tokens": 9727626.0,
      "step": 5275
    },
    {
      "entropy": 5.697809362411499,
      "epoch": 0.4436042848141147,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004985420050304312,
      "loss": 5.6983,
      "mean_token_accuracy": 0.14653837010264398,
      "num_tokens": 9737091.0,
      "step": 5280
    },
    {
      "entropy": 5.776859283447266,
      "epoch": 0.4440243646292796,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004985385994087462,
      "loss": 5.7667,
      "mean_token_accuracy": 0.14408197328448297,
      "num_tokens": 9746135.0,
      "step": 5285
    },
    {
      "entropy": 5.90691499710083,
      "epoch": 0.4444444444444444,
      "grad_norm": 1.5,
      "learning_rate": 0.0004985351898271901,
      "loss": 5.6909,
      "mean_token_accuracy": 0.14973507225513458,
      "num_tokens": 9754549.0,
      "step": 5290
    },
    {
      "entropy": 5.90071930885315,
      "epoch": 0.4448645242596093,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004985317762858231,
      "loss": 5.8522,
      "mean_token_accuracy": 0.1393520101904869,
      "num_tokens": 9764219.0,
      "step": 5295
    },
    {
      "entropy": 5.850748586654663,
      "epoch": 0.4452846040747742,
      "grad_norm": 1.390625,
      "learning_rate": 0.000498528358784706,
      "loss": 5.6867,
      "mean_token_accuracy": 0.15024393796920776,
      "num_tokens": 9772234.0,
      "step": 5300
    },
    {
      "entropy": 5.805743646621704,
      "epoch": 0.4457046838899391,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000498524937323899,
      "loss": 5.7412,
      "mean_token_accuracy": 0.1513320118188858,
      "num_tokens": 9781417.0,
      "step": 5305
    },
    {
      "entropy": 5.939965295791626,
      "epoch": 0.44612476370510395,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004985215119034628,
      "loss": 5.8569,
      "mean_token_accuracy": 0.14016103446483613,
      "num_tokens": 9791286.0,
      "step": 5310
    },
    {
      "entropy": 5.8461981296539305,
      "epoch": 0.44654484352026885,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004985180825234582,
      "loss": 5.8526,
      "mean_token_accuracy": 0.14801667556166648,
      "num_tokens": 9802157.0,
      "step": 5315
    },
    {
      "entropy": 5.973970174789429,
      "epoch": 0.44696492333543375,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004985146491839459,
      "loss": 5.8417,
      "mean_token_accuracy": 0.1345731109380722,
      "num_tokens": 9812646.0,
      "step": 5320
    },
    {
      "entropy": 5.948847103118896,
      "epoch": 0.4473850031505986,
      "grad_norm": 1.5,
      "learning_rate": 0.0004985112118849865,
      "loss": 5.8516,
      "mean_token_accuracy": 0.13398924767971038,
      "num_tokens": 9822274.0,
      "step": 5325
    },
    {
      "entropy": 5.782685327529907,
      "epoch": 0.4478050829657635,
      "grad_norm": 1.375,
      "learning_rate": 0.0004985077706266412,
      "loss": 5.6553,
      "mean_token_accuracy": 0.14450234919786453,
      "num_tokens": 9831337.0,
      "step": 5330
    },
    {
      "entropy": 5.838633728027344,
      "epoch": 0.4482251627809284,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004985043254089708,
      "loss": 5.7708,
      "mean_token_accuracy": 0.1397250510752201,
      "num_tokens": 9840798.0,
      "step": 5335
    },
    {
      "entropy": 5.778713369369507,
      "epoch": 0.44864524259609323,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004985008762320364,
      "loss": 5.7606,
      "mean_token_accuracy": 0.14193158969283104,
      "num_tokens": 9850117.0,
      "step": 5340
    },
    {
      "entropy": 5.850035953521728,
      "epoch": 0.4490653224112581,
      "grad_norm": 1.328125,
      "learning_rate": 0.000498497423095899,
      "loss": 5.7103,
      "mean_token_accuracy": 0.1520538941025734,
      "num_tokens": 9858227.0,
      "step": 5345
    },
    {
      "entropy": 5.814627599716187,
      "epoch": 0.449485402226423,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004984939660006199,
      "loss": 5.7869,
      "mean_token_accuracy": 0.14539863988757135,
      "num_tokens": 9867157.0,
      "step": 5350
    },
    {
      "entropy": 5.7682719230651855,
      "epoch": 0.4499054820415879,
      "grad_norm": 1.25,
      "learning_rate": 0.0004984905049462602,
      "loss": 5.7017,
      "mean_token_accuracy": 0.14406744837760926,
      "num_tokens": 9877045.0,
      "step": 5355
    },
    {
      "entropy": 5.952353000640869,
      "epoch": 0.45032556185675277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004984870399328814,
      "loss": 5.8435,
      "mean_token_accuracy": 0.14134326651692392,
      "num_tokens": 9886637.0,
      "step": 5360
    },
    {
      "entropy": 5.7878330707550045,
      "epoch": 0.45074564167191766,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004984835709605446,
      "loss": 5.7137,
      "mean_token_accuracy": 0.15287383496761323,
      "num_tokens": 9895601.0,
      "step": 5365
    },
    {
      "entropy": 5.821693420410156,
      "epoch": 0.45116572148708256,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004984800980293116,
      "loss": 5.8616,
      "mean_token_accuracy": 0.14498891532421113,
      "num_tokens": 9904775.0,
      "step": 5370
    },
    {
      "entropy": 5.841092920303344,
      "epoch": 0.4515858013022474,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004984766211392435,
      "loss": 5.7982,
      "mean_token_accuracy": 0.14115200862288474,
      "num_tokens": 9913795.0,
      "step": 5375
    },
    {
      "entropy": 5.857733345031738,
      "epoch": 0.4520058811174123,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004984731402904024,
      "loss": 5.6297,
      "mean_token_accuracy": 0.15439519211649894,
      "num_tokens": 9922576.0,
      "step": 5380
    },
    {
      "entropy": 5.776189756393433,
      "epoch": 0.4524259609325772,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004984696554828496,
      "loss": 5.6223,
      "mean_token_accuracy": 0.1497255176305771,
      "num_tokens": 9930971.0,
      "step": 5385
    },
    {
      "entropy": 5.789079189300537,
      "epoch": 0.4528460407477421,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004984661667166468,
      "loss": 5.7472,
      "mean_token_accuracy": 0.15051774978637694,
      "num_tokens": 9939628.0,
      "step": 5390
    },
    {
      "entropy": 5.814688301086425,
      "epoch": 0.45326612056290694,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004984626739918561,
      "loss": 5.7234,
      "mean_token_accuracy": 0.15411069244146347,
      "num_tokens": 9948397.0,
      "step": 5395
    },
    {
      "entropy": 5.903667163848877,
      "epoch": 0.45368620037807184,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004984591773085391,
      "loss": 5.7883,
      "mean_token_accuracy": 0.14472548216581343,
      "num_tokens": 9957683.0,
      "step": 5400
    },
    {
      "entropy": 5.878373336791992,
      "epoch": 0.45410628019323673,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004984556766667578,
      "loss": 5.775,
      "mean_token_accuracy": 0.14814501702785493,
      "num_tokens": 9966756.0,
      "step": 5405
    },
    {
      "entropy": 5.782114458084107,
      "epoch": 0.4545263600084016,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004984521720665743,
      "loss": 5.7889,
      "mean_token_accuracy": 0.14963518232107162,
      "num_tokens": 9976000.0,
      "step": 5410
    },
    {
      "entropy": 5.890175533294678,
      "epoch": 0.4549464398235665,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004984486635080507,
      "loss": 5.7788,
      "mean_token_accuracy": 0.14989694356918334,
      "num_tokens": 9985509.0,
      "step": 5415
    },
    {
      "entropy": 5.741406440734863,
      "epoch": 0.45536651963873137,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004984451509912489,
      "loss": 5.7261,
      "mean_token_accuracy": 0.14937452524900435,
      "num_tokens": 9994342.0,
      "step": 5420
    },
    {
      "entropy": 5.76695761680603,
      "epoch": 0.4557865994538962,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004984416345162315,
      "loss": 5.7717,
      "mean_token_accuracy": 0.14712392613291742,
      "num_tokens": 10004249.0,
      "step": 5425
    },
    {
      "entropy": 5.824491500854492,
      "epoch": 0.4562066792690611,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004984381140830605,
      "loss": 5.7345,
      "mean_token_accuracy": 0.14771454110741616,
      "num_tokens": 10012430.0,
      "step": 5430
    },
    {
      "entropy": 5.907290077209472,
      "epoch": 0.456626759084226,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004984345896917984,
      "loss": 5.745,
      "mean_token_accuracy": 0.14652188569307328,
      "num_tokens": 10021434.0,
      "step": 5435
    },
    {
      "entropy": 5.826172637939453,
      "epoch": 0.4570468388993909,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004984310613425076,
      "loss": 5.7465,
      "mean_token_accuracy": 0.1499892771244049,
      "num_tokens": 10030473.0,
      "step": 5440
    },
    {
      "entropy": 5.826717472076416,
      "epoch": 0.45746691871455575,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004984275290352506,
      "loss": 5.7186,
      "mean_token_accuracy": 0.15438036620616913,
      "num_tokens": 10039057.0,
      "step": 5445
    },
    {
      "entropy": 5.879296159744262,
      "epoch": 0.45788699852972065,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004984239927700899,
      "loss": 5.8158,
      "mean_token_accuracy": 0.14881263822317123,
      "num_tokens": 10047998.0,
      "step": 5450
    },
    {
      "entropy": 5.906938457489014,
      "epoch": 0.45830707834488554,
      "grad_norm": 1.5,
      "learning_rate": 0.0004984204525470883,
      "loss": 5.7609,
      "mean_token_accuracy": 0.14349722415208815,
      "num_tokens": 10057479.0,
      "step": 5455
    },
    {
      "entropy": 5.797386360168457,
      "epoch": 0.4587271581600504,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004984169083663084,
      "loss": 5.7284,
      "mean_token_accuracy": 0.14215826913714408,
      "num_tokens": 10067754.0,
      "step": 5460
    },
    {
      "entropy": 5.769140672683716,
      "epoch": 0.4591472379752153,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004984133602278129,
      "loss": 5.8074,
      "mean_token_accuracy": 0.14632865488529206,
      "num_tokens": 10076815.0,
      "step": 5465
    },
    {
      "entropy": 5.929787874221802,
      "epoch": 0.4595673177903802,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000498409808131665,
      "loss": 5.8172,
      "mean_token_accuracy": 0.14295997619628906,
      "num_tokens": 10086300.0,
      "step": 5470
    },
    {
      "entropy": 5.820594406127929,
      "epoch": 0.4599873976055451,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004984062520779272,
      "loss": 5.714,
      "mean_token_accuracy": 0.15191168189048768,
      "num_tokens": 10095383.0,
      "step": 5475
    },
    {
      "entropy": 5.724618530273437,
      "epoch": 0.4604074774207099,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004984026920666628,
      "loss": 5.6894,
      "mean_token_accuracy": 0.14800985455513,
      "num_tokens": 10103971.0,
      "step": 5480
    },
    {
      "entropy": 5.801212739944458,
      "epoch": 0.4608275572358748,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004983991280979347,
      "loss": 5.6852,
      "mean_token_accuracy": 0.15083771497011184,
      "num_tokens": 10113028.0,
      "step": 5485
    },
    {
      "entropy": 5.8347760200500485,
      "epoch": 0.4612476370510397,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004983955601718061,
      "loss": 5.6653,
      "mean_token_accuracy": 0.14897917956113815,
      "num_tokens": 10121890.0,
      "step": 5490
    },
    {
      "entropy": 5.866452312469482,
      "epoch": 0.46166771686620456,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004983919882883401,
      "loss": 5.792,
      "mean_token_accuracy": 0.14483870565891266,
      "num_tokens": 10131655.0,
      "step": 5495
    },
    {
      "entropy": 5.878374290466309,
      "epoch": 0.46208779668136946,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004983884124476,
      "loss": 5.7823,
      "mean_token_accuracy": 0.1475811406970024,
      "num_tokens": 10140778.0,
      "step": 5500
    },
    {
      "entropy": 5.852365493774414,
      "epoch": 0.46250787649653435,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004983848326496494,
      "loss": 5.841,
      "mean_token_accuracy": 0.14311327114701272,
      "num_tokens": 10150229.0,
      "step": 5505
    },
    {
      "entropy": 5.915929079055786,
      "epoch": 0.4629279563116992,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004983812488945513,
      "loss": 5.7457,
      "mean_token_accuracy": 0.1459451988339424,
      "num_tokens": 10158939.0,
      "step": 5510
    },
    {
      "entropy": 5.816072797775268,
      "epoch": 0.4633480361268641,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004983776611823696,
      "loss": 5.7201,
      "mean_token_accuracy": 0.1463111788034439,
      "num_tokens": 10168383.0,
      "step": 5515
    },
    {
      "entropy": 5.759098815917969,
      "epoch": 0.463768115942029,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004983740695131676,
      "loss": 5.7315,
      "mean_token_accuracy": 0.14871828705072404,
      "num_tokens": 10178678.0,
      "step": 5520
    },
    {
      "entropy": 5.8468421459197994,
      "epoch": 0.4641881957571939,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000498370473887009,
      "loss": 5.7383,
      "mean_token_accuracy": 0.14837254881858825,
      "num_tokens": 10188964.0,
      "step": 5525
    },
    {
      "entropy": 5.893353366851807,
      "epoch": 0.46460827557235873,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004983668743039573,
      "loss": 5.7476,
      "mean_token_accuracy": 0.15488529577851295,
      "num_tokens": 10198333.0,
      "step": 5530
    },
    {
      "entropy": 5.760245847702026,
      "epoch": 0.46502835538752363,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004983632707640766,
      "loss": 5.7603,
      "mean_token_accuracy": 0.14777156189084054,
      "num_tokens": 10207876.0,
      "step": 5535
    },
    {
      "entropy": 5.78720121383667,
      "epoch": 0.4654484352026885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004983596632674306,
      "loss": 5.699,
      "mean_token_accuracy": 0.15150602906942368,
      "num_tokens": 10216822.0,
      "step": 5540
    },
    {
      "entropy": 5.902374649047852,
      "epoch": 0.46586851501785337,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983560518140831,
      "loss": 5.8282,
      "mean_token_accuracy": 0.13706823736429213,
      "num_tokens": 10226887.0,
      "step": 5545
    },
    {
      "entropy": 5.864826297760009,
      "epoch": 0.46628859483301827,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004983524364040982,
      "loss": 5.6734,
      "mean_token_accuracy": 0.1513754189014435,
      "num_tokens": 10235935.0,
      "step": 5550
    },
    {
      "entropy": 5.801887845993042,
      "epoch": 0.46670867464818316,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004983488170375399,
      "loss": 5.6238,
      "mean_token_accuracy": 0.1499588668346405,
      "num_tokens": 10245590.0,
      "step": 5555
    },
    {
      "entropy": 5.719043874740601,
      "epoch": 0.46712875446334806,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004983451937144723,
      "loss": 5.7005,
      "mean_token_accuracy": 0.14591586142778395,
      "num_tokens": 10255104.0,
      "step": 5560
    },
    {
      "entropy": 5.659116125106811,
      "epoch": 0.4675488342785129,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983415664349595,
      "loss": 5.571,
      "mean_token_accuracy": 0.15797292441129684,
      "num_tokens": 10264236.0,
      "step": 5565
    },
    {
      "entropy": 5.817941188812256,
      "epoch": 0.4679689140936778,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983379351990659,
      "loss": 5.6847,
      "mean_token_accuracy": 0.1529952183365822,
      "num_tokens": 10273335.0,
      "step": 5570
    },
    {
      "entropy": 5.723038148880005,
      "epoch": 0.4683889939088427,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983343000068559,
      "loss": 5.6501,
      "mean_token_accuracy": 0.1551787719130516,
      "num_tokens": 10282206.0,
      "step": 5575
    },
    {
      "entropy": 5.658549833297729,
      "epoch": 0.46880907372400754,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983306608583937,
      "loss": 5.6064,
      "mean_token_accuracy": 0.1606610283255577,
      "num_tokens": 10290056.0,
      "step": 5580
    },
    {
      "entropy": 5.765953302383423,
      "epoch": 0.46922915353917244,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983270177537438,
      "loss": 5.688,
      "mean_token_accuracy": 0.14721163958311081,
      "num_tokens": 10299726.0,
      "step": 5585
    },
    {
      "entropy": 5.791565895080566,
      "epoch": 0.46964923335433734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004983233706929708,
      "loss": 5.7499,
      "mean_token_accuracy": 0.14841315150260925,
      "num_tokens": 10308696.0,
      "step": 5590
    },
    {
      "entropy": 5.832704639434814,
      "epoch": 0.4700693131695022,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004983197196761392,
      "loss": 5.8361,
      "mean_token_accuracy": 0.14296030402183532,
      "num_tokens": 10317845.0,
      "step": 5595
    },
    {
      "entropy": 5.783017778396607,
      "epoch": 0.4704893929846671,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004983160647033139,
      "loss": 5.7164,
      "mean_token_accuracy": 0.1538068726658821,
      "num_tokens": 10326563.0,
      "step": 5600
    },
    {
      "entropy": 5.82874116897583,
      "epoch": 0.470909472799832,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004983124057745595,
      "loss": 5.7062,
      "mean_token_accuracy": 0.14498351216316224,
      "num_tokens": 10335931.0,
      "step": 5605
    },
    {
      "entropy": 5.721787977218628,
      "epoch": 0.47132955261499687,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004983087428899408,
      "loss": 5.7048,
      "mean_token_accuracy": 0.1419524259865284,
      "num_tokens": 10344984.0,
      "step": 5610
    },
    {
      "entropy": 5.79919285774231,
      "epoch": 0.4717496324301617,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004983050760495227,
      "loss": 5.7392,
      "mean_token_accuracy": 0.1474568262696266,
      "num_tokens": 10353522.0,
      "step": 5615
    },
    {
      "entropy": 5.907325601577758,
      "epoch": 0.4721697122453266,
      "grad_norm": 1.375,
      "learning_rate": 0.0004983014052533702,
      "loss": 5.7558,
      "mean_token_accuracy": 0.15104926228523255,
      "num_tokens": 10363527.0,
      "step": 5620
    },
    {
      "entropy": 5.760684251785278,
      "epoch": 0.4725897920604915,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982977305015481,
      "loss": 5.6642,
      "mean_token_accuracy": 0.14723679050803185,
      "num_tokens": 10372040.0,
      "step": 5625
    },
    {
      "entropy": 5.758435773849487,
      "epoch": 0.47300987187565635,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004982940517941219,
      "loss": 5.6546,
      "mean_token_accuracy": 0.14704783335328103,
      "num_tokens": 10381279.0,
      "step": 5630
    },
    {
      "entropy": 5.856901502609253,
      "epoch": 0.47342995169082125,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004982903691311564,
      "loss": 5.812,
      "mean_token_accuracy": 0.14307899996638299,
      "num_tokens": 10390608.0,
      "step": 5635
    },
    {
      "entropy": 5.73579969406128,
      "epoch": 0.47385003150598615,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004982866825127172,
      "loss": 5.6083,
      "mean_token_accuracy": 0.1558460235595703,
      "num_tokens": 10399851.0,
      "step": 5640
    },
    {
      "entropy": 5.866163825988769,
      "epoch": 0.47427011132115104,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004982829919388692,
      "loss": 5.9015,
      "mean_token_accuracy": 0.1417351670563221,
      "num_tokens": 10410425.0,
      "step": 5645
    },
    {
      "entropy": 5.8129795551300045,
      "epoch": 0.4746901911363159,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004982792974096781,
      "loss": 5.6731,
      "mean_token_accuracy": 0.15132275298237802,
      "num_tokens": 10418783.0,
      "step": 5650
    },
    {
      "entropy": 5.84749026298523,
      "epoch": 0.4751102709514808,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000498275598925209,
      "loss": 5.8182,
      "mean_token_accuracy": 0.14057652279734612,
      "num_tokens": 10427360.0,
      "step": 5655
    },
    {
      "entropy": 5.890136814117431,
      "epoch": 0.4755303507666457,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004982718964855277,
      "loss": 5.7924,
      "mean_token_accuracy": 0.1445530578494072,
      "num_tokens": 10436613.0,
      "step": 5660
    },
    {
      "entropy": 5.866897964477539,
      "epoch": 0.4759504305818105,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004982681900907,
      "loss": 5.824,
      "mean_token_accuracy": 0.14708172976970674,
      "num_tokens": 10445055.0,
      "step": 5665
    },
    {
      "entropy": 5.747807741165161,
      "epoch": 0.4763705103969754,
      "grad_norm": 1.21875,
      "learning_rate": 0.000498264479740791,
      "loss": 5.6512,
      "mean_token_accuracy": 0.15550965517759324,
      "num_tokens": 10454516.0,
      "step": 5670
    },
    {
      "entropy": 5.870613145828247,
      "epoch": 0.4767905902121403,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004982607654358668,
      "loss": 5.7869,
      "mean_token_accuracy": 0.15260702520608901,
      "num_tokens": 10463771.0,
      "step": 5675
    },
    {
      "entropy": 5.8192380428314205,
      "epoch": 0.47721067002730516,
      "grad_norm": 1.484375,
      "learning_rate": 0.000498257047175993,
      "loss": 5.7218,
      "mean_token_accuracy": 0.1428370013833046,
      "num_tokens": 10473783.0,
      "step": 5680
    },
    {
      "entropy": 5.814655399322509,
      "epoch": 0.47763074984247006,
      "grad_norm": 1.5,
      "learning_rate": 0.0004982533249612357,
      "loss": 5.7006,
      "mean_token_accuracy": 0.14914624020457268,
      "num_tokens": 10483424.0,
      "step": 5685
    },
    {
      "entropy": 5.733596324920654,
      "epoch": 0.47805082965763496,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004982495987916607,
      "loss": 5.6225,
      "mean_token_accuracy": 0.15583412498235702,
      "num_tokens": 10492536.0,
      "step": 5690
    },
    {
      "entropy": 5.762752342224121,
      "epoch": 0.47847090947279985,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004982458686673339,
      "loss": 5.7437,
      "mean_token_accuracy": 0.14716439545154572,
      "num_tokens": 10501616.0,
      "step": 5695
    },
    {
      "entropy": 5.899823999404907,
      "epoch": 0.4788909892879647,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004982421345883217,
      "loss": 5.7689,
      "mean_token_accuracy": 0.14301676750183107,
      "num_tokens": 10511190.0,
      "step": 5700
    },
    {
      "entropy": 5.779811668395996,
      "epoch": 0.4793110691031296,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004982383965546898,
      "loss": 5.7065,
      "mean_token_accuracy": 0.14420118555426598,
      "num_tokens": 10520310.0,
      "step": 5705
    },
    {
      "entropy": 5.831453895568847,
      "epoch": 0.4797311489182945,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004982346545665048,
      "loss": 5.6901,
      "mean_token_accuracy": 0.15481160432100297,
      "num_tokens": 10528711.0,
      "step": 5710
    },
    {
      "entropy": 5.8021493434906,
      "epoch": 0.48015122873345933,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004982309086238328,
      "loss": 5.784,
      "mean_token_accuracy": 0.1420671336352825,
      "num_tokens": 10538484.0,
      "step": 5715
    },
    {
      "entropy": 5.836937665939331,
      "epoch": 0.48057130854862423,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004982271587267403,
      "loss": 5.7229,
      "mean_token_accuracy": 0.1457110583782196,
      "num_tokens": 10547623.0,
      "step": 5720
    },
    {
      "entropy": 5.850609970092774,
      "epoch": 0.48099138836378913,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004982234048752935,
      "loss": 5.6716,
      "mean_token_accuracy": 0.14635915756225587,
      "num_tokens": 10556234.0,
      "step": 5725
    },
    {
      "entropy": 5.8946315288543705,
      "epoch": 0.481411468178954,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000498219647069559,
      "loss": 5.8951,
      "mean_token_accuracy": 0.13762107565999032,
      "num_tokens": 10566308.0,
      "step": 5730
    },
    {
      "entropy": 5.863095760345459,
      "epoch": 0.48183154799411887,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004982158853096035,
      "loss": 5.8416,
      "mean_token_accuracy": 0.13780406042933463,
      "num_tokens": 10575212.0,
      "step": 5735
    },
    {
      "entropy": 5.846006250381469,
      "epoch": 0.48225162780928377,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004982121195954935,
      "loss": 5.6097,
      "mean_token_accuracy": 0.1565147191286087,
      "num_tokens": 10584590.0,
      "step": 5740
    },
    {
      "entropy": 5.7412327289581295,
      "epoch": 0.48267170762444866,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004982083499272957,
      "loss": 5.67,
      "mean_token_accuracy": 0.14907400235533713,
      "num_tokens": 10593997.0,
      "step": 5745
    },
    {
      "entropy": 5.807267189025879,
      "epoch": 0.4830917874396135,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004982045763050768,
      "loss": 5.8041,
      "mean_token_accuracy": 0.1457974396646023,
      "num_tokens": 10603299.0,
      "step": 5750
    },
    {
      "entropy": 5.819139719009399,
      "epoch": 0.4835118672547784,
      "grad_norm": 1.25,
      "learning_rate": 0.0004982007987289041,
      "loss": 5.7361,
      "mean_token_accuracy": 0.14838966429233552,
      "num_tokens": 10613546.0,
      "step": 5755
    },
    {
      "entropy": 5.766327381134033,
      "epoch": 0.4839319470699433,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004981970171988439,
      "loss": 5.7119,
      "mean_token_accuracy": 0.15573283806443214,
      "num_tokens": 10622966.0,
      "step": 5760
    },
    {
      "entropy": 5.8138025283813475,
      "epoch": 0.48435202688510814,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004981932317149636,
      "loss": 5.7886,
      "mean_token_accuracy": 0.14435389563441275,
      "num_tokens": 10633441.0,
      "step": 5765
    },
    {
      "entropy": 5.843524694442749,
      "epoch": 0.48477210670027304,
      "grad_norm": 1.46875,
      "learning_rate": 0.00049818944227733,
      "loss": 5.7587,
      "mean_token_accuracy": 0.14196690768003464,
      "num_tokens": 10643124.0,
      "step": 5770
    },
    {
      "entropy": 5.8232954978942875,
      "epoch": 0.48519218651543794,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004981856488860105,
      "loss": 5.7342,
      "mean_token_accuracy": 0.14407299906015397,
      "num_tokens": 10652517.0,
      "step": 5775
    },
    {
      "entropy": 5.807135486602784,
      "epoch": 0.48561226633060284,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004981818515410721,
      "loss": 5.7752,
      "mean_token_accuracy": 0.1420580416917801,
      "num_tokens": 10663352.0,
      "step": 5780
    },
    {
      "entropy": 5.849091625213623,
      "epoch": 0.4860323461457677,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004981780502425821,
      "loss": 5.7942,
      "mean_token_accuracy": 0.14775322377681732,
      "num_tokens": 10672430.0,
      "step": 5785
    },
    {
      "entropy": 5.835392618179322,
      "epoch": 0.4864524259609326,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004981742449906079,
      "loss": 5.7391,
      "mean_token_accuracy": 0.1527121603488922,
      "num_tokens": 10681908.0,
      "step": 5790
    },
    {
      "entropy": 5.863359975814819,
      "epoch": 0.4868725057760975,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004981704357852168,
      "loss": 5.7322,
      "mean_token_accuracy": 0.14731571227312087,
      "num_tokens": 10691259.0,
      "step": 5795
    },
    {
      "entropy": 5.765371608734131,
      "epoch": 0.4872925855912623,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004981666226264764,
      "loss": 5.6277,
      "mean_token_accuracy": 0.14454422146081924,
      "num_tokens": 10699668.0,
      "step": 5800
    },
    {
      "entropy": 5.800967359542847,
      "epoch": 0.4877126654064272,
      "grad_norm": 1.125,
      "learning_rate": 0.0004981628055144542,
      "loss": 5.6801,
      "mean_token_accuracy": 0.15341386646032334,
      "num_tokens": 10709146.0,
      "step": 5805
    },
    {
      "entropy": 5.810514068603515,
      "epoch": 0.4881327452215921,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004981589844492177,
      "loss": 5.7695,
      "mean_token_accuracy": 0.14522838592529297,
      "num_tokens": 10718724.0,
      "step": 5810
    },
    {
      "entropy": 5.761872911453247,
      "epoch": 0.488552825036757,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004981551594308349,
      "loss": 5.7216,
      "mean_token_accuracy": 0.147171813249588,
      "num_tokens": 10728101.0,
      "step": 5815
    },
    {
      "entropy": 5.862204074859619,
      "epoch": 0.48897290485192185,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004981513304593733,
      "loss": 5.7442,
      "mean_token_accuracy": 0.15811584144830704,
      "num_tokens": 10736750.0,
      "step": 5820
    },
    {
      "entropy": 5.867376232147217,
      "epoch": 0.48939298466708675,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004981474975349006,
      "loss": 5.9234,
      "mean_token_accuracy": 0.1416974514722824,
      "num_tokens": 10746914.0,
      "step": 5825
    },
    {
      "entropy": 5.859874296188354,
      "epoch": 0.48981306448225165,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000498143660657485,
      "loss": 5.7529,
      "mean_token_accuracy": 0.1483662724494934,
      "num_tokens": 10755786.0,
      "step": 5830
    },
    {
      "entropy": 5.708901691436767,
      "epoch": 0.4902331442974165,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004981398198271944,
      "loss": 5.62,
      "mean_token_accuracy": 0.15774477571249007,
      "num_tokens": 10764821.0,
      "step": 5835
    },
    {
      "entropy": 5.774704790115356,
      "epoch": 0.4906532241125814,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004981359750440968,
      "loss": 5.7095,
      "mean_token_accuracy": 0.1461211383342743,
      "num_tokens": 10773569.0,
      "step": 5840
    },
    {
      "entropy": 5.725243186950683,
      "epoch": 0.4910733039277463,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004981321263082603,
      "loss": 5.6856,
      "mean_token_accuracy": 0.1431654214859009,
      "num_tokens": 10782298.0,
      "step": 5845
    },
    {
      "entropy": 5.738628387451172,
      "epoch": 0.4914933837429111,
      "grad_norm": 1.28125,
      "learning_rate": 0.000498128273619753,
      "loss": 5.6828,
      "mean_token_accuracy": 0.14975926280021667,
      "num_tokens": 10792087.0,
      "step": 5850
    },
    {
      "entropy": 5.819939708709716,
      "epoch": 0.491913463558076,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004981244169786433,
      "loss": 5.7701,
      "mean_token_accuracy": 0.14303557947278023,
      "num_tokens": 10801641.0,
      "step": 5855
    },
    {
      "entropy": 5.9354065418243405,
      "epoch": 0.4923335433732409,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004981205563849994,
      "loss": 5.8349,
      "mean_token_accuracy": 0.14316534698009492,
      "num_tokens": 10811612.0,
      "step": 5860
    },
    {
      "entropy": 5.813403034210205,
      "epoch": 0.4927536231884058,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004981166918388897,
      "loss": 5.6588,
      "mean_token_accuracy": 0.15225751399993898,
      "num_tokens": 10821608.0,
      "step": 5865
    },
    {
      "entropy": 5.730116319656372,
      "epoch": 0.49317370300357066,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004981128233403828,
      "loss": 5.6081,
      "mean_token_accuracy": 0.15562243461608888,
      "num_tokens": 10830679.0,
      "step": 5870
    },
    {
      "entropy": 5.774923992156983,
      "epoch": 0.49359378281873556,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000498108950889547,
      "loss": 5.6988,
      "mean_token_accuracy": 0.14943243488669394,
      "num_tokens": 10839669.0,
      "step": 5875
    },
    {
      "entropy": 5.790045118331909,
      "epoch": 0.49401386263390046,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004981050744864512,
      "loss": 5.6749,
      "mean_token_accuracy": 0.14829235821962355,
      "num_tokens": 10849666.0,
      "step": 5880
    },
    {
      "entropy": 5.747752380371094,
      "epoch": 0.4944339424490653,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004981011941311638,
      "loss": 5.5867,
      "mean_token_accuracy": 0.15378802865743638,
      "num_tokens": 10858225.0,
      "step": 5885
    },
    {
      "entropy": 5.754449367523193,
      "epoch": 0.4948540222642302,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004980973098237535,
      "loss": 5.6944,
      "mean_token_accuracy": 0.14568711966276168,
      "num_tokens": 10867466.0,
      "step": 5890
    },
    {
      "entropy": 5.807771825790406,
      "epoch": 0.4952741020793951,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004980934215642894,
      "loss": 5.734,
      "mean_token_accuracy": 0.15061958581209184,
      "num_tokens": 10875850.0,
      "step": 5895
    },
    {
      "entropy": 5.799838733673096,
      "epoch": 0.49569418189456,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00049808952935284,
      "loss": 5.6623,
      "mean_token_accuracy": 0.15344742313027382,
      "num_tokens": 10885154.0,
      "step": 5900
    },
    {
      "entropy": 5.781932210922241,
      "epoch": 0.49611426170972484,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004980856331894747,
      "loss": 5.7613,
      "mean_token_accuracy": 0.1433933824300766,
      "num_tokens": 10894080.0,
      "step": 5905
    },
    {
      "entropy": 5.767203950881958,
      "epoch": 0.49653434152488973,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004980817330742621,
      "loss": 5.7532,
      "mean_token_accuracy": 0.14163193702697754,
      "num_tokens": 10903248.0,
      "step": 5910
    },
    {
      "entropy": 5.848739862442017,
      "epoch": 0.49695442134005463,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004980778290072716,
      "loss": 5.7241,
      "mean_token_accuracy": 0.14693543910980225,
      "num_tokens": 10912939.0,
      "step": 5915
    },
    {
      "entropy": 5.852625370025635,
      "epoch": 0.4973745011552195,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004980739209885722,
      "loss": 5.7214,
      "mean_token_accuracy": 0.15410863906145095,
      "num_tokens": 10921505.0,
      "step": 5920
    },
    {
      "entropy": 5.887701892852784,
      "epoch": 0.49779458097038437,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004980700090182331,
      "loss": 5.8102,
      "mean_token_accuracy": 0.15374419540166856,
      "num_tokens": 10931861.0,
      "step": 5925
    },
    {
      "entropy": 5.864461851119995,
      "epoch": 0.49821466078554927,
      "grad_norm": 1.5,
      "learning_rate": 0.0004980660930963238,
      "loss": 5.7275,
      "mean_token_accuracy": 0.14706841260194778,
      "num_tokens": 10940810.0,
      "step": 5930
    },
    {
      "entropy": 5.740756177902222,
      "epoch": 0.4986347406007141,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004980621732229133,
      "loss": 5.6064,
      "mean_token_accuracy": 0.15016069263219833,
      "num_tokens": 10949514.0,
      "step": 5935
    },
    {
      "entropy": 5.8102155208587645,
      "epoch": 0.499054820415879,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004980582493980714,
      "loss": 5.8295,
      "mean_token_accuracy": 0.14151460081338882,
      "num_tokens": 10959161.0,
      "step": 5940
    },
    {
      "entropy": 5.781671619415283,
      "epoch": 0.4994749002310439,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004980543216218674,
      "loss": 5.6979,
      "mean_token_accuracy": 0.15644685178995132,
      "num_tokens": 10968983.0,
      "step": 5945
    },
    {
      "entropy": 5.819313478469849,
      "epoch": 0.4998949800462088,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004980503898943711,
      "loss": 5.8145,
      "mean_token_accuracy": 0.14870925694704057,
      "num_tokens": 10978044.0,
      "step": 5950
    },
    {
      "entropy": 5.8994536876678465,
      "epoch": 0.5003150598613737,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004980464542156519,
      "loss": 5.7276,
      "mean_token_accuracy": 0.15195106863975524,
      "num_tokens": 10986980.0,
      "step": 5955
    },
    {
      "entropy": 5.804496479034424,
      "epoch": 0.5007351396765385,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004980425145857796,
      "loss": 5.6674,
      "mean_token_accuracy": 0.16026414483785628,
      "num_tokens": 10995163.0,
      "step": 5960
    },
    {
      "entropy": 5.705014753341675,
      "epoch": 0.5011552194917034,
      "grad_norm": 1.28125,
      "learning_rate": 0.000498038571004824,
      "loss": 5.6084,
      "mean_token_accuracy": 0.15652424693107606,
      "num_tokens": 11003722.0,
      "step": 5965
    },
    {
      "entropy": 5.725177145004272,
      "epoch": 0.5015752993068683,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004980346234728549,
      "loss": 5.6641,
      "mean_token_accuracy": 0.1604843556880951,
      "num_tokens": 11013176.0,
      "step": 5970
    },
    {
      "entropy": 5.815953350067138,
      "epoch": 0.5019953791220332,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980306719899424,
      "loss": 5.7337,
      "mean_token_accuracy": 0.14801866561174393,
      "num_tokens": 11022636.0,
      "step": 5975
    },
    {
      "entropy": 5.758071231842041,
      "epoch": 0.5024154589371981,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004980267165561564,
      "loss": 5.6897,
      "mean_token_accuracy": 0.1525172308087349,
      "num_tokens": 11031896.0,
      "step": 5980
    },
    {
      "entropy": 5.801525115966797,
      "epoch": 0.502835538752363,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004980227571715669,
      "loss": 5.7359,
      "mean_token_accuracy": 0.14860138446092605,
      "num_tokens": 11040802.0,
      "step": 5985
    },
    {
      "entropy": 5.774550342559815,
      "epoch": 0.5032556185675279,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004980187938362441,
      "loss": 5.6483,
      "mean_token_accuracy": 0.14492484778165818,
      "num_tokens": 11049701.0,
      "step": 5990
    },
    {
      "entropy": 5.841704797744751,
      "epoch": 0.5036756983826927,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004980148265502581,
      "loss": 5.831,
      "mean_token_accuracy": 0.14276057258248329,
      "num_tokens": 11059555.0,
      "step": 5995
    },
    {
      "entropy": 5.823680114746094,
      "epoch": 0.5040957781978576,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004980108553136795,
      "loss": 5.7444,
      "mean_token_accuracy": 0.14943215250968933,
      "num_tokens": 11068940.0,
      "step": 6000
    },
    {
      "epoch": 0.5040957781978576,
      "eval_entropy": 5.697147493315468,
      "eval_loss": 5.740706920623779,
      "eval_mean_token_accuracy": 0.15491213997795275,
      "eval_num_tokens": 11068940.0,
      "eval_runtime": 27.2377,
      "eval_samples_per_second": 1371.85,
      "eval_steps_per_second": 171.49,
      "step": 6000
    },
    {
      "entropy": 5.881861591339112,
      "epoch": 0.5045158580130225,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004980068801265783,
      "loss": 5.727,
      "mean_token_accuracy": 0.14888829812407495,
      "num_tokens": 11079014.0,
      "step": 6005
    },
    {
      "entropy": 5.835190010070801,
      "epoch": 0.5049359378281874,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980029009890251,
      "loss": 5.8148,
      "mean_token_accuracy": 0.14982240945100783,
      "num_tokens": 11089526.0,
      "step": 6010
    },
    {
      "entropy": 5.815138959884644,
      "epoch": 0.5053560176433523,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004979989179010904,
      "loss": 5.6997,
      "mean_token_accuracy": 0.15157256349921228,
      "num_tokens": 11099156.0,
      "step": 6015
    },
    {
      "entropy": 5.769509267807007,
      "epoch": 0.5057760974585171,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004979949308628445,
      "loss": 5.6704,
      "mean_token_accuracy": 0.1513045035302639,
      "num_tokens": 11108242.0,
      "step": 6020
    },
    {
      "entropy": 5.701773405075073,
      "epoch": 0.506196177273682,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979909398743584,
      "loss": 5.6759,
      "mean_token_accuracy": 0.15321220606565475,
      "num_tokens": 11118076.0,
      "step": 6025
    },
    {
      "entropy": 5.794755840301514,
      "epoch": 0.5066162570888468,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004979869449357026,
      "loss": 5.7614,
      "mean_token_accuracy": 0.15703760236501693,
      "num_tokens": 11127265.0,
      "step": 6030
    },
    {
      "entropy": 5.807633590698242,
      "epoch": 0.5070363369040117,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979829460469478,
      "loss": 5.6736,
      "mean_token_accuracy": 0.15248569697141648,
      "num_tokens": 11136429.0,
      "step": 6035
    },
    {
      "entropy": 5.786873388290405,
      "epoch": 0.5074564167191766,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004979789432081649,
      "loss": 5.6848,
      "mean_token_accuracy": 0.14916185587644576,
      "num_tokens": 11146201.0,
      "step": 6040
    },
    {
      "entropy": 5.832711553573608,
      "epoch": 0.5078764965343415,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000497974936419425,
      "loss": 5.7005,
      "mean_token_accuracy": 0.15159857869148255,
      "num_tokens": 11154867.0,
      "step": 6045
    },
    {
      "entropy": 5.749881458282471,
      "epoch": 0.5082965763495064,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004979709256807989,
      "loss": 5.7373,
      "mean_token_accuracy": 0.15010453313589095,
      "num_tokens": 11164092.0,
      "step": 6050
    },
    {
      "entropy": 5.8348558902740475,
      "epoch": 0.5087166561646713,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004979669109923575,
      "loss": 5.7561,
      "mean_token_accuracy": 0.14959800839424134,
      "num_tokens": 11173176.0,
      "step": 6055
    },
    {
      "entropy": 5.888238859176636,
      "epoch": 0.5091367359798362,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004979628923541721,
      "loss": 5.7312,
      "mean_token_accuracy": 0.15100948065519332,
      "num_tokens": 11182397.0,
      "step": 6060
    },
    {
      "entropy": 5.828408145904541,
      "epoch": 0.509556815795001,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000497958869766314,
      "loss": 5.7777,
      "mean_token_accuracy": 0.1484249010682106,
      "num_tokens": 11191790.0,
      "step": 6065
    },
    {
      "entropy": 5.782152128219605,
      "epoch": 0.5099768956101659,
      "grad_norm": 1.25,
      "learning_rate": 0.0004979548432288543,
      "loss": 5.6808,
      "mean_token_accuracy": 0.15421139895915986,
      "num_tokens": 11201104.0,
      "step": 6070
    },
    {
      "entropy": 5.781926536560059,
      "epoch": 0.5103969754253308,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004979508127418643,
      "loss": 5.6751,
      "mean_token_accuracy": 0.1530236378312111,
      "num_tokens": 11209578.0,
      "step": 6075
    },
    {
      "entropy": 5.7934770584106445,
      "epoch": 0.5108170552404957,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004979467783054155,
      "loss": 5.6411,
      "mean_token_accuracy": 0.15649499446153642,
      "num_tokens": 11218380.0,
      "step": 6080
    },
    {
      "entropy": 5.741991376876831,
      "epoch": 0.5112371350556606,
      "grad_norm": 1.25,
      "learning_rate": 0.0004979427399195793,
      "loss": 5.6468,
      "mean_token_accuracy": 0.15304491966962813,
      "num_tokens": 11227810.0,
      "step": 6085
    },
    {
      "entropy": 5.781096887588501,
      "epoch": 0.5116572148708255,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004979386975844274,
      "loss": 5.6714,
      "mean_token_accuracy": 0.14935449212789537,
      "num_tokens": 11236631.0,
      "step": 6090
    },
    {
      "entropy": 5.738012218475342,
      "epoch": 0.5120772946859904,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004979346513000311,
      "loss": 5.7303,
      "mean_token_accuracy": 0.14597853422164916,
      "num_tokens": 11247418.0,
      "step": 6095
    },
    {
      "entropy": 5.700581455230713,
      "epoch": 0.5124973745011552,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979306010664623,
      "loss": 5.6292,
      "mean_token_accuracy": 0.1570621207356453,
      "num_tokens": 11256246.0,
      "step": 6100
    },
    {
      "entropy": 5.715303707122803,
      "epoch": 0.5129174543163201,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004979265468837927,
      "loss": 5.6189,
      "mean_token_accuracy": 0.153353688120842,
      "num_tokens": 11265980.0,
      "step": 6105
    },
    {
      "entropy": 5.795161724090576,
      "epoch": 0.513337534131485,
      "grad_norm": 1.25,
      "learning_rate": 0.000497922488752094,
      "loss": 5.6695,
      "mean_token_accuracy": 0.15220673233270646,
      "num_tokens": 11276158.0,
      "step": 6110
    },
    {
      "entropy": 5.751807403564453,
      "epoch": 0.5137576139466499,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004979184266714383,
      "loss": 5.5778,
      "mean_token_accuracy": 0.15824360698461531,
      "num_tokens": 11284957.0,
      "step": 6115
    },
    {
      "entropy": 5.663950729370117,
      "epoch": 0.5141776937618148,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004979143606418974,
      "loss": 5.5969,
      "mean_token_accuracy": 0.15413714349269866,
      "num_tokens": 11294340.0,
      "step": 6120
    },
    {
      "entropy": 5.871919870376587,
      "epoch": 0.5145977735769797,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004979102906635435,
      "loss": 5.8581,
      "mean_token_accuracy": 0.1427499048411846,
      "num_tokens": 11303344.0,
      "step": 6125
    },
    {
      "entropy": 5.876821136474609,
      "epoch": 0.5150178533921445,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004979062167364486,
      "loss": 5.718,
      "mean_token_accuracy": 0.15818116441369057,
      "num_tokens": 11311338.0,
      "step": 6130
    },
    {
      "entropy": 5.703396987915039,
      "epoch": 0.5154379332073094,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004979021388606847,
      "loss": 5.5589,
      "mean_token_accuracy": 0.16222982257604598,
      "num_tokens": 11320194.0,
      "step": 6135
    },
    {
      "entropy": 5.746637916564941,
      "epoch": 0.5158580130224742,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004978980570363243,
      "loss": 5.7316,
      "mean_token_accuracy": 0.1521230459213257,
      "num_tokens": 11329952.0,
      "step": 6140
    },
    {
      "entropy": 5.715192699432373,
      "epoch": 0.5162780928376391,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004978939712634396,
      "loss": 5.6894,
      "mean_token_accuracy": 0.1550623059272766,
      "num_tokens": 11339384.0,
      "step": 6145
    },
    {
      "entropy": 5.8614537715911865,
      "epoch": 0.516698172652804,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004978898815421029,
      "loss": 5.8578,
      "mean_token_accuracy": 0.14535280168056489,
      "num_tokens": 11348409.0,
      "step": 6150
    },
    {
      "entropy": 5.916036462783813,
      "epoch": 0.5171182524679689,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004978857878723867,
      "loss": 5.7757,
      "mean_token_accuracy": 0.1502631425857544,
      "num_tokens": 11357478.0,
      "step": 6155
    },
    {
      "entropy": 5.791261196136475,
      "epoch": 0.5175383322831338,
      "grad_norm": 1.25,
      "learning_rate": 0.0004978816902543636,
      "loss": 5.7553,
      "mean_token_accuracy": 0.15079374462366105,
      "num_tokens": 11366379.0,
      "step": 6160
    },
    {
      "entropy": 5.864686632156372,
      "epoch": 0.5179584120982986,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978775886881062,
      "loss": 5.7858,
      "mean_token_accuracy": 0.14658837914466857,
      "num_tokens": 11376357.0,
      "step": 6165
    },
    {
      "entropy": 5.7728368759155275,
      "epoch": 0.5183784919134635,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000497873483173687,
      "loss": 5.6671,
      "mean_token_accuracy": 0.1539960816502571,
      "num_tokens": 11384995.0,
      "step": 6170
    },
    {
      "entropy": 5.737247657775879,
      "epoch": 0.5187985717286284,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978693737111787,
      "loss": 5.6705,
      "mean_token_accuracy": 0.14912281930446625,
      "num_tokens": 11395363.0,
      "step": 6175
    },
    {
      "entropy": 5.844261121749878,
      "epoch": 0.5192186515437933,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004978652603006543,
      "loss": 5.6595,
      "mean_token_accuracy": 0.14817970544099807,
      "num_tokens": 11404511.0,
      "step": 6180
    },
    {
      "entropy": 5.792019557952881,
      "epoch": 0.5196387313589582,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004978611429421866,
      "loss": 5.706,
      "mean_token_accuracy": 0.14902769923210143,
      "num_tokens": 11413400.0,
      "step": 6185
    },
    {
      "entropy": 5.796417903900147,
      "epoch": 0.5200588111741231,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004978570216358485,
      "loss": 5.7593,
      "mean_token_accuracy": 0.14429776668548583,
      "num_tokens": 11423693.0,
      "step": 6190
    },
    {
      "entropy": 5.864574718475342,
      "epoch": 0.520478890989288,
      "grad_norm": 1.484375,
      "learning_rate": 0.000497852896381713,
      "loss": 5.7084,
      "mean_token_accuracy": 0.14549012184143068,
      "num_tokens": 11433195.0,
      "step": 6195
    },
    {
      "entropy": 5.842799043655395,
      "epoch": 0.5208989708044528,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004978487671798531,
      "loss": 5.8297,
      "mean_token_accuracy": 0.13647647053003312,
      "num_tokens": 11443416.0,
      "step": 6200
    },
    {
      "entropy": 5.879248857498169,
      "epoch": 0.5213190506196177,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004978446340303422,
      "loss": 5.7058,
      "mean_token_accuracy": 0.1528090812265873,
      "num_tokens": 11452487.0,
      "step": 6205
    },
    {
      "entropy": 5.78897385597229,
      "epoch": 0.5217391304347826,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004978404969332533,
      "loss": 5.7399,
      "mean_token_accuracy": 0.1569316253066063,
      "num_tokens": 11461893.0,
      "step": 6210
    },
    {
      "entropy": 5.697704410552978,
      "epoch": 0.5221592102499475,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004978363558886597,
      "loss": 5.6464,
      "mean_token_accuracy": 0.1460499122738838,
      "num_tokens": 11471238.0,
      "step": 6215
    },
    {
      "entropy": 5.79966549873352,
      "epoch": 0.5225792900651124,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004978322108966348,
      "loss": 5.752,
      "mean_token_accuracy": 0.14630683958530427,
      "num_tokens": 11480571.0,
      "step": 6220
    },
    {
      "entropy": 5.799462223052979,
      "epoch": 0.5229993698802773,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004978280619572521,
      "loss": 5.7488,
      "mean_token_accuracy": 0.14263924062252045,
      "num_tokens": 11489552.0,
      "step": 6225
    },
    {
      "entropy": 5.8670618534088135,
      "epoch": 0.5234194496954422,
      "grad_norm": 1.4375,
      "learning_rate": 0.000497823909070585,
      "loss": 5.7876,
      "mean_token_accuracy": 0.14341011941432952,
      "num_tokens": 11498715.0,
      "step": 6230
    },
    {
      "entropy": 5.83455605506897,
      "epoch": 0.523839529510607,
      "grad_norm": 1.375,
      "learning_rate": 0.0004978197522367071,
      "loss": 5.733,
      "mean_token_accuracy": 0.1471528485417366,
      "num_tokens": 11508472.0,
      "step": 6235
    },
    {
      "entropy": 5.869316053390503,
      "epoch": 0.5242596093257719,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004978155914556919,
      "loss": 5.6768,
      "mean_token_accuracy": 0.15946109294891359,
      "num_tokens": 11517620.0,
      "step": 6240
    },
    {
      "entropy": 5.756000518798828,
      "epoch": 0.5246796891409368,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004978114267276134,
      "loss": 5.7077,
      "mean_token_accuracy": 0.14685916304588317,
      "num_tokens": 11526106.0,
      "step": 6245
    },
    {
      "entropy": 5.81321964263916,
      "epoch": 0.5250997689561017,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004978072580525451,
      "loss": 5.7369,
      "mean_token_accuracy": 0.15604251325130464,
      "num_tokens": 11535840.0,
      "step": 6250
    },
    {
      "entropy": 5.809660863876343,
      "epoch": 0.5255198487712666,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000497803085430561,
      "loss": 5.7363,
      "mean_token_accuracy": 0.14954072907567023,
      "num_tokens": 11545110.0,
      "step": 6255
    },
    {
      "entropy": 5.943748712539673,
      "epoch": 0.5259399285864315,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004977989088617349,
      "loss": 5.7661,
      "mean_token_accuracy": 0.14624925330281258,
      "num_tokens": 11554382.0,
      "step": 6260
    },
    {
      "entropy": 5.762158155441284,
      "epoch": 0.5263600084015964,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000497794728346141,
      "loss": 5.6224,
      "mean_token_accuracy": 0.158403742313385,
      "num_tokens": 11562821.0,
      "step": 6265
    },
    {
      "entropy": 5.863851022720337,
      "epoch": 0.5267800882167611,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004977905438838531,
      "loss": 5.8503,
      "mean_token_accuracy": 0.14122226759791373,
      "num_tokens": 11571705.0,
      "step": 6270
    },
    {
      "entropy": 5.719424200057984,
      "epoch": 0.527200168031926,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004977863554749453,
      "loss": 5.6546,
      "mean_token_accuracy": 0.14733370542526245,
      "num_tokens": 11580692.0,
      "step": 6275
    },
    {
      "entropy": 5.734767818450928,
      "epoch": 0.5276202478470909,
      "grad_norm": 1.625,
      "learning_rate": 0.0004977821631194922,
      "loss": 5.6721,
      "mean_token_accuracy": 0.14979589208960534,
      "num_tokens": 11589966.0,
      "step": 6280
    },
    {
      "entropy": 5.853599739074707,
      "epoch": 0.5280403276622558,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004977779668175677,
      "loss": 5.7372,
      "mean_token_accuracy": 0.147355617582798,
      "num_tokens": 11599627.0,
      "step": 6285
    },
    {
      "entropy": 5.828013181686401,
      "epoch": 0.5284604074774207,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004977737665692461,
      "loss": 5.7248,
      "mean_token_accuracy": 0.15135419219732285,
      "num_tokens": 11608431.0,
      "step": 6290
    },
    {
      "entropy": 5.765194272994995,
      "epoch": 0.5288804872925856,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004977695623746021,
      "loss": 5.6002,
      "mean_token_accuracy": 0.1498238652944565,
      "num_tokens": 11617552.0,
      "step": 6295
    },
    {
      "entropy": 5.738640642166137,
      "epoch": 0.5293005671077504,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004977653542337099,
      "loss": 5.6352,
      "mean_token_accuracy": 0.15803537368774415,
      "num_tokens": 11626828.0,
      "step": 6300
    },
    {
      "entropy": 5.746969509124756,
      "epoch": 0.5297206469229153,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004977611421466443,
      "loss": 5.7259,
      "mean_token_accuracy": 0.14914170354604722,
      "num_tokens": 11635867.0,
      "step": 6305
    },
    {
      "entropy": 5.846399927139283,
      "epoch": 0.5301407267380802,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004977569261134797,
      "loss": 5.6433,
      "mean_token_accuracy": 0.1556470736861229,
      "num_tokens": 11644711.0,
      "step": 6310
    },
    {
      "entropy": 5.829954338073731,
      "epoch": 0.5305608065532451,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004977527061342908,
      "loss": 5.7155,
      "mean_token_accuracy": 0.15602043867111207,
      "num_tokens": 11653320.0,
      "step": 6315
    },
    {
      "entropy": 5.789897012710571,
      "epoch": 0.53098088636841,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004977484822091524,
      "loss": 5.6774,
      "mean_token_accuracy": 0.15594944804906846,
      "num_tokens": 11662753.0,
      "step": 6320
    },
    {
      "entropy": 5.8104596614837645,
      "epoch": 0.5314009661835749,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004977442543381394,
      "loss": 5.7164,
      "mean_token_accuracy": 0.14611099809408187,
      "num_tokens": 11671622.0,
      "step": 6325
    },
    {
      "entropy": 5.828975057601928,
      "epoch": 0.5318210459987398,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004977400225213266,
      "loss": 5.7045,
      "mean_token_accuracy": 0.14795105382800103,
      "num_tokens": 11679964.0,
      "step": 6330
    },
    {
      "entropy": 5.745060634613037,
      "epoch": 0.5322411258139046,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000497735786758789,
      "loss": 5.6717,
      "mean_token_accuracy": 0.15165670067071915,
      "num_tokens": 11688700.0,
      "step": 6335
    },
    {
      "entropy": 5.83441858291626,
      "epoch": 0.5326612056290695,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004977315470506016,
      "loss": 5.7835,
      "mean_token_accuracy": 0.14904531762003898,
      "num_tokens": 11698425.0,
      "step": 6340
    },
    {
      "entropy": 5.9017280578613285,
      "epoch": 0.5330812854442344,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004977273033968397,
      "loss": 5.7618,
      "mean_token_accuracy": 0.14042024686932564,
      "num_tokens": 11707705.0,
      "step": 6345
    },
    {
      "entropy": 5.7629967212677,
      "epoch": 0.5335013652593993,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004977230557975782,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15279789865016938,
      "num_tokens": 11717079.0,
      "step": 6350
    },
    {
      "entropy": 5.769342231750488,
      "epoch": 0.5339214450745642,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004977188042528923,
      "loss": 5.6452,
      "mean_token_accuracy": 0.15056213662028312,
      "num_tokens": 11725504.0,
      "step": 6355
    },
    {
      "entropy": 5.820083951950073,
      "epoch": 0.5343415248897291,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004977145487628576,
      "loss": 5.7328,
      "mean_token_accuracy": 0.14796183705329896,
      "num_tokens": 11735282.0,
      "step": 6360
    },
    {
      "entropy": 5.833522939682007,
      "epoch": 0.534761604704894,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004977102893275494,
      "loss": 5.7212,
      "mean_token_accuracy": 0.1473264567553997,
      "num_tokens": 11744827.0,
      "step": 6365
    },
    {
      "entropy": 5.850900793075562,
      "epoch": 0.5351816845200588,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000497706025947043,
      "loss": 5.6722,
      "mean_token_accuracy": 0.15216659232974053,
      "num_tokens": 11753066.0,
      "step": 6370
    },
    {
      "entropy": 5.759478998184204,
      "epoch": 0.5356017643352237,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004977017586214142,
      "loss": 5.6991,
      "mean_token_accuracy": 0.1468222975730896,
      "num_tokens": 11761190.0,
      "step": 6375
    },
    {
      "entropy": 5.775899982452392,
      "epoch": 0.5360218441503886,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004976974873507382,
      "loss": 5.6625,
      "mean_token_accuracy": 0.15434601306915283,
      "num_tokens": 11770321.0,
      "step": 6380
    },
    {
      "entropy": 5.77639217376709,
      "epoch": 0.5364419239655535,
      "grad_norm": 1.234375,
      "learning_rate": 0.000497693212135091,
      "loss": 5.7275,
      "mean_token_accuracy": 0.1511888399720192,
      "num_tokens": 11778388.0,
      "step": 6385
    },
    {
      "entropy": 5.850775814056396,
      "epoch": 0.5368620037807184,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004976889329745482,
      "loss": 5.6031,
      "mean_token_accuracy": 0.15313038155436515,
      "num_tokens": 11786250.0,
      "step": 6390
    },
    {
      "entropy": 5.698163795471191,
      "epoch": 0.5372820835958833,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004976846498691857,
      "loss": 5.5642,
      "mean_token_accuracy": 0.1565365344285965,
      "num_tokens": 11794831.0,
      "step": 6395
    },
    {
      "entropy": 5.7417463779449465,
      "epoch": 0.5377021634110482,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004976803628190792,
      "loss": 5.6156,
      "mean_token_accuracy": 0.15607637539505959,
      "num_tokens": 11803550.0,
      "step": 6400
    },
    {
      "entropy": 5.770711898803711,
      "epoch": 0.5381222432262129,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004976760718243047,
      "loss": 5.6882,
      "mean_token_accuracy": 0.14991564601659774,
      "num_tokens": 11812478.0,
      "step": 6405
    },
    {
      "entropy": 5.792763185501099,
      "epoch": 0.5385423230413778,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004976717768849383,
      "loss": 5.66,
      "mean_token_accuracy": 0.14323782697319984,
      "num_tokens": 11822463.0,
      "step": 6410
    },
    {
      "entropy": 5.790409803390503,
      "epoch": 0.5389624028565427,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004976674780010561,
      "loss": 5.6988,
      "mean_token_accuracy": 0.14533982276916504,
      "num_tokens": 11831853.0,
      "step": 6415
    },
    {
      "entropy": 5.887389659881592,
      "epoch": 0.5393824826717076,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000497663175172734,
      "loss": 5.732,
      "mean_token_accuracy": 0.1425001822412014,
      "num_tokens": 11841574.0,
      "step": 6420
    },
    {
      "entropy": 5.900919437408447,
      "epoch": 0.5398025624868725,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004976588684000486,
      "loss": 5.8292,
      "mean_token_accuracy": 0.13499653488397598,
      "num_tokens": 11852489.0,
      "step": 6425
    },
    {
      "entropy": 5.774700355529785,
      "epoch": 0.5402226423020374,
      "grad_norm": 1.25,
      "learning_rate": 0.0004976545576830759,
      "loss": 5.6866,
      "mean_token_accuracy": 0.14809116050601007,
      "num_tokens": 11861499.0,
      "step": 6430
    },
    {
      "entropy": 5.837946367263794,
      "epoch": 0.5406427221172023,
      "grad_norm": 1.25,
      "learning_rate": 0.0004976502430218924,
      "loss": 5.7586,
      "mean_token_accuracy": 0.1402556501328945,
      "num_tokens": 11871685.0,
      "step": 6435
    },
    {
      "entropy": 5.817818880081177,
      "epoch": 0.5410628019323671,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004976459244165744,
      "loss": 5.6802,
      "mean_token_accuracy": 0.15272180885076522,
      "num_tokens": 11881340.0,
      "step": 6440
    },
    {
      "entropy": 5.70283875465393,
      "epoch": 0.541482881747532,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004976416018671986,
      "loss": 5.6817,
      "mean_token_accuracy": 0.1477773442864418,
      "num_tokens": 11890700.0,
      "step": 6445
    },
    {
      "entropy": 5.776293468475342,
      "epoch": 0.5419029615626969,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004976372753738415,
      "loss": 5.6896,
      "mean_token_accuracy": 0.1432950407266617,
      "num_tokens": 11900329.0,
      "step": 6450
    },
    {
      "entropy": 5.955761337280274,
      "epoch": 0.5423230413778618,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004976329449365795,
      "loss": 5.7217,
      "mean_token_accuracy": 0.14292674511671066,
      "num_tokens": 11909915.0,
      "step": 6455
    },
    {
      "entropy": 5.778019428253174,
      "epoch": 0.5427431211930267,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004976286105554897,
      "loss": 5.7279,
      "mean_token_accuracy": 0.15403613746166228,
      "num_tokens": 11918302.0,
      "step": 6460
    },
    {
      "entropy": 5.734020948410034,
      "epoch": 0.5431632010081916,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004976242722306487,
      "loss": 5.6907,
      "mean_token_accuracy": 0.15065288841724395,
      "num_tokens": 11927794.0,
      "step": 6465
    },
    {
      "entropy": 5.854168796539307,
      "epoch": 0.5435832808233564,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004976199299621333,
      "loss": 5.7169,
      "mean_token_accuracy": 0.1502951353788376,
      "num_tokens": 11937701.0,
      "step": 6470
    },
    {
      "entropy": 5.70092806816101,
      "epoch": 0.5440033606385213,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004976155837500205,
      "loss": 5.6271,
      "mean_token_accuracy": 0.15380519181489943,
      "num_tokens": 11946106.0,
      "step": 6475
    },
    {
      "entropy": 5.770787239074707,
      "epoch": 0.5444234404536862,
      "grad_norm": 1.625,
      "learning_rate": 0.0004976112335943872,
      "loss": 5.5681,
      "mean_token_accuracy": 0.15551166981458664,
      "num_tokens": 11954604.0,
      "step": 6480
    },
    {
      "entropy": 5.674026393890381,
      "epoch": 0.5448435202688511,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004976068794953106,
      "loss": 5.623,
      "mean_token_accuracy": 0.15597020387649535,
      "num_tokens": 11963664.0,
      "step": 6485
    },
    {
      "entropy": 5.792567873001099,
      "epoch": 0.545263600084016,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004976025214528677,
      "loss": 5.6413,
      "mean_token_accuracy": 0.15098657310009003,
      "num_tokens": 11973426.0,
      "step": 6490
    },
    {
      "entropy": 5.726925039291382,
      "epoch": 0.5456836798991809,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004975981594671359,
      "loss": 5.6801,
      "mean_token_accuracy": 0.14519474059343337,
      "num_tokens": 11982339.0,
      "step": 6495
    },
    {
      "entropy": 5.827198791503906,
      "epoch": 0.5461037597143458,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004975937935381921,
      "loss": 5.7182,
      "mean_token_accuracy": 0.1518527202308178,
      "num_tokens": 11992016.0,
      "step": 6500
    },
    {
      "entropy": 5.757827472686768,
      "epoch": 0.5465238395295106,
      "grad_norm": 1.5,
      "learning_rate": 0.000497589423666114,
      "loss": 5.7083,
      "mean_token_accuracy": 0.14981078058481218,
      "num_tokens": 12000616.0,
      "step": 6505
    },
    {
      "entropy": 5.628077507019043,
      "epoch": 0.5469439193446755,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004975850498509789,
      "loss": 5.6115,
      "mean_token_accuracy": 0.155507330596447,
      "num_tokens": 12009717.0,
      "step": 6510
    },
    {
      "entropy": 5.750427150726319,
      "epoch": 0.5473639991598404,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004975806720928642,
      "loss": 5.6897,
      "mean_token_accuracy": 0.15005779415369033,
      "num_tokens": 12018020.0,
      "step": 6515
    },
    {
      "entropy": 5.818803167343139,
      "epoch": 0.5477840789750053,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004975762903918475,
      "loss": 5.6906,
      "mean_token_accuracy": 0.14765678942203522,
      "num_tokens": 12027119.0,
      "step": 6520
    },
    {
      "entropy": 5.7886378288269045,
      "epoch": 0.5482041587901701,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004975719047480064,
      "loss": 5.6594,
      "mean_token_accuracy": 0.1583278015255928,
      "num_tokens": 12035566.0,
      "step": 6525
    },
    {
      "entropy": 5.725280332565307,
      "epoch": 0.548624238605335,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004975675151614187,
      "loss": 5.5804,
      "mean_token_accuracy": 0.1575550004839897,
      "num_tokens": 12044505.0,
      "step": 6530
    },
    {
      "entropy": 5.66980390548706,
      "epoch": 0.5490443184204999,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000497563121632162,
      "loss": 5.6574,
      "mean_token_accuracy": 0.1514866828918457,
      "num_tokens": 12053338.0,
      "step": 6535
    },
    {
      "entropy": 5.768905687332153,
      "epoch": 0.5494643982356647,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004975587241603142,
      "loss": 5.6602,
      "mean_token_accuracy": 0.14860893040895462,
      "num_tokens": 12063235.0,
      "step": 6540
    },
    {
      "entropy": 5.842012786865235,
      "epoch": 0.5498844780508296,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004975543227459533,
      "loss": 5.7328,
      "mean_token_accuracy": 0.14544900879263878,
      "num_tokens": 12072490.0,
      "step": 6545
    },
    {
      "entropy": 5.826513576507568,
      "epoch": 0.5503045578659945,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004975499173891571,
      "loss": 5.7879,
      "mean_token_accuracy": 0.14479437321424485,
      "num_tokens": 12081474.0,
      "step": 6550
    },
    {
      "entropy": 5.7784984588623045,
      "epoch": 0.5507246376811594,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004975455080900037,
      "loss": 5.643,
      "mean_token_accuracy": 0.15413787513971328,
      "num_tokens": 12090963.0,
      "step": 6555
    },
    {
      "entropy": 5.757335090637207,
      "epoch": 0.5511447174963243,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004975410948485713,
      "loss": 5.6689,
      "mean_token_accuracy": 0.14914965778589248,
      "num_tokens": 12099786.0,
      "step": 6560
    },
    {
      "entropy": 5.719509267807007,
      "epoch": 0.5515647973114892,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004975366776649379,
      "loss": 5.6816,
      "mean_token_accuracy": 0.15208746641874313,
      "num_tokens": 12108469.0,
      "step": 6565
    },
    {
      "entropy": 5.7668781757354735,
      "epoch": 0.5519848771266541,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004975322565391818,
      "loss": 5.6629,
      "mean_token_accuracy": 0.15438627898693086,
      "num_tokens": 12118287.0,
      "step": 6570
    },
    {
      "entropy": 5.862242794036865,
      "epoch": 0.5524049569418189,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004975278314713814,
      "loss": 5.8088,
      "mean_token_accuracy": 0.1440727099776268,
      "num_tokens": 12127122.0,
      "step": 6575
    },
    {
      "entropy": 5.806251859664917,
      "epoch": 0.5528250367569838,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004975234024616152,
      "loss": 5.6977,
      "mean_token_accuracy": 0.15571955889463424,
      "num_tokens": 12136395.0,
      "step": 6580
    },
    {
      "entropy": 5.695988988876342,
      "epoch": 0.5532451165721487,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004975189695099613,
      "loss": 5.683,
      "mean_token_accuracy": 0.1485990047454834,
      "num_tokens": 12145025.0,
      "step": 6585
    },
    {
      "entropy": 5.828900909423828,
      "epoch": 0.5536651963873136,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004975145326164985,
      "loss": 5.7345,
      "mean_token_accuracy": 0.1466663308441639,
      "num_tokens": 12154352.0,
      "step": 6590
    },
    {
      "entropy": 5.78888168334961,
      "epoch": 0.5540852762024785,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004975100917813055,
      "loss": 5.6427,
      "mean_token_accuracy": 0.1504230782389641,
      "num_tokens": 12163802.0,
      "step": 6595
    },
    {
      "entropy": 5.719293403625488,
      "epoch": 0.5545053560176434,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004975056470044606,
      "loss": 5.6601,
      "mean_token_accuracy": 0.15201830267906188,
      "num_tokens": 12173111.0,
      "step": 6600
    },
    {
      "entropy": 5.788504505157471,
      "epoch": 0.5549254358328082,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004975011982860428,
      "loss": 5.7128,
      "mean_token_accuracy": 0.15041669309139252,
      "num_tokens": 12182048.0,
      "step": 6605
    },
    {
      "entropy": 5.795574474334717,
      "epoch": 0.5553455156479731,
      "grad_norm": 1.75,
      "learning_rate": 0.0004974967456261309,
      "loss": 5.6928,
      "mean_token_accuracy": 0.1525782212615013,
      "num_tokens": 12191501.0,
      "step": 6610
    },
    {
      "entropy": 5.853242444992065,
      "epoch": 0.555765595463138,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004974922890248036,
      "loss": 5.7179,
      "mean_token_accuracy": 0.15184747502207757,
      "num_tokens": 12201132.0,
      "step": 6615
    },
    {
      "entropy": 5.852621126174927,
      "epoch": 0.5561856752783029,
      "grad_norm": 1.40625,
      "learning_rate": 0.00049748782848214,
      "loss": 5.8361,
      "mean_token_accuracy": 0.14677973017096518,
      "num_tokens": 12211082.0,
      "step": 6620
    },
    {
      "entropy": 5.785615491867065,
      "epoch": 0.5566057550934678,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004974833639982192,
      "loss": 5.6457,
      "mean_token_accuracy": 0.15385498329997063,
      "num_tokens": 12219946.0,
      "step": 6625
    },
    {
      "entropy": 5.848658609390259,
      "epoch": 0.5570258349086327,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00049747889557312,
      "loss": 5.7623,
      "mean_token_accuracy": 0.14732073992490768,
      "num_tokens": 12229668.0,
      "step": 6630
    },
    {
      "entropy": 5.834868860244751,
      "epoch": 0.5574459147237976,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004974744232069219,
      "loss": 5.756,
      "mean_token_accuracy": 0.15054013729095458,
      "num_tokens": 12238750.0,
      "step": 6635
    },
    {
      "entropy": 5.799629974365234,
      "epoch": 0.5578659945389624,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004974699468997038,
      "loss": 5.6952,
      "mean_token_accuracy": 0.14818740338087083,
      "num_tokens": 12246825.0,
      "step": 6640
    },
    {
      "entropy": 5.715137672424317,
      "epoch": 0.5582860743541272,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004974654666515452,
      "loss": 5.6398,
      "mean_token_accuracy": 0.14599109143018724,
      "num_tokens": 12256413.0,
      "step": 6645
    },
    {
      "entropy": 5.772627592086792,
      "epoch": 0.5587061541692921,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004974609824625254,
      "loss": 5.6647,
      "mean_token_accuracy": 0.15836252719163896,
      "num_tokens": 12265458.0,
      "step": 6650
    },
    {
      "entropy": 5.642200517654419,
      "epoch": 0.559126233984457,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004974564943327239,
      "loss": 5.6032,
      "mean_token_accuracy": 0.1530925676226616,
      "num_tokens": 12274124.0,
      "step": 6655
    },
    {
      "entropy": 5.643485116958618,
      "epoch": 0.5595463137996219,
      "grad_norm": 1.3125,
      "learning_rate": 0.00049745200226222,
      "loss": 5.5692,
      "mean_token_accuracy": 0.1617315962910652,
      "num_tokens": 12283513.0,
      "step": 6660
    },
    {
      "entropy": 5.769548034667968,
      "epoch": 0.5599663936147868,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004974475062510936,
      "loss": 5.6941,
      "mean_token_accuracy": 0.15532324314117432,
      "num_tokens": 12292396.0,
      "step": 6665
    },
    {
      "entropy": 5.82047667503357,
      "epoch": 0.5603864734299517,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004974430062994242,
      "loss": 5.7225,
      "mean_token_accuracy": 0.15115630030632018,
      "num_tokens": 12301604.0,
      "step": 6670
    },
    {
      "entropy": 5.886059427261353,
      "epoch": 0.5608065532451165,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004974385024072912,
      "loss": 5.7476,
      "mean_token_accuracy": 0.14723663330078124,
      "num_tokens": 12310458.0,
      "step": 6675
    },
    {
      "entropy": 5.850595569610595,
      "epoch": 0.5612266330602814,
      "grad_norm": 1.296875,
      "learning_rate": 0.000497433994574775,
      "loss": 5.7541,
      "mean_token_accuracy": 0.14729131162166595,
      "num_tokens": 12319620.0,
      "step": 6680
    },
    {
      "entropy": 5.768581199645996,
      "epoch": 0.5616467128754463,
      "grad_norm": 1.296875,
      "learning_rate": 0.000497429482801955,
      "loss": 5.8087,
      "mean_token_accuracy": 0.14342449381947517,
      "num_tokens": 12329518.0,
      "step": 6685
    },
    {
      "entropy": 5.740103340148925,
      "epoch": 0.5620667926906112,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004974249670889111,
      "loss": 5.6224,
      "mean_token_accuracy": 0.14958661496639253,
      "num_tokens": 12338244.0,
      "step": 6690
    },
    {
      "entropy": 5.922348546981811,
      "epoch": 0.5624868725057761,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004974204474357237,
      "loss": 5.8038,
      "mean_token_accuracy": 0.14635297060012817,
      "num_tokens": 12347962.0,
      "step": 6695
    },
    {
      "entropy": 5.858186721801758,
      "epoch": 0.562906952320941,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004974159238424723,
      "loss": 5.7152,
      "mean_token_accuracy": 0.14685503840446473,
      "num_tokens": 12357020.0,
      "step": 6700
    },
    {
      "entropy": 5.7741382122039795,
      "epoch": 0.5633270321361059,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004974113963092376,
      "loss": 5.6818,
      "mean_token_accuracy": 0.15062317550182341,
      "num_tokens": 12366108.0,
      "step": 6705
    },
    {
      "entropy": 5.800469636917114,
      "epoch": 0.5637471119512707,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004974068648360995,
      "loss": 5.6174,
      "mean_token_accuracy": 0.16103638261556624,
      "num_tokens": 12374508.0,
      "step": 6710
    },
    {
      "entropy": 5.716407442092896,
      "epoch": 0.5641671917664356,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004974023294231383,
      "loss": 5.6307,
      "mean_token_accuracy": 0.15565890967845916,
      "num_tokens": 12383555.0,
      "step": 6715
    },
    {
      "entropy": 5.739489316940308,
      "epoch": 0.5645872715816005,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004973977900704342,
      "loss": 5.7318,
      "mean_token_accuracy": 0.14744656831026076,
      "num_tokens": 12392680.0,
      "step": 6720
    },
    {
      "entropy": 5.8620096206665036,
      "epoch": 0.5650073513967654,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004973932467780679,
      "loss": 5.7678,
      "mean_token_accuracy": 0.14722730666399003,
      "num_tokens": 12401881.0,
      "step": 6725
    },
    {
      "entropy": 5.841188859939575,
      "epoch": 0.5654274312119303,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004973886995461197,
      "loss": 5.7751,
      "mean_token_accuracy": 0.14041661322116852,
      "num_tokens": 12411487.0,
      "step": 6730
    },
    {
      "entropy": 5.73593955039978,
      "epoch": 0.5658475110270952,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004973841483746703,
      "loss": 5.5558,
      "mean_token_accuracy": 0.1630205363035202,
      "num_tokens": 12420376.0,
      "step": 6735
    },
    {
      "entropy": 5.657348871231079,
      "epoch": 0.5662675908422601,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004973795932638001,
      "loss": 5.5997,
      "mean_token_accuracy": 0.1595743790268898,
      "num_tokens": 12429518.0,
      "step": 6740
    },
    {
      "entropy": 5.758114671707153,
      "epoch": 0.5666876706574249,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00049737503421359,
      "loss": 5.591,
      "mean_token_accuracy": 0.15658295452594756,
      "num_tokens": 12438952.0,
      "step": 6745
    },
    {
      "entropy": 5.740869188308716,
      "epoch": 0.5671077504725898,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004973704712241206,
      "loss": 5.6004,
      "mean_token_accuracy": 0.15196311324834824,
      "num_tokens": 12448576.0,
      "step": 6750
    },
    {
      "entropy": 5.697316026687622,
      "epoch": 0.5675278302877547,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004973659042954729,
      "loss": 5.651,
      "mean_token_accuracy": 0.1531267300248146,
      "num_tokens": 12458166.0,
      "step": 6755
    },
    {
      "entropy": 5.621774578094483,
      "epoch": 0.5679479101029196,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004973613334277277,
      "loss": 5.5773,
      "mean_token_accuracy": 0.15975697934627534,
      "num_tokens": 12467271.0,
      "step": 6760
    },
    {
      "entropy": 5.788559246063232,
      "epoch": 0.5683679899180845,
      "grad_norm": 1.375,
      "learning_rate": 0.0004973567586209658,
      "loss": 5.7353,
      "mean_token_accuracy": 0.14790709912776948,
      "num_tokens": 12476255.0,
      "step": 6765
    },
    {
      "entropy": 5.805595254898071,
      "epoch": 0.5687880697332494,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004973521798752686,
      "loss": 5.6924,
      "mean_token_accuracy": 0.14894914180040358,
      "num_tokens": 12485096.0,
      "step": 6770
    },
    {
      "entropy": 5.907484722137451,
      "epoch": 0.5692081495484141,
      "grad_norm": 1.484375,
      "learning_rate": 0.000497347597190717,
      "loss": 5.7415,
      "mean_token_accuracy": 0.1513557866215706,
      "num_tokens": 12494405.0,
      "step": 6775
    },
    {
      "entropy": 5.706912994384766,
      "epoch": 0.569628229363579,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004973430105673921,
      "loss": 5.664,
      "mean_token_accuracy": 0.14975441992282867,
      "num_tokens": 12503349.0,
      "step": 6780
    },
    {
      "entropy": 5.7776655673980715,
      "epoch": 0.5700483091787439,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004973384200053754,
      "loss": 5.7322,
      "mean_token_accuracy": 0.14976128786802292,
      "num_tokens": 12513122.0,
      "step": 6785
    },
    {
      "entropy": 5.756508922576904,
      "epoch": 0.5704683889939088,
      "grad_norm": 1.40625,
      "learning_rate": 0.000497333825504748,
      "loss": 5.6834,
      "mean_token_accuracy": 0.15084973052144052,
      "num_tokens": 12523614.0,
      "step": 6790
    },
    {
      "entropy": 5.739598941802979,
      "epoch": 0.5708884688090737,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004973292270655914,
      "loss": 5.7215,
      "mean_token_accuracy": 0.1441544845700264,
      "num_tokens": 12532031.0,
      "step": 6795
    },
    {
      "entropy": 5.866616153717041,
      "epoch": 0.5713085486242386,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000497324624687987,
      "loss": 5.8053,
      "mean_token_accuracy": 0.14371495842933654,
      "num_tokens": 12542239.0,
      "step": 6800
    },
    {
      "entropy": 5.8771069049835205,
      "epoch": 0.5717286284394035,
      "grad_norm": 1.375,
      "learning_rate": 0.0004973200183720164,
      "loss": 5.7128,
      "mean_token_accuracy": 0.14354896992444993,
      "num_tokens": 12552608.0,
      "step": 6805
    },
    {
      "entropy": 5.6871239185333256,
      "epoch": 0.5721487082545683,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004973154081177611,
      "loss": 5.5667,
      "mean_token_accuracy": 0.15022857040166854,
      "num_tokens": 12562020.0,
      "step": 6810
    },
    {
      "entropy": 5.765218448638916,
      "epoch": 0.5725687880697332,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004973107939253027,
      "loss": 5.6358,
      "mean_token_accuracy": 0.16130239069461821,
      "num_tokens": 12570519.0,
      "step": 6815
    },
    {
      "entropy": 5.67629337310791,
      "epoch": 0.5729888678848981,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004973061757947233,
      "loss": 5.6383,
      "mean_token_accuracy": 0.15358631312847137,
      "num_tokens": 12579324.0,
      "step": 6820
    },
    {
      "entropy": 5.70035400390625,
      "epoch": 0.573408947700063,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004973015537261043,
      "loss": 5.7002,
      "mean_token_accuracy": 0.15523785203695298,
      "num_tokens": 12588014.0,
      "step": 6825
    },
    {
      "entropy": 5.787239074707031,
      "epoch": 0.5738290275152279,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004972969277195279,
      "loss": 5.6983,
      "mean_token_accuracy": 0.15244747400283815,
      "num_tokens": 12596882.0,
      "step": 6830
    },
    {
      "entropy": 5.7592627048492435,
      "epoch": 0.5742491073303928,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004972922977750757,
      "loss": 5.6187,
      "mean_token_accuracy": 0.15164665579795839,
      "num_tokens": 12606069.0,
      "step": 6835
    },
    {
      "entropy": 5.75448842048645,
      "epoch": 0.5746691871455577,
      "grad_norm": 1.9375,
      "learning_rate": 0.00049728766389283,
      "loss": 5.6692,
      "mean_token_accuracy": 0.1461893856525421,
      "num_tokens": 12615167.0,
      "step": 6840
    },
    {
      "entropy": 5.7733536720275875,
      "epoch": 0.5750892669607225,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004972830260728729,
      "loss": 5.6936,
      "mean_token_accuracy": 0.15012103021144868,
      "num_tokens": 12624230.0,
      "step": 6845
    },
    {
      "entropy": 5.8393168449401855,
      "epoch": 0.5755093467758874,
      "grad_norm": 1.5,
      "learning_rate": 0.0004972783843152863,
      "loss": 5.6759,
      "mean_token_accuracy": 0.15672653466463088,
      "num_tokens": 12633158.0,
      "step": 6850
    },
    {
      "entropy": 5.7473838329315186,
      "epoch": 0.5759294265910523,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004972737386201527,
      "loss": 5.6159,
      "mean_token_accuracy": 0.1514609858393669,
      "num_tokens": 12641465.0,
      "step": 6855
    },
    {
      "entropy": 5.694714307785034,
      "epoch": 0.5763495064062172,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004972690889875541,
      "loss": 5.576,
      "mean_token_accuracy": 0.1582634076476097,
      "num_tokens": 12650437.0,
      "step": 6860
    },
    {
      "entropy": 5.887161779403686,
      "epoch": 0.5767695862213821,
      "grad_norm": 1.25,
      "learning_rate": 0.0004972644354175732,
      "loss": 5.8013,
      "mean_token_accuracy": 0.1454438552260399,
      "num_tokens": 12660072.0,
      "step": 6865
    },
    {
      "entropy": 5.8891956329345705,
      "epoch": 0.577189666036547,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004972597779102922,
      "loss": 5.8047,
      "mean_token_accuracy": 0.1495344288647175,
      "num_tokens": 12670405.0,
      "step": 6870
    },
    {
      "entropy": 5.794363021850586,
      "epoch": 0.5776097458517119,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004972551164657937,
      "loss": 5.6863,
      "mean_token_accuracy": 0.1481093443930149,
      "num_tokens": 12679992.0,
      "step": 6875
    },
    {
      "entropy": 5.821959114074707,
      "epoch": 0.5780298256668767,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004972504510841602,
      "loss": 5.741,
      "mean_token_accuracy": 0.14912082627415657,
      "num_tokens": 12690289.0,
      "step": 6880
    },
    {
      "entropy": 5.85949215888977,
      "epoch": 0.5784499054820416,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004972457817654745,
      "loss": 5.7464,
      "mean_token_accuracy": 0.14608799815177917,
      "num_tokens": 12700518.0,
      "step": 6885
    },
    {
      "entropy": 5.854418659210205,
      "epoch": 0.5788699852972065,
      "grad_norm": 1.375,
      "learning_rate": 0.0004972411085098191,
      "loss": 5.8047,
      "mean_token_accuracy": 0.13859488815069199,
      "num_tokens": 12710603.0,
      "step": 6890
    },
    {
      "entropy": 5.80083737373352,
      "epoch": 0.5792900651123714,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000497236431317277,
      "loss": 5.6832,
      "mean_token_accuracy": 0.15335874259471893,
      "num_tokens": 12719298.0,
      "step": 6895
    },
    {
      "entropy": 5.793888568878174,
      "epoch": 0.5797101449275363,
      "grad_norm": 1.4375,
      "learning_rate": 0.000497231750187931,
      "loss": 5.679,
      "mean_token_accuracy": 0.14987643510103227,
      "num_tokens": 12728368.0,
      "step": 6900
    },
    {
      "entropy": 5.803576946258545,
      "epoch": 0.5801302247427012,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004972270651218638,
      "loss": 5.7501,
      "mean_token_accuracy": 0.15091048330068588,
      "num_tokens": 12737898.0,
      "step": 6905
    },
    {
      "entropy": 5.823219203948975,
      "epoch": 0.580550304557866,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004972223761191587,
      "loss": 5.6763,
      "mean_token_accuracy": 0.14775750860571862,
      "num_tokens": 12746761.0,
      "step": 6910
    },
    {
      "entropy": 5.743349599838257,
      "epoch": 0.5809703843730308,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004972176831798986,
      "loss": 5.598,
      "mean_token_accuracy": 0.16036793515086173,
      "num_tokens": 12755128.0,
      "step": 6915
    },
    {
      "entropy": 5.7417994976043705,
      "epoch": 0.5813904641881957,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004972129863041667,
      "loss": 5.7851,
      "mean_token_accuracy": 0.14712280929088592,
      "num_tokens": 12764727.0,
      "step": 6920
    },
    {
      "entropy": 5.813629150390625,
      "epoch": 0.5818105440033606,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004972082854920462,
      "loss": 5.6555,
      "mean_token_accuracy": 0.15503355711698533,
      "num_tokens": 12773557.0,
      "step": 6925
    },
    {
      "entropy": 5.765597820281982,
      "epoch": 0.5822306238185255,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004972035807436203,
      "loss": 5.6281,
      "mean_token_accuracy": 0.15963127166032792,
      "num_tokens": 12782525.0,
      "step": 6930
    },
    {
      "entropy": 5.826959037780762,
      "epoch": 0.5826507036336904,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004971988720589723,
      "loss": 5.7343,
      "mean_token_accuracy": 0.14858472794294358,
      "num_tokens": 12791534.0,
      "step": 6935
    },
    {
      "entropy": 5.811407089233398,
      "epoch": 0.5830707834488553,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004971941594381858,
      "loss": 5.6378,
      "mean_token_accuracy": 0.15586716905236245,
      "num_tokens": 12800662.0,
      "step": 6940
    },
    {
      "entropy": 5.8181867599487305,
      "epoch": 0.5834908632640201,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004971894428813441,
      "loss": 5.6793,
      "mean_token_accuracy": 0.15237494111061095,
      "num_tokens": 12809440.0,
      "step": 6945
    },
    {
      "entropy": 5.825567674636841,
      "epoch": 0.583910943079185,
      "grad_norm": 1.4375,
      "learning_rate": 0.000497184722388531,
      "loss": 5.7669,
      "mean_token_accuracy": 0.14334406405687333,
      "num_tokens": 12818560.0,
      "step": 6950
    },
    {
      "entropy": 5.836616373062133,
      "epoch": 0.5843310228943499,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004971799979598297,
      "loss": 5.7006,
      "mean_token_accuracy": 0.14847517311573027,
      "num_tokens": 12827898.0,
      "step": 6955
    },
    {
      "entropy": 5.705964803695679,
      "epoch": 0.5847511027095148,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004971752695953243,
      "loss": 5.6278,
      "mean_token_accuracy": 0.15511309504508972,
      "num_tokens": 12837199.0,
      "step": 6960
    },
    {
      "entropy": 5.776910972595215,
      "epoch": 0.5851711825246797,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004971705372950984,
      "loss": 5.6685,
      "mean_token_accuracy": 0.15222510397434236,
      "num_tokens": 12846493.0,
      "step": 6965
    },
    {
      "entropy": 5.805812168121338,
      "epoch": 0.5855912623398446,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004971658010592358,
      "loss": 5.661,
      "mean_token_accuracy": 0.14836213737726212,
      "num_tokens": 12855026.0,
      "step": 6970
    },
    {
      "entropy": 5.7935222625732425,
      "epoch": 0.5860113421550095,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004971610608878205,
      "loss": 5.7399,
      "mean_token_accuracy": 0.14885252118110656,
      "num_tokens": 12864563.0,
      "step": 6975
    },
    {
      "entropy": 5.890350008010865,
      "epoch": 0.5864314219701743,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004971563167809363,
      "loss": 5.7027,
      "mean_token_accuracy": 0.15145302265882493,
      "num_tokens": 12874358.0,
      "step": 6980
    },
    {
      "entropy": 5.775705432891845,
      "epoch": 0.5868515017853392,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004971515687386674,
      "loss": 5.6936,
      "mean_token_accuracy": 0.14974414557218552,
      "num_tokens": 12883110.0,
      "step": 6985
    },
    {
      "entropy": 5.7816088676452635,
      "epoch": 0.5872715816005041,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004971468167610978,
      "loss": 5.7569,
      "mean_token_accuracy": 0.15565478801727295,
      "num_tokens": 12892977.0,
      "step": 6990
    },
    {
      "entropy": 5.780284309387207,
      "epoch": 0.587691661415669,
      "grad_norm": 1.25,
      "learning_rate": 0.0004971420608483117,
      "loss": 5.557,
      "mean_token_accuracy": 0.15739939212799073,
      "num_tokens": 12902327.0,
      "step": 6995
    },
    {
      "entropy": 5.6380256652832035,
      "epoch": 0.5881117412308339,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004971373010003936,
      "loss": 5.5792,
      "mean_token_accuracy": 0.16282830759882927,
      "num_tokens": 12911957.0,
      "step": 7000
    },
    {
      "entropy": 5.734653520584106,
      "epoch": 0.5885318210459988,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004971325372174274,
      "loss": 5.6761,
      "mean_token_accuracy": 0.14648626297712325,
      "num_tokens": 12920380.0,
      "step": 7005
    },
    {
      "entropy": 5.786434602737427,
      "epoch": 0.5889519008611637,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004971277694994976,
      "loss": 5.7325,
      "mean_token_accuracy": 0.15107914805412292,
      "num_tokens": 12929670.0,
      "step": 7010
    },
    {
      "entropy": 5.806161260604858,
      "epoch": 0.5893719806763285,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000497122997846689,
      "loss": 5.6395,
      "mean_token_accuracy": 0.15913416296243668,
      "num_tokens": 12938185.0,
      "step": 7015
    },
    {
      "entropy": 5.785260486602783,
      "epoch": 0.5897920604914934,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004971182222590857,
      "loss": 5.6737,
      "mean_token_accuracy": 0.15860777348279953,
      "num_tokens": 12947706.0,
      "step": 7020
    },
    {
      "entropy": 5.699825477600098,
      "epoch": 0.5902121403066583,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004971134427367725,
      "loss": 5.6616,
      "mean_token_accuracy": 0.15406165570020675,
      "num_tokens": 12957393.0,
      "step": 7025
    },
    {
      "entropy": 5.792528390884399,
      "epoch": 0.5906322201218231,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000497108659279834,
      "loss": 5.5571,
      "mean_token_accuracy": 0.1610640689730644,
      "num_tokens": 12967165.0,
      "step": 7030
    },
    {
      "entropy": 5.813483619689942,
      "epoch": 0.591052299936988,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004971038718883551,
      "loss": 5.6962,
      "mean_token_accuracy": 0.144030924141407,
      "num_tokens": 12976490.0,
      "step": 7035
    },
    {
      "entropy": 5.810479640960693,
      "epoch": 0.5914723797521529,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004970990805624203,
      "loss": 5.6798,
      "mean_token_accuracy": 0.14762643873691558,
      "num_tokens": 12985423.0,
      "step": 7040
    },
    {
      "entropy": 5.7493095874786375,
      "epoch": 0.5918924595673178,
      "grad_norm": 1.375,
      "learning_rate": 0.0004970942853021147,
      "loss": 5.586,
      "mean_token_accuracy": 0.15507588982582093,
      "num_tokens": 12994510.0,
      "step": 7045
    },
    {
      "entropy": 5.759291505813598,
      "epoch": 0.5923125393824826,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004970894861075232,
      "loss": 5.6933,
      "mean_token_accuracy": 0.15216402411460878,
      "num_tokens": 13003383.0,
      "step": 7050
    },
    {
      "entropy": 5.7968847274780275,
      "epoch": 0.5927326191976475,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004970846829787309,
      "loss": 5.6424,
      "mean_token_accuracy": 0.15637523382902146,
      "num_tokens": 13012550.0,
      "step": 7055
    },
    {
      "entropy": 5.814409351348877,
      "epoch": 0.5931526990128124,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004970798759158227,
      "loss": 5.717,
      "mean_token_accuracy": 0.14880919829010963,
      "num_tokens": 13022066.0,
      "step": 7060
    },
    {
      "entropy": 5.810356950759887,
      "epoch": 0.5935727788279773,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004970750649188839,
      "loss": 5.6906,
      "mean_token_accuracy": 0.1568908266723156,
      "num_tokens": 13031008.0,
      "step": 7065
    },
    {
      "entropy": 5.721108341217041,
      "epoch": 0.5939928586431422,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004970702499879998,
      "loss": 5.6632,
      "mean_token_accuracy": 0.15371381118893623,
      "num_tokens": 13040366.0,
      "step": 7070
    },
    {
      "entropy": 5.732092952728271,
      "epoch": 0.5944129384583071,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004970654311232554,
      "loss": 5.6904,
      "mean_token_accuracy": 0.15139214396476747,
      "num_tokens": 13051140.0,
      "step": 7075
    },
    {
      "entropy": 5.8530854225158695,
      "epoch": 0.594833018273472,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004970606083247362,
      "loss": 5.6124,
      "mean_token_accuracy": 0.1549868643283844,
      "num_tokens": 13059835.0,
      "step": 7080
    },
    {
      "entropy": 5.682567930221557,
      "epoch": 0.5952530980886368,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004970557815925278,
      "loss": 5.5625,
      "mean_token_accuracy": 0.15498380362987518,
      "num_tokens": 13068909.0,
      "step": 7085
    },
    {
      "entropy": 5.665302896499634,
      "epoch": 0.5956731779038017,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004970509509267155,
      "loss": 5.6395,
      "mean_token_accuracy": 0.15328293964266776,
      "num_tokens": 13078380.0,
      "step": 7090
    },
    {
      "entropy": 5.856351041793824,
      "epoch": 0.5960932577189666,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004970461163273849,
      "loss": 5.6873,
      "mean_token_accuracy": 0.15368358492851258,
      "num_tokens": 13087774.0,
      "step": 7095
    },
    {
      "entropy": 5.7456623077392575,
      "epoch": 0.5965133375341315,
      "grad_norm": 1.75,
      "learning_rate": 0.0004970412777946219,
      "loss": 5.5125,
      "mean_token_accuracy": 0.1564317002892494,
      "num_tokens": 13095938.0,
      "step": 7100
    },
    {
      "entropy": 5.6623311042785645,
      "epoch": 0.5969334173492964,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004970364353285117,
      "loss": 5.6649,
      "mean_token_accuracy": 0.15760405361652374,
      "num_tokens": 13104661.0,
      "step": 7105
    },
    {
      "entropy": 5.799459838867188,
      "epoch": 0.5973534971644613,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004970315889291405,
      "loss": 5.6498,
      "mean_token_accuracy": 0.1500176966190338,
      "num_tokens": 13114505.0,
      "step": 7110
    },
    {
      "entropy": 5.67414870262146,
      "epoch": 0.5977735769796261,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004970267385965941,
      "loss": 5.5967,
      "mean_token_accuracy": 0.15627617239952088,
      "num_tokens": 13124590.0,
      "step": 7115
    },
    {
      "entropy": 5.709624147415161,
      "epoch": 0.598193656794791,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004970218843309583,
      "loss": 5.5727,
      "mean_token_accuracy": 0.16128048151731492,
      "num_tokens": 13134026.0,
      "step": 7120
    },
    {
      "entropy": 5.858076572418213,
      "epoch": 0.5986137366099559,
      "grad_norm": 1.375,
      "learning_rate": 0.0004970170261323192,
      "loss": 5.7412,
      "mean_token_accuracy": 0.15393318980932236,
      "num_tokens": 13142654.0,
      "step": 7125
    },
    {
      "entropy": 5.691959428787231,
      "epoch": 0.5990338164251208,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004970121640007627,
      "loss": 5.6406,
      "mean_token_accuracy": 0.1516393780708313,
      "num_tokens": 13151177.0,
      "step": 7130
    },
    {
      "entropy": 5.724456405639648,
      "epoch": 0.5994538962402857,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004970072979363751,
      "loss": 5.6415,
      "mean_token_accuracy": 0.15101922899484635,
      "num_tokens": 13159689.0,
      "step": 7135
    },
    {
      "entropy": 5.751378107070923,
      "epoch": 0.5998739760554506,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004970024279392425,
      "loss": 5.6808,
      "mean_token_accuracy": 0.14915067553520203,
      "num_tokens": 13168601.0,
      "step": 7140
    },
    {
      "entropy": 5.811783504486084,
      "epoch": 0.6002940558706155,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004969975540094513,
      "loss": 5.65,
      "mean_token_accuracy": 0.15383494645357132,
      "num_tokens": 13177035.0,
      "step": 7145
    },
    {
      "entropy": 5.808345937728882,
      "epoch": 0.6007141356857802,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004969926761470876,
      "loss": 5.6175,
      "mean_token_accuracy": 0.16169959604740142,
      "num_tokens": 13185444.0,
      "step": 7150
    },
    {
      "entropy": 5.739962339401245,
      "epoch": 0.6011342155009451,
      "grad_norm": 1.375,
      "learning_rate": 0.000496987794352238,
      "loss": 5.6269,
      "mean_token_accuracy": 0.158029805123806,
      "num_tokens": 13194987.0,
      "step": 7155
    },
    {
      "entropy": 5.7030730724334715,
      "epoch": 0.60155429531611,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004969829086249889,
      "loss": 5.6506,
      "mean_token_accuracy": 0.14698696434497832,
      "num_tokens": 13203807.0,
      "step": 7160
    },
    {
      "entropy": 5.818206834793091,
      "epoch": 0.6019743751312749,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000496978018965427,
      "loss": 5.759,
      "mean_token_accuracy": 0.14981004670262338,
      "num_tokens": 13214362.0,
      "step": 7165
    },
    {
      "entropy": 5.880270910263062,
      "epoch": 0.6023944549464398,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004969731253736387,
      "loss": 5.7679,
      "mean_token_accuracy": 0.1471377916634083,
      "num_tokens": 13224192.0,
      "step": 7170
    },
    {
      "entropy": 5.780945491790772,
      "epoch": 0.6028145347616047,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004969682278497109,
      "loss": 5.7083,
      "mean_token_accuracy": 0.15497646927833558,
      "num_tokens": 13234430.0,
      "step": 7175
    },
    {
      "entropy": 5.719972944259643,
      "epoch": 0.6032346145767696,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004969633263937301,
      "loss": 5.6096,
      "mean_token_accuracy": 0.15373648703098297,
      "num_tokens": 13243681.0,
      "step": 7180
    },
    {
      "entropy": 5.953779697418213,
      "epoch": 0.6036546943919344,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004969584210057832,
      "loss": 5.9049,
      "mean_token_accuracy": 0.14223207384347916,
      "num_tokens": 13254334.0,
      "step": 7185
    },
    {
      "entropy": 5.90763521194458,
      "epoch": 0.6040747742070993,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004969535116859573,
      "loss": 5.69,
      "mean_token_accuracy": 0.15484999641776084,
      "num_tokens": 13263781.0,
      "step": 7190
    },
    {
      "entropy": 5.698655986785889,
      "epoch": 0.6044948540222642,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004969485984343392,
      "loss": 5.623,
      "mean_token_accuracy": 0.153725266456604,
      "num_tokens": 13272831.0,
      "step": 7195
    },
    {
      "entropy": 5.807196807861328,
      "epoch": 0.6049149338374291,
      "grad_norm": 1.5,
      "learning_rate": 0.000496943681251016,
      "loss": 5.6458,
      "mean_token_accuracy": 0.15263475701212884,
      "num_tokens": 13281621.0,
      "step": 7200
    },
    {
      "entropy": 5.737537956237793,
      "epoch": 0.605335013652594,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004969387601360747,
      "loss": 5.6339,
      "mean_token_accuracy": 0.1460575617849827,
      "num_tokens": 13291021.0,
      "step": 7205
    },
    {
      "entropy": 5.792116641998291,
      "epoch": 0.6057550934677589,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004969338350896026,
      "loss": 5.6657,
      "mean_token_accuracy": 0.15297241359949112,
      "num_tokens": 13299752.0,
      "step": 7210
    },
    {
      "entropy": 5.843834066390992,
      "epoch": 0.6061751732829238,
      "grad_norm": 1.5,
      "learning_rate": 0.0004969289061116869,
      "loss": 5.692,
      "mean_token_accuracy": 0.14698407873511316,
      "num_tokens": 13309112.0,
      "step": 7215
    },
    {
      "entropy": 5.78744421005249,
      "epoch": 0.6065952530980886,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004969239732024148,
      "loss": 5.6878,
      "mean_token_accuracy": 0.15506349503993988,
      "num_tokens": 13318328.0,
      "step": 7220
    },
    {
      "entropy": 5.677953767776489,
      "epoch": 0.6070153329132535,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004969190363618739,
      "loss": 5.5847,
      "mean_token_accuracy": 0.1533963978290558,
      "num_tokens": 13328940.0,
      "step": 7225
    },
    {
      "entropy": 5.71037449836731,
      "epoch": 0.6074354127284184,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004969140955901516,
      "loss": 5.6079,
      "mean_token_accuracy": 0.15790644884109498,
      "num_tokens": 13337829.0,
      "step": 7230
    },
    {
      "entropy": 5.844385385513306,
      "epoch": 0.6078554925435833,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004969091508873352,
      "loss": 5.7794,
      "mean_token_accuracy": 0.15029568076133729,
      "num_tokens": 13348289.0,
      "step": 7235
    },
    {
      "entropy": 5.801990795135498,
      "epoch": 0.6082755723587482,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004969042022535126,
      "loss": 5.6815,
      "mean_token_accuracy": 0.15717827379703522,
      "num_tokens": 13357292.0,
      "step": 7240
    },
    {
      "entropy": 5.778904294967651,
      "epoch": 0.6086956521739131,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004968992496887713,
      "loss": 5.7374,
      "mean_token_accuracy": 0.1471494920551777,
      "num_tokens": 13366640.0,
      "step": 7245
    },
    {
      "entropy": 5.804550743103027,
      "epoch": 0.609115731989078,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004968942931931989,
      "loss": 5.6431,
      "mean_token_accuracy": 0.15975850373506545,
      "num_tokens": 13377509.0,
      "step": 7250
    },
    {
      "entropy": 5.821708917617798,
      "epoch": 0.6095358118042428,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004968893327668835,
      "loss": 5.7281,
      "mean_token_accuracy": 0.14594982862472533,
      "num_tokens": 13386573.0,
      "step": 7255
    },
    {
      "entropy": 5.695713996887207,
      "epoch": 0.6099558916194077,
      "grad_norm": 1.25,
      "learning_rate": 0.0004968843684099128,
      "loss": 5.5984,
      "mean_token_accuracy": 0.15431449562311172,
      "num_tokens": 13395790.0,
      "step": 7260
    },
    {
      "entropy": 5.6999798774719235,
      "epoch": 0.6103759714345726,
      "grad_norm": 1.625,
      "learning_rate": 0.0004968794001223747,
      "loss": 5.659,
      "mean_token_accuracy": 0.15051533728837968,
      "num_tokens": 13405265.0,
      "step": 7265
    },
    {
      "entropy": 5.771698093414306,
      "epoch": 0.6107960512497375,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004968744279043574,
      "loss": 5.6372,
      "mean_token_accuracy": 0.1554905042052269,
      "num_tokens": 13413796.0,
      "step": 7270
    },
    {
      "entropy": 5.878541707992554,
      "epoch": 0.6112161310649024,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004968694517559488,
      "loss": 5.6955,
      "mean_token_accuracy": 0.15228554159402846,
      "num_tokens": 13423299.0,
      "step": 7275
    },
    {
      "entropy": 5.71440634727478,
      "epoch": 0.6116362108800673,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004968644716772371,
      "loss": 5.6168,
      "mean_token_accuracy": 0.15852915048599242,
      "num_tokens": 13432267.0,
      "step": 7280
    },
    {
      "entropy": 5.70712103843689,
      "epoch": 0.612056290695232,
      "grad_norm": 1.5,
      "learning_rate": 0.0004968594876683105,
      "loss": 5.7059,
      "mean_token_accuracy": 0.1497867949306965,
      "num_tokens": 13442332.0,
      "step": 7285
    },
    {
      "entropy": 5.776857042312622,
      "epoch": 0.6124763705103969,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004968544997292572,
      "loss": 5.65,
      "mean_token_accuracy": 0.1574311837553978,
      "num_tokens": 13451700.0,
      "step": 7290
    },
    {
      "entropy": 5.810136556625366,
      "epoch": 0.6128964503255618,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004968495078601659,
      "loss": 5.7362,
      "mean_token_accuracy": 0.14313261955976486,
      "num_tokens": 13461009.0,
      "step": 7295
    },
    {
      "entropy": 5.852288770675659,
      "epoch": 0.6133165301407267,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004968445120611247,
      "loss": 5.7511,
      "mean_token_accuracy": 0.15160316973924637,
      "num_tokens": 13470341.0,
      "step": 7300
    },
    {
      "entropy": 5.815583086013794,
      "epoch": 0.6137366099558916,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004968395123322223,
      "loss": 5.6802,
      "mean_token_accuracy": 0.15470752790570258,
      "num_tokens": 13479898.0,
      "step": 7305
    },
    {
      "entropy": 5.749247694015503,
      "epoch": 0.6141566897710565,
      "grad_norm": 1.453125,
      "learning_rate": 0.000496834508673547,
      "loss": 5.584,
      "mean_token_accuracy": 0.15403383672237397,
      "num_tokens": 13488116.0,
      "step": 7310
    },
    {
      "entropy": 5.759438514709473,
      "epoch": 0.6145767695862214,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004968295010851877,
      "loss": 5.6356,
      "mean_token_accuracy": 0.15597606748342513,
      "num_tokens": 13497814.0,
      "step": 7315
    },
    {
      "entropy": 5.793872499465943,
      "epoch": 0.6149968494013862,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004968244895672331,
      "loss": 5.6334,
      "mean_token_accuracy": 0.14847874492406846,
      "num_tokens": 13506617.0,
      "step": 7320
    },
    {
      "entropy": 5.775398635864258,
      "epoch": 0.6154169292165511,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004968194741197718,
      "loss": 5.788,
      "mean_token_accuracy": 0.14676353633403777,
      "num_tokens": 13516632.0,
      "step": 7325
    },
    {
      "entropy": 5.834484338760376,
      "epoch": 0.615837009031716,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004968144547428927,
      "loss": 5.7015,
      "mean_token_accuracy": 0.1523831970989704,
      "num_tokens": 13526452.0,
      "step": 7330
    },
    {
      "entropy": 5.787644147872925,
      "epoch": 0.6162570888468809,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004968094314366848,
      "loss": 5.6104,
      "mean_token_accuracy": 0.1564450517296791,
      "num_tokens": 13535663.0,
      "step": 7335
    },
    {
      "entropy": 5.663765478134155,
      "epoch": 0.6166771686620458,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000496804404201237,
      "loss": 5.5183,
      "mean_token_accuracy": 0.1632273629307747,
      "num_tokens": 13544574.0,
      "step": 7340
    },
    {
      "entropy": 5.863478326797486,
      "epoch": 0.6170972484772107,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004967993730366385,
      "loss": 5.7136,
      "mean_token_accuracy": 0.1504902571439743,
      "num_tokens": 13553041.0,
      "step": 7345
    },
    {
      "entropy": 5.755918788909912,
      "epoch": 0.6175173282923756,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004967943379429781,
      "loss": 5.6698,
      "mean_token_accuracy": 0.1484751097857952,
      "num_tokens": 13562108.0,
      "step": 7350
    },
    {
      "entropy": 5.919857025146484,
      "epoch": 0.6179374081075404,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004967892989203454,
      "loss": 5.8215,
      "mean_token_accuracy": 0.1457022547721863,
      "num_tokens": 13571500.0,
      "step": 7355
    },
    {
      "entropy": 5.858237886428833,
      "epoch": 0.6183574879227053,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004967842559688295,
      "loss": 5.7304,
      "mean_token_accuracy": 0.1467614322900772,
      "num_tokens": 13581304.0,
      "step": 7360
    },
    {
      "entropy": 5.7586640357971195,
      "epoch": 0.6187775677378702,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004967792090885195,
      "loss": 5.6034,
      "mean_token_accuracy": 0.15647478327155112,
      "num_tokens": 13590734.0,
      "step": 7365
    },
    {
      "entropy": 5.683297252655029,
      "epoch": 0.6191976475530351,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004967741582795052,
      "loss": 5.6578,
      "mean_token_accuracy": 0.15261005461215973,
      "num_tokens": 13600486.0,
      "step": 7370
    },
    {
      "entropy": 5.855695104598999,
      "epoch": 0.6196177273682,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004967691035418758,
      "loss": 5.6842,
      "mean_token_accuracy": 0.14570327550172807,
      "num_tokens": 13610542.0,
      "step": 7375
    },
    {
      "entropy": 5.7847048282623295,
      "epoch": 0.6200378071833649,
      "grad_norm": 1.890625,
      "learning_rate": 0.000496764044875721,
      "loss": 5.6311,
      "mean_token_accuracy": 0.15491739362478257,
      "num_tokens": 13619431.0,
      "step": 7380
    },
    {
      "entropy": 5.743926477432251,
      "epoch": 0.6204578869985298,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004967589822811303,
      "loss": 5.6796,
      "mean_token_accuracy": 0.1506201907992363,
      "num_tokens": 13629930.0,
      "step": 7385
    },
    {
      "entropy": 5.907500982284546,
      "epoch": 0.6208779668136946,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004967539157581934,
      "loss": 5.811,
      "mean_token_accuracy": 0.14877504855394363,
      "num_tokens": 13639439.0,
      "step": 7390
    },
    {
      "entropy": 5.859608602523804,
      "epoch": 0.6212980466288595,
      "grad_norm": 1.453125,
      "learning_rate": 0.000496748845307,
      "loss": 5.7233,
      "mean_token_accuracy": 0.15566221624612808,
      "num_tokens": 13648548.0,
      "step": 7395
    },
    {
      "entropy": 5.76888256072998,
      "epoch": 0.6217181264440244,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004967437709276401,
      "loss": 5.7631,
      "mean_token_accuracy": 0.15250276327133178,
      "num_tokens": 13657658.0,
      "step": 7400
    },
    {
      "entropy": 5.7022544860839846,
      "epoch": 0.6221382062591893,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004967386926202034,
      "loss": 5.5529,
      "mean_token_accuracy": 0.15799887329339982,
      "num_tokens": 13666763.0,
      "step": 7405
    },
    {
      "entropy": 5.85063681602478,
      "epoch": 0.6225582860743542,
      "grad_norm": 1.484375,
      "learning_rate": 0.00049673361038478,
      "loss": 5.7924,
      "mean_token_accuracy": 0.14320089891552926,
      "num_tokens": 13676527.0,
      "step": 7410
    },
    {
      "entropy": 5.783917856216431,
      "epoch": 0.622978365889519,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004967285242214599,
      "loss": 5.7395,
      "mean_token_accuracy": 0.152333851903677,
      "num_tokens": 13685404.0,
      "step": 7415
    },
    {
      "entropy": 5.781331348419189,
      "epoch": 0.6233984457046838,
      "grad_norm": 1.484375,
      "learning_rate": 0.000496723434130333,
      "loss": 5.5582,
      "mean_token_accuracy": 0.15735181719064711,
      "num_tokens": 13693118.0,
      "step": 7420
    },
    {
      "entropy": 5.746804714202881,
      "epoch": 0.6238185255198487,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004967183401114898,
      "loss": 5.6315,
      "mean_token_accuracy": 0.1518068231642246,
      "num_tokens": 13702015.0,
      "step": 7425
    },
    {
      "entropy": 5.754427099227906,
      "epoch": 0.6242386053350136,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004967132421650203,
      "loss": 5.6635,
      "mean_token_accuracy": 0.15001336112618446,
      "num_tokens": 13711658.0,
      "step": 7430
    },
    {
      "entropy": 5.663620853424073,
      "epoch": 0.6246586851501785,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004967081402910149,
      "loss": 5.6707,
      "mean_token_accuracy": 0.1505231335759163,
      "num_tokens": 13720718.0,
      "step": 7435
    },
    {
      "entropy": 5.7571845054626465,
      "epoch": 0.6250787649653434,
      "grad_norm": 1.484375,
      "learning_rate": 0.000496703034489564,
      "loss": 5.5426,
      "mean_token_accuracy": 0.15831270217895507,
      "num_tokens": 13729364.0,
      "step": 7440
    },
    {
      "entropy": 5.881094646453858,
      "epoch": 0.6254988447805083,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004966979247607579,
      "loss": 5.8386,
      "mean_token_accuracy": 0.14335498884320258,
      "num_tokens": 13739436.0,
      "step": 7445
    },
    {
      "entropy": 5.8375767230987545,
      "epoch": 0.6259189245956732,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004966928111046873,
      "loss": 5.7527,
      "mean_token_accuracy": 0.15689793825149537,
      "num_tokens": 13749196.0,
      "step": 7450
    },
    {
      "entropy": 5.786605453491211,
      "epoch": 0.626339004410838,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004966876935214426,
      "loss": 5.6034,
      "mean_token_accuracy": 0.15611332505941392,
      "num_tokens": 13758414.0,
      "step": 7455
    },
    {
      "entropy": 5.758475589752197,
      "epoch": 0.6267590842260029,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004966825720111147,
      "loss": 5.6428,
      "mean_token_accuracy": 0.1506047248840332,
      "num_tokens": 13767496.0,
      "step": 7460
    },
    {
      "entropy": 5.810634279251099,
      "epoch": 0.6271791640411678,
      "grad_norm": 1.375,
      "learning_rate": 0.0004966774465737942,
      "loss": 5.7739,
      "mean_token_accuracy": 0.15019554272294044,
      "num_tokens": 13777033.0,
      "step": 7465
    },
    {
      "entropy": 5.827461242675781,
      "epoch": 0.6275992438563327,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004966723172095717,
      "loss": 5.7321,
      "mean_token_accuracy": 0.14653541669249534,
      "num_tokens": 13786313.0,
      "step": 7470
    },
    {
      "entropy": 5.787878465652466,
      "epoch": 0.6280193236714976,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004966671839185384,
      "loss": 5.6609,
      "mean_token_accuracy": 0.15652250424027442,
      "num_tokens": 13795257.0,
      "step": 7475
    },
    {
      "entropy": 5.715157699584961,
      "epoch": 0.6284394034866625,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004966620467007851,
      "loss": 5.5822,
      "mean_token_accuracy": 0.15817383229732512,
      "num_tokens": 13804582.0,
      "step": 7480
    },
    {
      "entropy": 5.718271636962891,
      "epoch": 0.6288594833018274,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004966569055564027,
      "loss": 5.5588,
      "mean_token_accuracy": 0.15505106449127198,
      "num_tokens": 13813248.0,
      "step": 7485
    },
    {
      "entropy": 5.8337007522583,
      "epoch": 0.6292795631169922,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004966517604854823,
      "loss": 5.8426,
      "mean_token_accuracy": 0.13890477865934373,
      "num_tokens": 13823301.0,
      "step": 7490
    },
    {
      "entropy": 5.777707862854004,
      "epoch": 0.6296996429321571,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004966466114881152,
      "loss": 5.5601,
      "mean_token_accuracy": 0.16184355765581132,
      "num_tokens": 13832040.0,
      "step": 7495
    },
    {
      "entropy": 5.751884889602661,
      "epoch": 0.630119722747322,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004966414585643925,
      "loss": 5.7486,
      "mean_token_accuracy": 0.1461435317993164,
      "num_tokens": 13841874.0,
      "step": 7500
    },
    {
      "entropy": 5.743130207061768,
      "epoch": 0.6305398025624869,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004966363017144055,
      "loss": 5.5699,
      "mean_token_accuracy": 0.1629691332578659,
      "num_tokens": 13850755.0,
      "step": 7505
    },
    {
      "entropy": 5.75084810256958,
      "epoch": 0.6309598823776518,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004966311409382455,
      "loss": 5.6466,
      "mean_token_accuracy": 0.15385955274105073,
      "num_tokens": 13860009.0,
      "step": 7510
    },
    {
      "entropy": 5.6666340827941895,
      "epoch": 0.6313799621928167,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004966259762360039,
      "loss": 5.5514,
      "mean_token_accuracy": 0.16524431109428406,
      "num_tokens": 13868476.0,
      "step": 7515
    },
    {
      "entropy": 5.6674237728118895,
      "epoch": 0.6318000420079816,
      "grad_norm": 1.375,
      "learning_rate": 0.0004966208076077723,
      "loss": 5.5884,
      "mean_token_accuracy": 0.1589062973856926,
      "num_tokens": 13877367.0,
      "step": 7520
    },
    {
      "entropy": 5.764949607849121,
      "epoch": 0.6322201218231464,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004966156350536422,
      "loss": 5.657,
      "mean_token_accuracy": 0.1521471455693245,
      "num_tokens": 13885985.0,
      "step": 7525
    },
    {
      "entropy": 5.7143798828125,
      "epoch": 0.6326402016383113,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004966104585737054,
      "loss": 5.5793,
      "mean_token_accuracy": 0.1567079693078995,
      "num_tokens": 13895059.0,
      "step": 7530
    },
    {
      "entropy": 5.7325562000274655,
      "epoch": 0.6330602814534761,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004966052781680534,
      "loss": 5.639,
      "mean_token_accuracy": 0.1523404136300087,
      "num_tokens": 13903789.0,
      "step": 7535
    },
    {
      "entropy": 5.77357177734375,
      "epoch": 0.633480361268641,
      "grad_norm": 1.375,
      "learning_rate": 0.0004966000938367778,
      "loss": 5.6303,
      "mean_token_accuracy": 0.15177295506000518,
      "num_tokens": 13913377.0,
      "step": 7540
    },
    {
      "entropy": 5.674224042892456,
      "epoch": 0.6339004410838059,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004965949055799708,
      "loss": 5.5907,
      "mean_token_accuracy": 0.16200509965419768,
      "num_tokens": 13922141.0,
      "step": 7545
    },
    {
      "entropy": 5.762849998474121,
      "epoch": 0.6343205208989708,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004965897133977241,
      "loss": 5.6311,
      "mean_token_accuracy": 0.14147766605019568,
      "num_tokens": 13930717.0,
      "step": 7550
    },
    {
      "entropy": 5.803768348693848,
      "epoch": 0.6347406007141357,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004965845172901298,
      "loss": 5.7098,
      "mean_token_accuracy": 0.15163939744234084,
      "num_tokens": 13940344.0,
      "step": 7555
    },
    {
      "entropy": 5.696092939376831,
      "epoch": 0.6351606805293005,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004965793172572798,
      "loss": 5.5525,
      "mean_token_accuracy": 0.15789541453123093,
      "num_tokens": 13948400.0,
      "step": 7560
    },
    {
      "entropy": 5.68128399848938,
      "epoch": 0.6355807603444654,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004965741132992663,
      "loss": 5.6728,
      "mean_token_accuracy": 0.14891874939203262,
      "num_tokens": 13957939.0,
      "step": 7565
    },
    {
      "entropy": 5.8004132270812985,
      "epoch": 0.6360008401596303,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004965689054161814,
      "loss": 5.6336,
      "mean_token_accuracy": 0.15334389358758926,
      "num_tokens": 13966943.0,
      "step": 7570
    },
    {
      "entropy": 5.748394632339478,
      "epoch": 0.6364209199747952,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004965636936081176,
      "loss": 5.5608,
      "mean_token_accuracy": 0.15567472875118255,
      "num_tokens": 13975850.0,
      "step": 7575
    },
    {
      "entropy": 5.780313920974732,
      "epoch": 0.6368409997899601,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000496558477875167,
      "loss": 5.6499,
      "mean_token_accuracy": 0.1615470714867115,
      "num_tokens": 13985059.0,
      "step": 7580
    },
    {
      "entropy": 5.7933934211730955,
      "epoch": 0.637261079605125,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000496553258217422,
      "loss": 5.7004,
      "mean_token_accuracy": 0.1459404468536377,
      "num_tokens": 13993571.0,
      "step": 7585
    },
    {
      "entropy": 5.8100879192352295,
      "epoch": 0.6376811594202898,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004965480346349751,
      "loss": 5.6982,
      "mean_token_accuracy": 0.1491829678416252,
      "num_tokens": 14002326.0,
      "step": 7590
    },
    {
      "entropy": 5.909585475921631,
      "epoch": 0.6381012392354547,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000496542807127919,
      "loss": 5.8381,
      "mean_token_accuracy": 0.1500192791223526,
      "num_tokens": 14012002.0,
      "step": 7595
    },
    {
      "entropy": 5.814299058914185,
      "epoch": 0.6385213190506196,
      "grad_norm": 1.4375,
      "learning_rate": 0.000496537575696346,
      "loss": 5.7004,
      "mean_token_accuracy": 0.14627951383590698,
      "num_tokens": 14022085.0,
      "step": 7600
    },
    {
      "entropy": 5.694448709487915,
      "epoch": 0.6389413988657845,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004965323403403488,
      "loss": 5.5641,
      "mean_token_accuracy": 0.15740595012903214,
      "num_tokens": 14030706.0,
      "step": 7605
    },
    {
      "entropy": 5.74091477394104,
      "epoch": 0.6393614786809494,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004965271010600205,
      "loss": 5.5936,
      "mean_token_accuracy": 0.15839929282665252,
      "num_tokens": 14039520.0,
      "step": 7610
    },
    {
      "entropy": 5.824947929382324,
      "epoch": 0.6397815584961143,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004965218578554535,
      "loss": 5.7052,
      "mean_token_accuracy": 0.15347498506307602,
      "num_tokens": 14048407.0,
      "step": 7615
    },
    {
      "entropy": 5.705046701431274,
      "epoch": 0.6402016383112792,
      "grad_norm": 1.328125,
      "learning_rate": 0.000496516610726741,
      "loss": 5.6076,
      "mean_token_accuracy": 0.15854684114456177,
      "num_tokens": 14057534.0,
      "step": 7620
    },
    {
      "entropy": 5.763433980941772,
      "epoch": 0.640621718126444,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004965113596739759,
      "loss": 5.5959,
      "mean_token_accuracy": 0.1591039463877678,
      "num_tokens": 14065992.0,
      "step": 7625
    },
    {
      "entropy": 5.7118175506591795,
      "epoch": 0.6410417979416089,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004965061046972508,
      "loss": 5.5862,
      "mean_token_accuracy": 0.1558775633573532,
      "num_tokens": 14074806.0,
      "step": 7630
    },
    {
      "entropy": 5.717222595214844,
      "epoch": 0.6414618777567738,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004965008457966594,
      "loss": 5.6274,
      "mean_token_accuracy": 0.15323564708232879,
      "num_tokens": 14083813.0,
      "step": 7635
    },
    {
      "entropy": 5.750378274917603,
      "epoch": 0.6418819575719387,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004964955829722945,
      "loss": 5.5653,
      "mean_token_accuracy": 0.15824178904294967,
      "num_tokens": 14092193.0,
      "step": 7640
    },
    {
      "entropy": 5.85125765800476,
      "epoch": 0.6423020373871036,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004964903162242493,
      "loss": 5.7825,
      "mean_token_accuracy": 0.1446760669350624,
      "num_tokens": 14102797.0,
      "step": 7645
    },
    {
      "entropy": 5.775562286376953,
      "epoch": 0.6427221172022685,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004964850455526173,
      "loss": 5.6487,
      "mean_token_accuracy": 0.1530699238181114,
      "num_tokens": 14112226.0,
      "step": 7650
    },
    {
      "entropy": 5.671306467056274,
      "epoch": 0.6431421970174334,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004964797709574917,
      "loss": 5.5723,
      "mean_token_accuracy": 0.15474275052547454,
      "num_tokens": 14121775.0,
      "step": 7655
    },
    {
      "entropy": 5.7099405288696286,
      "epoch": 0.6435622768325981,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000496474492438966,
      "loss": 5.5616,
      "mean_token_accuracy": 0.15505539178848265,
      "num_tokens": 14130415.0,
      "step": 7660
    },
    {
      "entropy": 5.719390869140625,
      "epoch": 0.643982356647763,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004964692099971338,
      "loss": 5.5889,
      "mean_token_accuracy": 0.15883549749851228,
      "num_tokens": 14140204.0,
      "step": 7665
    },
    {
      "entropy": 5.727085542678833,
      "epoch": 0.6444024364629279,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004964639236320885,
      "loss": 5.5326,
      "mean_token_accuracy": 0.15387142151594163,
      "num_tokens": 14149595.0,
      "step": 7670
    },
    {
      "entropy": 5.671725845336914,
      "epoch": 0.6448225162780928,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004964586333439239,
      "loss": 5.6096,
      "mean_token_accuracy": 0.15158815383911134,
      "num_tokens": 14158865.0,
      "step": 7675
    },
    {
      "entropy": 5.69843373298645,
      "epoch": 0.6452425960932577,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004964533391327335,
      "loss": 5.5645,
      "mean_token_accuracy": 0.16603838801383972,
      "num_tokens": 14167962.0,
      "step": 7680
    },
    {
      "entropy": 5.727479887008667,
      "epoch": 0.6456626759084226,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004964480409986113,
      "loss": 5.6263,
      "mean_token_accuracy": 0.1631734326481819,
      "num_tokens": 14176479.0,
      "step": 7685
    },
    {
      "entropy": 5.830864524841308,
      "epoch": 0.6460827557235875,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004964427389416512,
      "loss": 5.6441,
      "mean_token_accuracy": 0.15119680762290955,
      "num_tokens": 14185408.0,
      "step": 7690
    },
    {
      "entropy": 5.700068235397339,
      "epoch": 0.6465028355387523,
      "grad_norm": 1.65625,
      "learning_rate": 0.000496437432961947,
      "loss": 5.6396,
      "mean_token_accuracy": 0.15550505965948105,
      "num_tokens": 14194155.0,
      "step": 7695
    },
    {
      "entropy": 5.704675006866455,
      "epoch": 0.6469229153539172,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004964321230595925,
      "loss": 5.6697,
      "mean_token_accuracy": 0.15572153329849242,
      "num_tokens": 14202779.0,
      "step": 7700
    },
    {
      "entropy": 5.89739408493042,
      "epoch": 0.6473429951690821,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004964268092346821,
      "loss": 5.856,
      "mean_token_accuracy": 0.1441698580980301,
      "num_tokens": 14212552.0,
      "step": 7705
    },
    {
      "entropy": 5.918906402587891,
      "epoch": 0.647763074984247,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004964214914873098,
      "loss": 5.6441,
      "mean_token_accuracy": 0.14508990049362183,
      "num_tokens": 14222783.0,
      "step": 7710
    },
    {
      "entropy": 5.6578751564025875,
      "epoch": 0.6481831547994119,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004964161698175697,
      "loss": 5.5291,
      "mean_token_accuracy": 0.15217974185943603,
      "num_tokens": 14232085.0,
      "step": 7715
    },
    {
      "entropy": 5.666456794738769,
      "epoch": 0.6486032346145768,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004964108442255562,
      "loss": 5.6923,
      "mean_token_accuracy": 0.14667234867811202,
      "num_tokens": 14241969.0,
      "step": 7720
    },
    {
      "entropy": 5.712312459945679,
      "epoch": 0.6490233144297417,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004964055147113637,
      "loss": 5.6018,
      "mean_token_accuracy": 0.15744369924068452,
      "num_tokens": 14251012.0,
      "step": 7725
    },
    {
      "entropy": 5.881955242156982,
      "epoch": 0.6494433942449065,
      "grad_norm": 2.125,
      "learning_rate": 0.0004964001812750864,
      "loss": 5.7226,
      "mean_token_accuracy": 0.14762117117643356,
      "num_tokens": 14261110.0,
      "step": 7730
    },
    {
      "entropy": 5.768258047103882,
      "epoch": 0.6498634740600714,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000496394843916819,
      "loss": 5.7015,
      "mean_token_accuracy": 0.14753929674625396,
      "num_tokens": 14270869.0,
      "step": 7735
    },
    {
      "entropy": 5.736590814590454,
      "epoch": 0.6502835538752363,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004963895026366558,
      "loss": 5.6374,
      "mean_token_accuracy": 0.14768418967723845,
      "num_tokens": 14279607.0,
      "step": 7740
    },
    {
      "entropy": 5.7367908477783205,
      "epoch": 0.6507036336904012,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004963841574346917,
      "loss": 5.6264,
      "mean_token_accuracy": 0.14881719276309013,
      "num_tokens": 14289282.0,
      "step": 7745
    },
    {
      "entropy": 5.697269630432129,
      "epoch": 0.6511237135055661,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004963788083110212,
      "loss": 5.5701,
      "mean_token_accuracy": 0.15828859508037568,
      "num_tokens": 14298658.0,
      "step": 7750
    },
    {
      "entropy": 5.868004083633423,
      "epoch": 0.651543793320731,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000496373455265739,
      "loss": 5.6507,
      "mean_token_accuracy": 0.1502188928425312,
      "num_tokens": 14307832.0,
      "step": 7755
    },
    {
      "entropy": 5.7541135311126705,
      "epoch": 0.6519638731358958,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004963680982989402,
      "loss": 5.5594,
      "mean_token_accuracy": 0.15743454396724701,
      "num_tokens": 14317122.0,
      "step": 7760
    },
    {
      "entropy": 5.668973541259765,
      "epoch": 0.6523839529510607,
      "grad_norm": 1.5,
      "learning_rate": 0.0004963627374107195,
      "loss": 5.6035,
      "mean_token_accuracy": 0.15760391652584077,
      "num_tokens": 14326069.0,
      "step": 7765
    },
    {
      "entropy": 5.701200485229492,
      "epoch": 0.6528040327662256,
      "grad_norm": 1.625,
      "learning_rate": 0.0004963573726011717,
      "loss": 5.604,
      "mean_token_accuracy": 0.1542496606707573,
      "num_tokens": 14335260.0,
      "step": 7770
    },
    {
      "entropy": 5.85819001197815,
      "epoch": 0.6532241125813905,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004963520038703922,
      "loss": 5.6911,
      "mean_token_accuracy": 0.14053659662604331,
      "num_tokens": 14345823.0,
      "step": 7775
    },
    {
      "entropy": 5.743032455444336,
      "epoch": 0.6536441923965554,
      "grad_norm": 1.609375,
      "learning_rate": 0.000496346631218476,
      "loss": 5.5588,
      "mean_token_accuracy": 0.15523895770311355,
      "num_tokens": 14354316.0,
      "step": 7780
    },
    {
      "entropy": 5.75605149269104,
      "epoch": 0.6540642722117203,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000496341254645518,
      "loss": 5.6096,
      "mean_token_accuracy": 0.1583119735121727,
      "num_tokens": 14364539.0,
      "step": 7785
    },
    {
      "entropy": 5.746860456466675,
      "epoch": 0.6544843520268852,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004963358741516138,
      "loss": 5.728,
      "mean_token_accuracy": 0.14681073948740958,
      "num_tokens": 14374081.0,
      "step": 7790
    },
    {
      "entropy": 5.743828630447387,
      "epoch": 0.6549044318420499,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004963304897368585,
      "loss": 5.6157,
      "mean_token_accuracy": 0.15063438564538956,
      "num_tokens": 14383255.0,
      "step": 7795
    },
    {
      "entropy": 5.8674695014953615,
      "epoch": 0.6553245116572148,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004963251014013475,
      "loss": 5.7454,
      "mean_token_accuracy": 0.1499713510274887,
      "num_tokens": 14392417.0,
      "step": 7800
    },
    {
      "entropy": 5.904078197479248,
      "epoch": 0.6557445914723797,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004963197091451763,
      "loss": 5.7859,
      "mean_token_accuracy": 0.13848365619778633,
      "num_tokens": 14401899.0,
      "step": 7805
    },
    {
      "entropy": 5.860718584060669,
      "epoch": 0.6561646712875446,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004963143129684405,
      "loss": 5.7538,
      "mean_token_accuracy": 0.14672411531209945,
      "num_tokens": 14411245.0,
      "step": 7810
    },
    {
      "entropy": 5.703166818618774,
      "epoch": 0.6565847511027095,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004963089128712355,
      "loss": 5.6064,
      "mean_token_accuracy": 0.15734720826148987,
      "num_tokens": 14419710.0,
      "step": 7815
    },
    {
      "entropy": 5.71895399093628,
      "epoch": 0.6570048309178744,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004963035088536571,
      "loss": 5.5915,
      "mean_token_accuracy": 0.15892575681209564,
      "num_tokens": 14430266.0,
      "step": 7820
    },
    {
      "entropy": 5.801663017272949,
      "epoch": 0.6574249107330393,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004962981009158012,
      "loss": 5.5594,
      "mean_token_accuracy": 0.14950107038021088,
      "num_tokens": 14439515.0,
      "step": 7825
    },
    {
      "entropy": 5.776130819320679,
      "epoch": 0.6578449905482041,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004962926890577632,
      "loss": 5.625,
      "mean_token_accuracy": 0.15389978587627412,
      "num_tokens": 14448091.0,
      "step": 7830
    },
    {
      "entropy": 5.71961612701416,
      "epoch": 0.658265070363369,
      "grad_norm": 2.921875,
      "learning_rate": 0.000496287273279639,
      "loss": 5.6595,
      "mean_token_accuracy": 0.14972689151763915,
      "num_tokens": 14457744.0,
      "step": 7835
    },
    {
      "entropy": 5.747187852859497,
      "epoch": 0.6586851501785339,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000496281853581525,
      "loss": 5.6368,
      "mean_token_accuracy": 0.15765562653541565,
      "num_tokens": 14467597.0,
      "step": 7840
    },
    {
      "entropy": 5.736932706832886,
      "epoch": 0.6591052299936988,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004962764299635168,
      "loss": 5.6237,
      "mean_token_accuracy": 0.15176722705364226,
      "num_tokens": 14476662.0,
      "step": 7845
    },
    {
      "entropy": 5.84397029876709,
      "epoch": 0.6595253098088637,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004962710024257105,
      "loss": 5.7182,
      "mean_token_accuracy": 0.1437528148293495,
      "num_tokens": 14486583.0,
      "step": 7850
    },
    {
      "entropy": 5.882024717330933,
      "epoch": 0.6599453896240286,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004962655709682025,
      "loss": 5.7135,
      "mean_token_accuracy": 0.14967520385980607,
      "num_tokens": 14496528.0,
      "step": 7855
    },
    {
      "entropy": 5.7820171356201175,
      "epoch": 0.6603654694391935,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004962601355910887,
      "loss": 5.6997,
      "mean_token_accuracy": 0.14425860121846198,
      "num_tokens": 14507026.0,
      "step": 7860
    },
    {
      "entropy": 5.658552885055542,
      "epoch": 0.6607855492543583,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004962546962944656,
      "loss": 5.5687,
      "mean_token_accuracy": 0.1535755679011345,
      "num_tokens": 14516480.0,
      "step": 7865
    },
    {
      "entropy": 5.741037130355835,
      "epoch": 0.6612056290695232,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004962492530784295,
      "loss": 5.5149,
      "mean_token_accuracy": 0.16426774710416794,
      "num_tokens": 14525068.0,
      "step": 7870
    },
    {
      "entropy": 5.7695718765258786,
      "epoch": 0.6616257088846881,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004962438059430768,
      "loss": 5.6342,
      "mean_token_accuracy": 0.15432853996753693,
      "num_tokens": 14534441.0,
      "step": 7875
    },
    {
      "entropy": 5.770703077316284,
      "epoch": 0.662045788699853,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004962383548885039,
      "loss": 5.7265,
      "mean_token_accuracy": 0.15264071598649026,
      "num_tokens": 14543026.0,
      "step": 7880
    },
    {
      "entropy": 5.728618335723877,
      "epoch": 0.6624658685150179,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004962328999148075,
      "loss": 5.5953,
      "mean_token_accuracy": 0.1570552945137024,
      "num_tokens": 14552068.0,
      "step": 7885
    },
    {
      "entropy": 5.7635719776153564,
      "epoch": 0.6628859483301828,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004962274410220842,
      "loss": 5.7341,
      "mean_token_accuracy": 0.14789471104741098,
      "num_tokens": 14561587.0,
      "step": 7890
    },
    {
      "entropy": 5.859791803359985,
      "epoch": 0.6633060281453477,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004962219782104308,
      "loss": 5.7256,
      "mean_token_accuracy": 0.16066627949476242,
      "num_tokens": 14571020.0,
      "step": 7895
    },
    {
      "entropy": 5.814104175567627,
      "epoch": 0.6637261079605125,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004962165114799439,
      "loss": 5.6841,
      "mean_token_accuracy": 0.1433543421328068,
      "num_tokens": 14580638.0,
      "step": 7900
    },
    {
      "entropy": 5.693702554702758,
      "epoch": 0.6641461877756774,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004962110408307204,
      "loss": 5.6214,
      "mean_token_accuracy": 0.15168005228042603,
      "num_tokens": 14590173.0,
      "step": 7905
    },
    {
      "entropy": 5.696859216690063,
      "epoch": 0.6645662675908423,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004962055662628571,
      "loss": 5.5978,
      "mean_token_accuracy": 0.15115945935249328,
      "num_tokens": 14598635.0,
      "step": 7910
    },
    {
      "entropy": 5.789608335494995,
      "epoch": 0.6649863474060071,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004962000877764513,
      "loss": 5.6362,
      "mean_token_accuracy": 0.15672800540924073,
      "num_tokens": 14607233.0,
      "step": 7915
    },
    {
      "entropy": 5.904106950759887,
      "epoch": 0.665406427221172,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004961946053715998,
      "loss": 5.8042,
      "mean_token_accuracy": 0.14238024279475212,
      "num_tokens": 14617483.0,
      "step": 7920
    },
    {
      "entropy": 5.73907699584961,
      "epoch": 0.665826507036337,
      "grad_norm": 1.625,
      "learning_rate": 0.0004961891190483997,
      "loss": 5.6099,
      "mean_token_accuracy": 0.15576208233833314,
      "num_tokens": 14625805.0,
      "step": 7925
    },
    {
      "entropy": 5.668482398986816,
      "epoch": 0.6662465868515017,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004961836288069483,
      "loss": 5.5303,
      "mean_token_accuracy": 0.15367640256881715,
      "num_tokens": 14634605.0,
      "step": 7930
    },
    {
      "entropy": 5.802455711364746,
      "epoch": 0.6666666666666666,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004961781346473428,
      "loss": 5.7424,
      "mean_token_accuracy": 0.1421908602118492,
      "num_tokens": 14644970.0,
      "step": 7935
    },
    {
      "entropy": 5.835074090957642,
      "epoch": 0.6670867464818315,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004961726365696805,
      "loss": 5.6277,
      "mean_token_accuracy": 0.15110262408852576,
      "num_tokens": 14655043.0,
      "step": 7940
    },
    {
      "entropy": 5.85212664604187,
      "epoch": 0.6675068262969964,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004961671345740589,
      "loss": 5.6034,
      "mean_token_accuracy": 0.15058621391654015,
      "num_tokens": 14663994.0,
      "step": 7945
    },
    {
      "entropy": 5.6571033000946045,
      "epoch": 0.6679269061121613,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004961616286605753,
      "loss": 5.6122,
      "mean_token_accuracy": 0.1486305832862854,
      "num_tokens": 14674101.0,
      "step": 7950
    },
    {
      "entropy": 5.6964271068573,
      "epoch": 0.6683469859273262,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004961561188293273,
      "loss": 5.6979,
      "mean_token_accuracy": 0.14713332504034043,
      "num_tokens": 14684156.0,
      "step": 7955
    },
    {
      "entropy": 5.670635604858399,
      "epoch": 0.6687670657424911,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004961506050804126,
      "loss": 5.6027,
      "mean_token_accuracy": 0.15685120671987535,
      "num_tokens": 14693223.0,
      "step": 7960
    },
    {
      "entropy": 5.775451517105102,
      "epoch": 0.6691871455576559,
      "grad_norm": 1.71875,
      "learning_rate": 0.000496145087413929,
      "loss": 5.604,
      "mean_token_accuracy": 0.15279402881860732,
      "num_tokens": 14702959.0,
      "step": 7965
    },
    {
      "entropy": 5.8267169952392575,
      "epoch": 0.6696072253728208,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004961395658299737,
      "loss": 5.7168,
      "mean_token_accuracy": 0.14913576021790503,
      "num_tokens": 14712146.0,
      "step": 7970
    },
    {
      "entropy": 5.767063426971435,
      "epoch": 0.6700273051879857,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004961340403286451,
      "loss": 5.6427,
      "mean_token_accuracy": 0.1493378534913063,
      "num_tokens": 14721932.0,
      "step": 7975
    },
    {
      "entropy": 5.712933492660523,
      "epoch": 0.6704473850031506,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004961285109100408,
      "loss": 5.5679,
      "mean_token_accuracy": 0.16175169199705125,
      "num_tokens": 14731080.0,
      "step": 7980
    },
    {
      "entropy": 5.712076044082641,
      "epoch": 0.6708674648183155,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004961229775742587,
      "loss": 5.5804,
      "mean_token_accuracy": 0.16035438925027848,
      "num_tokens": 14740057.0,
      "step": 7985
    },
    {
      "entropy": 5.814006376266479,
      "epoch": 0.6712875446334804,
      "grad_norm": 1.484375,
      "learning_rate": 0.000496117440321397,
      "loss": 5.6576,
      "mean_token_accuracy": 0.15759425461292267,
      "num_tokens": 14748399.0,
      "step": 7990
    },
    {
      "entropy": 5.742432641983032,
      "epoch": 0.6717076244486453,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004961118991515537,
      "loss": 5.6627,
      "mean_token_accuracy": 0.1502150148153305,
      "num_tokens": 14757215.0,
      "step": 7995
    },
    {
      "entropy": 5.684312534332276,
      "epoch": 0.6721277042638101,
      "grad_norm": 1.640625,
      "learning_rate": 0.000496106354064827,
      "loss": 5.6795,
      "mean_token_accuracy": 0.15305250138044357,
      "num_tokens": 14766191.0,
      "step": 8000
    },
    {
      "entropy": 5.875341749191284,
      "epoch": 0.672547784078975,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004961008050613149,
      "loss": 5.7247,
      "mean_token_accuracy": 0.1427694909274578,
      "num_tokens": 14775220.0,
      "step": 8005
    },
    {
      "entropy": 5.894955635070801,
      "epoch": 0.6729678638941399,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004960952521411161,
      "loss": 5.6944,
      "mean_token_accuracy": 0.14855958074331282,
      "num_tokens": 14784287.0,
      "step": 8010
    },
    {
      "entropy": 5.868842029571534,
      "epoch": 0.6733879437093048,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004960896953043287,
      "loss": 5.7613,
      "mean_token_accuracy": 0.14570847973227502,
      "num_tokens": 14794219.0,
      "step": 8015
    },
    {
      "entropy": 5.783145427703857,
      "epoch": 0.6738080235244697,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004960841345510511,
      "loss": 5.6466,
      "mean_token_accuracy": 0.1534015029668808,
      "num_tokens": 14803324.0,
      "step": 8020
    },
    {
      "entropy": 5.758127498626709,
      "epoch": 0.6742281033396346,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000496078569881382,
      "loss": 5.669,
      "mean_token_accuracy": 0.15486637651920318,
      "num_tokens": 14811963.0,
      "step": 8025
    },
    {
      "entropy": 5.706368112564087,
      "epoch": 0.6746481831547995,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004960730012954198,
      "loss": 5.6159,
      "mean_token_accuracy": 0.15371300429105758,
      "num_tokens": 14821903.0,
      "step": 8030
    },
    {
      "entropy": 5.682500696182251,
      "epoch": 0.6750682629699643,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004960674287932634,
      "loss": 5.6036,
      "mean_token_accuracy": 0.15207039266824723,
      "num_tokens": 14831215.0,
      "step": 8035
    },
    {
      "entropy": 5.7689878940582275,
      "epoch": 0.6754883427851291,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004960618523750111,
      "loss": 5.5344,
      "mean_token_accuracy": 0.16025708019733428,
      "num_tokens": 14840354.0,
      "step": 8040
    },
    {
      "entropy": 5.807526636123657,
      "epoch": 0.675908422600294,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000496056272040762,
      "loss": 5.7256,
      "mean_token_accuracy": 0.1501130685210228,
      "num_tokens": 14849660.0,
      "step": 8045
    },
    {
      "entropy": 5.7956421852111815,
      "epoch": 0.6763285024154589,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004960506877906149,
      "loss": 5.6257,
      "mean_token_accuracy": 0.14838283210992814,
      "num_tokens": 14859819.0,
      "step": 8050
    },
    {
      "entropy": 5.757112693786621,
      "epoch": 0.6767485822306238,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004960450996246686,
      "loss": 5.6388,
      "mean_token_accuracy": 0.15627321898937224,
      "num_tokens": 14869260.0,
      "step": 8055
    },
    {
      "entropy": 5.7164520740509035,
      "epoch": 0.6771686620457887,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004960395075430222,
      "loss": 5.614,
      "mean_token_accuracy": 0.15564015433192252,
      "num_tokens": 14878685.0,
      "step": 8060
    },
    {
      "entropy": 5.777329826354981,
      "epoch": 0.6775887418609536,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004960339115457748,
      "loss": 5.6172,
      "mean_token_accuracy": 0.1507098227739334,
      "num_tokens": 14888456.0,
      "step": 8065
    },
    {
      "entropy": 5.80708589553833,
      "epoch": 0.6780088216761184,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004960283116330255,
      "loss": 5.7205,
      "mean_token_accuracy": 0.14664217829704285,
      "num_tokens": 14897401.0,
      "step": 8070
    },
    {
      "entropy": 5.84255633354187,
      "epoch": 0.6784289014912833,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004960227078048735,
      "loss": 5.6226,
      "mean_token_accuracy": 0.15697504729032516,
      "num_tokens": 14906741.0,
      "step": 8075
    },
    {
      "entropy": 5.720810079574585,
      "epoch": 0.6788489813064482,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004960171000614179,
      "loss": 5.5069,
      "mean_token_accuracy": 0.16248478889465331,
      "num_tokens": 14916002.0,
      "step": 8080
    },
    {
      "entropy": 5.585197591781617,
      "epoch": 0.6792690611216131,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004960114884027583,
      "loss": 5.4659,
      "mean_token_accuracy": 0.1668694093823433,
      "num_tokens": 14925247.0,
      "step": 8085
    },
    {
      "entropy": 5.697068214416504,
      "epoch": 0.679689140936778,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004960058728289939,
      "loss": 5.5785,
      "mean_token_accuracy": 0.1530070647597313,
      "num_tokens": 14933925.0,
      "step": 8090
    },
    {
      "entropy": 5.845986604690552,
      "epoch": 0.6801092207519429,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004960002533402243,
      "loss": 5.6551,
      "mean_token_accuracy": 0.1536601111292839,
      "num_tokens": 14943368.0,
      "step": 8095
    },
    {
      "entropy": 5.7944434642791744,
      "epoch": 0.6805293005671077,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004959946299365491,
      "loss": 5.6835,
      "mean_token_accuracy": 0.147177092730999,
      "num_tokens": 14953710.0,
      "step": 8100
    },
    {
      "entropy": 5.828194952011108,
      "epoch": 0.6809493803822726,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004959890026180677,
      "loss": 5.691,
      "mean_token_accuracy": 0.1521795153617859,
      "num_tokens": 14962814.0,
      "step": 8105
    },
    {
      "entropy": 5.653000354766846,
      "epoch": 0.6813694601974375,
      "grad_norm": 1.46875,
      "learning_rate": 0.00049598337138488,
      "loss": 5.5758,
      "mean_token_accuracy": 0.15957341492176055,
      "num_tokens": 14971631.0,
      "step": 8110
    },
    {
      "entropy": 5.725569152832032,
      "epoch": 0.6817895400126024,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004959777362370855,
      "loss": 5.5609,
      "mean_token_accuracy": 0.1567662850022316,
      "num_tokens": 14980528.0,
      "step": 8115
    },
    {
      "entropy": 5.756756782531738,
      "epoch": 0.6822096198277673,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004959720971747843,
      "loss": 5.5812,
      "mean_token_accuracy": 0.15582162886857986,
      "num_tokens": 14989331.0,
      "step": 8120
    },
    {
      "entropy": 5.713353538513184,
      "epoch": 0.6826296996429322,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004959664541980762,
      "loss": 5.5831,
      "mean_token_accuracy": 0.15746317207813262,
      "num_tokens": 14999403.0,
      "step": 8125
    },
    {
      "entropy": 5.753393220901489,
      "epoch": 0.6830497794580971,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004959608073070612,
      "loss": 5.6785,
      "mean_token_accuracy": 0.14850740507245064,
      "num_tokens": 15009388.0,
      "step": 8130
    },
    {
      "entropy": 5.798916053771973,
      "epoch": 0.6834698592732619,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004959551565018392,
      "loss": 5.5994,
      "mean_token_accuracy": 0.1609173148870468,
      "num_tokens": 15018586.0,
      "step": 8135
    },
    {
      "entropy": 5.737577104568482,
      "epoch": 0.6838899390884268,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004959495017825104,
      "loss": 5.611,
      "mean_token_accuracy": 0.1578981950879097,
      "num_tokens": 15027982.0,
      "step": 8140
    },
    {
      "entropy": 5.684477043151856,
      "epoch": 0.6843100189035917,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004959438431491749,
      "loss": 5.6098,
      "mean_token_accuracy": 0.1577099531888962,
      "num_tokens": 15037103.0,
      "step": 8145
    },
    {
      "entropy": 5.697509860992431,
      "epoch": 0.6847300987187566,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000495938180601933,
      "loss": 5.7013,
      "mean_token_accuracy": 0.1484990067780018,
      "num_tokens": 15046739.0,
      "step": 8150
    },
    {
      "entropy": 5.794940948486328,
      "epoch": 0.6851501785339215,
      "grad_norm": 1.875,
      "learning_rate": 0.0004959325141408851,
      "loss": 5.6373,
      "mean_token_accuracy": 0.15557540208101273,
      "num_tokens": 15056586.0,
      "step": 8155
    },
    {
      "entropy": 5.724914121627807,
      "epoch": 0.6855702583490864,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004959268437661313,
      "loss": 5.6297,
      "mean_token_accuracy": 0.15379536151885986,
      "num_tokens": 15066622.0,
      "step": 8160
    },
    {
      "entropy": 5.794011211395263,
      "epoch": 0.6859903381642513,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004959211694777724,
      "loss": 5.6106,
      "mean_token_accuracy": 0.15808048397302626,
      "num_tokens": 15075415.0,
      "step": 8165
    },
    {
      "entropy": 5.751935195922852,
      "epoch": 0.686410417979416,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004959154912759086,
      "loss": 5.59,
      "mean_token_accuracy": 0.15190571993589402,
      "num_tokens": 15085087.0,
      "step": 8170
    },
    {
      "entropy": 5.736638736724854,
      "epoch": 0.6868304977945809,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004959098091606406,
      "loss": 5.6039,
      "mean_token_accuracy": 0.15554135516285897,
      "num_tokens": 15093580.0,
      "step": 8175
    },
    {
      "entropy": 5.630563879013062,
      "epoch": 0.6872505776097458,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004959041231320692,
      "loss": 5.5887,
      "mean_token_accuracy": 0.16016243845224382,
      "num_tokens": 15104033.0,
      "step": 8180
    },
    {
      "entropy": 5.761919355392456,
      "epoch": 0.6876706574249107,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004958984331902951,
      "loss": 5.6592,
      "mean_token_accuracy": 0.14851288944482804,
      "num_tokens": 15113164.0,
      "step": 8185
    },
    {
      "entropy": 5.72430567741394,
      "epoch": 0.6880907372400756,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004958927393354188,
      "loss": 5.6022,
      "mean_token_accuracy": 0.15889295786619187,
      "num_tokens": 15122215.0,
      "step": 8190
    },
    {
      "entropy": 5.736812925338745,
      "epoch": 0.6885108170552405,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004958870415675415,
      "loss": 5.5838,
      "mean_token_accuracy": 0.15310561507940293,
      "num_tokens": 15130877.0,
      "step": 8195
    },
    {
      "entropy": 5.734648513793945,
      "epoch": 0.6889308968704054,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004958813398867639,
      "loss": 5.5748,
      "mean_token_accuracy": 0.1602252170443535,
      "num_tokens": 15140227.0,
      "step": 8200
    },
    {
      "entropy": 5.8443381786346436,
      "epoch": 0.6893509766855702,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004958756342931872,
      "loss": 5.7447,
      "mean_token_accuracy": 0.1436396934092045,
      "num_tokens": 15150006.0,
      "step": 8205
    },
    {
      "entropy": 5.749261903762817,
      "epoch": 0.6897710565007351,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004958699247869122,
      "loss": 5.6456,
      "mean_token_accuracy": 0.15174706876277924,
      "num_tokens": 15160032.0,
      "step": 8210
    },
    {
      "entropy": 5.75321626663208,
      "epoch": 0.6901911363159,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004958642113680404,
      "loss": 5.58,
      "mean_token_accuracy": 0.16214127987623214,
      "num_tokens": 15168966.0,
      "step": 8215
    },
    {
      "entropy": 5.870010280609131,
      "epoch": 0.6906112161310649,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004958584940366727,
      "loss": 5.7572,
      "mean_token_accuracy": 0.14761550426483155,
      "num_tokens": 15179337.0,
      "step": 8220
    },
    {
      "entropy": 5.803754615783691,
      "epoch": 0.6910312959462298,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004958527727929106,
      "loss": 5.6701,
      "mean_token_accuracy": 0.1573884293437004,
      "num_tokens": 15188395.0,
      "step": 8225
    },
    {
      "entropy": 5.746722412109375,
      "epoch": 0.6914513757613947,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004958470476368552,
      "loss": 5.6025,
      "mean_token_accuracy": 0.16065260544419288,
      "num_tokens": 15198669.0,
      "step": 8230
    },
    {
      "entropy": 5.7148209571838375,
      "epoch": 0.6918714555765595,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004958413185686082,
      "loss": 5.6004,
      "mean_token_accuracy": 0.15373986065387726,
      "num_tokens": 15207371.0,
      "step": 8235
    },
    {
      "entropy": 5.766255283355713,
      "epoch": 0.6922915353917244,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004958355855882709,
      "loss": 5.6358,
      "mean_token_accuracy": 0.1554594576358795,
      "num_tokens": 15215694.0,
      "step": 8240
    },
    {
      "entropy": 5.766929721832275,
      "epoch": 0.6927116152068893,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000495829848695945,
      "loss": 5.6244,
      "mean_token_accuracy": 0.15312992185354232,
      "num_tokens": 15224963.0,
      "step": 8245
    },
    {
      "entropy": 5.648315715789795,
      "epoch": 0.6931316950220542,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000495824107891732,
      "loss": 5.43,
      "mean_token_accuracy": 0.1609826758503914,
      "num_tokens": 15233569.0,
      "step": 8250
    },
    {
      "entropy": 5.702505493164063,
      "epoch": 0.6935517748372191,
      "grad_norm": 1.75,
      "learning_rate": 0.0004958183631757336,
      "loss": 5.6174,
      "mean_token_accuracy": 0.1533745378255844,
      "num_tokens": 15242671.0,
      "step": 8255
    },
    {
      "entropy": 5.715704250335693,
      "epoch": 0.693971854652384,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004958126145480517,
      "loss": 5.5655,
      "mean_token_accuracy": 0.1603790357708931,
      "num_tokens": 15251698.0,
      "step": 8260
    },
    {
      "entropy": 5.793996429443359,
      "epoch": 0.6943919344675489,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004958068620087879,
      "loss": 5.7033,
      "mean_token_accuracy": 0.15312060862779617,
      "num_tokens": 15260608.0,
      "step": 8265
    },
    {
      "entropy": 5.725742197036743,
      "epoch": 0.6948120142827137,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004958011055580443,
      "loss": 5.5631,
      "mean_token_accuracy": 0.15500877648591996,
      "num_tokens": 15268866.0,
      "step": 8270
    },
    {
      "entropy": 5.648518466949463,
      "epoch": 0.6952320940978786,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004957953451959229,
      "loss": 5.5272,
      "mean_token_accuracy": 0.16275469362735748,
      "num_tokens": 15277600.0,
      "step": 8275
    },
    {
      "entropy": 5.730481147766113,
      "epoch": 0.6956521739130435,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004957895809225254,
      "loss": 5.5546,
      "mean_token_accuracy": 0.15924442559480667,
      "num_tokens": 15286016.0,
      "step": 8280
    },
    {
      "entropy": 5.750138473510742,
      "epoch": 0.6960722537282084,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004957838127379544,
      "loss": 5.6063,
      "mean_token_accuracy": 0.1584298923611641,
      "num_tokens": 15294676.0,
      "step": 8285
    },
    {
      "entropy": 5.749259424209595,
      "epoch": 0.6964923335433733,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004957780406423118,
      "loss": 5.5886,
      "mean_token_accuracy": 0.1535394623875618,
      "num_tokens": 15304084.0,
      "step": 8290
    },
    {
      "entropy": 5.743555212020874,
      "epoch": 0.6969124133585382,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004957722646356999,
      "loss": 5.5989,
      "mean_token_accuracy": 0.15085933804512025,
      "num_tokens": 15314182.0,
      "step": 8295
    },
    {
      "entropy": 5.774941158294678,
      "epoch": 0.697332493173703,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004957664847182209,
      "loss": 5.7173,
      "mean_token_accuracy": 0.14531809762120246,
      "num_tokens": 15324213.0,
      "step": 8300
    },
    {
      "entropy": 5.836635112762451,
      "epoch": 0.6977525729888678,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004957607008899774,
      "loss": 5.6305,
      "mean_token_accuracy": 0.14892250299453735,
      "num_tokens": 15333122.0,
      "step": 8305
    },
    {
      "entropy": 5.788960552215576,
      "epoch": 0.6981726528040327,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004957549131510717,
      "loss": 5.7278,
      "mean_token_accuracy": 0.14765103310346603,
      "num_tokens": 15342199.0,
      "step": 8310
    },
    {
      "entropy": 5.840366983413697,
      "epoch": 0.6985927326191976,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004957491215016065,
      "loss": 5.6792,
      "mean_token_accuracy": 0.15214879661798478,
      "num_tokens": 15352463.0,
      "step": 8315
    },
    {
      "entropy": 5.700908565521241,
      "epoch": 0.6990128124343625,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004957433259416841,
      "loss": 5.5231,
      "mean_token_accuracy": 0.1583988130092621,
      "num_tokens": 15361815.0,
      "step": 8320
    },
    {
      "entropy": 5.767258024215698,
      "epoch": 0.6994328922495274,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004957375264714075,
      "loss": 5.6439,
      "mean_token_accuracy": 0.14546388015151024,
      "num_tokens": 15371773.0,
      "step": 8325
    },
    {
      "entropy": 5.697303724288941,
      "epoch": 0.6998529720646923,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004957317230908792,
      "loss": 5.5916,
      "mean_token_accuracy": 0.15627994984388352,
      "num_tokens": 15380881.0,
      "step": 8330
    },
    {
      "entropy": 5.672244882583618,
      "epoch": 0.7002730518798572,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004957259158002022,
      "loss": 5.4522,
      "mean_token_accuracy": 0.16274693012237548,
      "num_tokens": 15389310.0,
      "step": 8335
    },
    {
      "entropy": 5.663344621658325,
      "epoch": 0.700693131695022,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004957201045994791,
      "loss": 5.5548,
      "mean_token_accuracy": 0.15098515748977662,
      "num_tokens": 15398584.0,
      "step": 8340
    },
    {
      "entropy": 5.729818964004517,
      "epoch": 0.7011132115101869,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004957142894888131,
      "loss": 5.5935,
      "mean_token_accuracy": 0.16061321198940276,
      "num_tokens": 15407208.0,
      "step": 8345
    },
    {
      "entropy": 5.745049524307251,
      "epoch": 0.7015332913253518,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004957084704683071,
      "loss": 5.639,
      "mean_token_accuracy": 0.15271784067153932,
      "num_tokens": 15416474.0,
      "step": 8350
    },
    {
      "entropy": 5.730433702468872,
      "epoch": 0.7019533711405167,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004957026475380642,
      "loss": 5.6296,
      "mean_token_accuracy": 0.15648382306098937,
      "num_tokens": 15426101.0,
      "step": 8355
    },
    {
      "entropy": 5.7902411937713625,
      "epoch": 0.7023734509556816,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004956968206981875,
      "loss": 5.6721,
      "mean_token_accuracy": 0.15503493845462799,
      "num_tokens": 15435910.0,
      "step": 8360
    },
    {
      "entropy": 5.79659628868103,
      "epoch": 0.7027935307708465,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004956909899487803,
      "loss": 5.6948,
      "mean_token_accuracy": 0.1535898119211197,
      "num_tokens": 15445494.0,
      "step": 8365
    },
    {
      "entropy": 5.741454792022705,
      "epoch": 0.7032136105860114,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004956851552899459,
      "loss": 5.5931,
      "mean_token_accuracy": 0.16135696619749068,
      "num_tokens": 15455332.0,
      "step": 8370
    },
    {
      "entropy": 5.763624382019043,
      "epoch": 0.7036336904011762,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004956793167217874,
      "loss": 5.65,
      "mean_token_accuracy": 0.1525983467698097,
      "num_tokens": 15464241.0,
      "step": 8375
    },
    {
      "entropy": 5.82790732383728,
      "epoch": 0.7040537702163411,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004956734742444087,
      "loss": 5.6567,
      "mean_token_accuracy": 0.1559632509946823,
      "num_tokens": 15473473.0,
      "step": 8380
    },
    {
      "entropy": 5.716184997558594,
      "epoch": 0.704473850031506,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004956676278579129,
      "loss": 5.5329,
      "mean_token_accuracy": 0.16085591912269592,
      "num_tokens": 15482494.0,
      "step": 8385
    },
    {
      "entropy": 5.663648986816407,
      "epoch": 0.7048939298466709,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004956617775624037,
      "loss": 5.5538,
      "mean_token_accuracy": 0.15378274619579316,
      "num_tokens": 15491180.0,
      "step": 8390
    },
    {
      "entropy": 5.7593092918396,
      "epoch": 0.7053140096618358,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004956559233579848,
      "loss": 5.6075,
      "mean_token_accuracy": 0.15238080322742462,
      "num_tokens": 15501035.0,
      "step": 8395
    },
    {
      "entropy": 5.747204065322876,
      "epoch": 0.7057340894770007,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004956500652447598,
      "loss": 5.5831,
      "mean_token_accuracy": 0.15643986463546752,
      "num_tokens": 15510191.0,
      "step": 8400
    },
    {
      "entropy": 5.691559076309204,
      "epoch": 0.7061541692921655,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004956442032228324,
      "loss": 5.6635,
      "mean_token_accuracy": 0.15496474727988244,
      "num_tokens": 15519253.0,
      "step": 8405
    },
    {
      "entropy": 5.722681283950806,
      "epoch": 0.7065742491073304,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004956383372923067,
      "loss": 5.627,
      "mean_token_accuracy": 0.15085585564374923,
      "num_tokens": 15528348.0,
      "step": 8410
    },
    {
      "entropy": 5.935694408416748,
      "epoch": 0.7069943289224953,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004956324674532864,
      "loss": 5.7185,
      "mean_token_accuracy": 0.14439277797937394,
      "num_tokens": 15537557.0,
      "step": 8415
    },
    {
      "entropy": 5.8467813491821286,
      "epoch": 0.7074144087376601,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004956265937058757,
      "loss": 5.654,
      "mean_token_accuracy": 0.15044144690036773,
      "num_tokens": 15546745.0,
      "step": 8420
    },
    {
      "entropy": 5.744271945953369,
      "epoch": 0.707834488552825,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004956207160501784,
      "loss": 5.5384,
      "mean_token_accuracy": 0.16095120459794998,
      "num_tokens": 15555532.0,
      "step": 8425
    },
    {
      "entropy": 5.7119933605194095,
      "epoch": 0.70825456836799,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004956148344862987,
      "loss": 5.5878,
      "mean_token_accuracy": 0.16213922202587128,
      "num_tokens": 15564189.0,
      "step": 8430
    },
    {
      "entropy": 5.626652574539184,
      "epoch": 0.7086746481831548,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004956089490143408,
      "loss": 5.6232,
      "mean_token_accuracy": 0.15468233972787857,
      "num_tokens": 15574116.0,
      "step": 8435
    },
    {
      "entropy": 5.845960283279419,
      "epoch": 0.7090947279983196,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004956030596344089,
      "loss": 5.6162,
      "mean_token_accuracy": 0.15749153792858123,
      "num_tokens": 15583031.0,
      "step": 8440
    },
    {
      "entropy": 5.82060866355896,
      "epoch": 0.7095148078134845,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004955971663466075,
      "loss": 5.7384,
      "mean_token_accuracy": 0.1541066735982895,
      "num_tokens": 15592576.0,
      "step": 8445
    },
    {
      "entropy": 5.800911283493042,
      "epoch": 0.7099348876286494,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004955912691510407,
      "loss": 5.6751,
      "mean_token_accuracy": 0.15478230714797975,
      "num_tokens": 15601065.0,
      "step": 8450
    },
    {
      "entropy": 5.75335373878479,
      "epoch": 0.7103549674438143,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004955853680478134,
      "loss": 5.5987,
      "mean_token_accuracy": 0.1501637101173401,
      "num_tokens": 15610112.0,
      "step": 8455
    },
    {
      "entropy": 5.803425455093384,
      "epoch": 0.7107750472589792,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004955794630370297,
      "loss": 5.5779,
      "mean_token_accuracy": 0.15560339987277985,
      "num_tokens": 15618890.0,
      "step": 8460
    },
    {
      "entropy": 5.72348952293396,
      "epoch": 0.7111951270741441,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004955735541187945,
      "loss": 5.6142,
      "mean_token_accuracy": 0.15508751571178436,
      "num_tokens": 15627678.0,
      "step": 8465
    },
    {
      "entropy": 5.737122440338135,
      "epoch": 0.711615206889309,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004955676412932124,
      "loss": 5.5994,
      "mean_token_accuracy": 0.15686758160591124,
      "num_tokens": 15636833.0,
      "step": 8470
    },
    {
      "entropy": 5.708105039596558,
      "epoch": 0.7120352867044738,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004955617245603881,
      "loss": 5.6109,
      "mean_token_accuracy": 0.1494527980685234,
      "num_tokens": 15646571.0,
      "step": 8475
    },
    {
      "entropy": 5.771750783920288,
      "epoch": 0.7124553665196387,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004955558039204263,
      "loss": 5.6511,
      "mean_token_accuracy": 0.16095416396856307,
      "num_tokens": 15654907.0,
      "step": 8480
    },
    {
      "entropy": 5.806697988510132,
      "epoch": 0.7128754463348036,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004955498793734321,
      "loss": 5.5932,
      "mean_token_accuracy": 0.15864523649215698,
      "num_tokens": 15664336.0,
      "step": 8485
    },
    {
      "entropy": 5.8073060512542725,
      "epoch": 0.7132955261499685,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004955439509195103,
      "loss": 5.6601,
      "mean_token_accuracy": 0.15415582954883575,
      "num_tokens": 15674000.0,
      "step": 8490
    },
    {
      "entropy": 5.826253652572632,
      "epoch": 0.7137156059651334,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004955380185587661,
      "loss": 5.6496,
      "mean_token_accuracy": 0.15672145187854766,
      "num_tokens": 15684214.0,
      "step": 8495
    },
    {
      "entropy": 5.762752962112427,
      "epoch": 0.7141356857802983,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004955320822913043,
      "loss": 5.6571,
      "mean_token_accuracy": 0.1523528926074505,
      "num_tokens": 15693546.0,
      "step": 8500
    },
    {
      "entropy": 5.711481142044067,
      "epoch": 0.7145557655954632,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004955261421172302,
      "loss": 5.5738,
      "mean_token_accuracy": 0.15294577926397324,
      "num_tokens": 15702310.0,
      "step": 8505
    },
    {
      "entropy": 5.757646560668945,
      "epoch": 0.714975845410628,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004955201980366493,
      "loss": 5.647,
      "mean_token_accuracy": 0.154372838139534,
      "num_tokens": 15711544.0,
      "step": 8510
    },
    {
      "entropy": 5.690837240219116,
      "epoch": 0.7153959252257929,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004955142500496665,
      "loss": 5.5216,
      "mean_token_accuracy": 0.15888839215040207,
      "num_tokens": 15720914.0,
      "step": 8515
    },
    {
      "entropy": 5.809607696533203,
      "epoch": 0.7158160050409578,
      "grad_norm": 1.5,
      "learning_rate": 0.0004955082981563872,
      "loss": 5.6077,
      "mean_token_accuracy": 0.1486271560192108,
      "num_tokens": 15729825.0,
      "step": 8520
    },
    {
      "entropy": 5.724713563919067,
      "epoch": 0.7162360848561227,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000495502342356917,
      "loss": 5.6217,
      "mean_token_accuracy": 0.1557765081524849,
      "num_tokens": 15739649.0,
      "step": 8525
    },
    {
      "entropy": 5.726509523391724,
      "epoch": 0.7166561646712876,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004954963826513614,
      "loss": 5.5154,
      "mean_token_accuracy": 0.1566923290491104,
      "num_tokens": 15747805.0,
      "step": 8530
    },
    {
      "entropy": 5.80653567314148,
      "epoch": 0.7170762444864525,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000495490419039826,
      "loss": 5.6602,
      "mean_token_accuracy": 0.15109436362981796,
      "num_tokens": 15757267.0,
      "step": 8535
    },
    {
      "entropy": 5.7422764778137205,
      "epoch": 0.7174963243016174,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004954844515224162,
      "loss": 5.6074,
      "mean_token_accuracy": 0.1531700074672699,
      "num_tokens": 15767412.0,
      "step": 8540
    },
    {
      "entropy": 5.669848108291626,
      "epoch": 0.7179164041167821,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004954784800992379,
      "loss": 5.6081,
      "mean_token_accuracy": 0.1509648084640503,
      "num_tokens": 15776813.0,
      "step": 8545
    },
    {
      "entropy": 5.8059522151947025,
      "epoch": 0.718336483931947,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004954725047703969,
      "loss": 5.6531,
      "mean_token_accuracy": 0.15384986773133277,
      "num_tokens": 15786258.0,
      "step": 8550
    },
    {
      "entropy": 5.800448894500732,
      "epoch": 0.7187565637471119,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000495466525535999,
      "loss": 5.6218,
      "mean_token_accuracy": 0.1543135389685631,
      "num_tokens": 15795673.0,
      "step": 8555
    },
    {
      "entropy": 5.757376861572266,
      "epoch": 0.7191766435622768,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004954605423961501,
      "loss": 5.634,
      "mean_token_accuracy": 0.1545005664229393,
      "num_tokens": 15805050.0,
      "step": 8560
    },
    {
      "entropy": 5.695555400848389,
      "epoch": 0.7195967233774417,
      "grad_norm": 1.5,
      "learning_rate": 0.0004954545553509562,
      "loss": 5.5764,
      "mean_token_accuracy": 0.16470458656549453,
      "num_tokens": 15813347.0,
      "step": 8565
    },
    {
      "entropy": 5.851338243484497,
      "epoch": 0.7200168031926066,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004954485644005235,
      "loss": 5.7053,
      "mean_token_accuracy": 0.14976215064525605,
      "num_tokens": 15823528.0,
      "step": 8570
    },
    {
      "entropy": 5.810816144943237,
      "epoch": 0.7204368830077714,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004954425695449578,
      "loss": 5.597,
      "mean_token_accuracy": 0.15381426066160203,
      "num_tokens": 15832727.0,
      "step": 8575
    },
    {
      "entropy": 5.766140604019165,
      "epoch": 0.7208569628229363,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004954365707843657,
      "loss": 5.6661,
      "mean_token_accuracy": 0.1463111013174057,
      "num_tokens": 15842402.0,
      "step": 8580
    },
    {
      "entropy": 5.698196029663086,
      "epoch": 0.7212770426381012,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004954305681188531,
      "loss": 5.5389,
      "mean_token_accuracy": 0.1566384069621563,
      "num_tokens": 15850886.0,
      "step": 8585
    },
    {
      "entropy": 5.9173768043518065,
      "epoch": 0.7216971224532661,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004954245615485265,
      "loss": 5.8176,
      "mean_token_accuracy": 0.1536174975335598,
      "num_tokens": 15860093.0,
      "step": 8590
    },
    {
      "entropy": 5.758410358428955,
      "epoch": 0.722117202268431,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004954185510734924,
      "loss": 5.5379,
      "mean_token_accuracy": 0.1607280820608139,
      "num_tokens": 15868681.0,
      "step": 8595
    },
    {
      "entropy": 5.739753103256225,
      "epoch": 0.7225372820835959,
      "grad_norm": 1.5,
      "learning_rate": 0.0004954125366938571,
      "loss": 5.6224,
      "mean_token_accuracy": 0.1589464083313942,
      "num_tokens": 15878041.0,
      "step": 8600
    },
    {
      "entropy": 5.717496919631958,
      "epoch": 0.7229573618987608,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004954065184097271,
      "loss": 5.6144,
      "mean_token_accuracy": 0.15493135452270507,
      "num_tokens": 15887562.0,
      "step": 8605
    },
    {
      "entropy": 5.765741586685181,
      "epoch": 0.7233774417139256,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004954004962212092,
      "loss": 5.5471,
      "mean_token_accuracy": 0.16325860619544982,
      "num_tokens": 15896480.0,
      "step": 8610
    },
    {
      "entropy": 5.920656204223633,
      "epoch": 0.7237975215290905,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004953944701284101,
      "loss": 5.7553,
      "mean_token_accuracy": 0.14824794083833695,
      "num_tokens": 15906743.0,
      "step": 8615
    },
    {
      "entropy": 5.760147953033448,
      "epoch": 0.7242176013442554,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004953884401314363,
      "loss": 5.7146,
      "mean_token_accuracy": 0.1402403138577938,
      "num_tokens": 15915981.0,
      "step": 8620
    },
    {
      "entropy": 5.728390407562256,
      "epoch": 0.7246376811594203,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004953824062303949,
      "loss": 5.5455,
      "mean_token_accuracy": 0.15434545278549194,
      "num_tokens": 15924117.0,
      "step": 8625
    },
    {
      "entropy": 5.702328252792358,
      "epoch": 0.7250577609745852,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004953763684253926,
      "loss": 5.5834,
      "mean_token_accuracy": 0.1618655726313591,
      "num_tokens": 15933124.0,
      "step": 8630
    },
    {
      "entropy": 5.671606540679932,
      "epoch": 0.7254778407897501,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004953703267165364,
      "loss": 5.4707,
      "mean_token_accuracy": 0.15732883363962175,
      "num_tokens": 15942422.0,
      "step": 8635
    },
    {
      "entropy": 5.790132761001587,
      "epoch": 0.725897920604915,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004953642811039332,
      "loss": 5.6894,
      "mean_token_accuracy": 0.15469479262828828,
      "num_tokens": 15950989.0,
      "step": 8640
    },
    {
      "entropy": 5.856866359710693,
      "epoch": 0.7263180004200798,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004953582315876904,
      "loss": 5.6948,
      "mean_token_accuracy": 0.15004287138581276,
      "num_tokens": 15959659.0,
      "step": 8645
    },
    {
      "entropy": 5.730572128295899,
      "epoch": 0.7267380802352447,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000495352178167915,
      "loss": 5.5797,
      "mean_token_accuracy": 0.16224073469638825,
      "num_tokens": 15968102.0,
      "step": 8650
    },
    {
      "entropy": 5.820402717590332,
      "epoch": 0.7271581600504096,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004953461208447143,
      "loss": 5.6882,
      "mean_token_accuracy": 0.14849866628646852,
      "num_tokens": 15977705.0,
      "step": 8655
    },
    {
      "entropy": 5.7769698143005375,
      "epoch": 0.7275782398655745,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004953400596181953,
      "loss": 5.6721,
      "mean_token_accuracy": 0.1488412395119667,
      "num_tokens": 15986703.0,
      "step": 8660
    },
    {
      "entropy": 5.752463150024414,
      "epoch": 0.7279983196807394,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004953339944884657,
      "loss": 5.5871,
      "mean_token_accuracy": 0.1577693849802017,
      "num_tokens": 15995672.0,
      "step": 8665
    },
    {
      "entropy": 5.61082501411438,
      "epoch": 0.7284183994959043,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004953279254556329,
      "loss": 5.5399,
      "mean_token_accuracy": 0.1641200602054596,
      "num_tokens": 16004437.0,
      "step": 8670
    },
    {
      "entropy": 5.707981443405151,
      "epoch": 0.7288384793110692,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004953218525198043,
      "loss": 5.575,
      "mean_token_accuracy": 0.15932455360889436,
      "num_tokens": 16012847.0,
      "step": 8675
    },
    {
      "entropy": 5.810865259170532,
      "epoch": 0.7292585591262339,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004953157756810876,
      "loss": 5.6175,
      "mean_token_accuracy": 0.15550039410591127,
      "num_tokens": 16022213.0,
      "step": 8680
    },
    {
      "entropy": 5.780650091171265,
      "epoch": 0.7296786389413988,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004953096949395902,
      "loss": 5.6738,
      "mean_token_accuracy": 0.15553813129663469,
      "num_tokens": 16031411.0,
      "step": 8685
    },
    {
      "entropy": 5.803027772903443,
      "epoch": 0.7300987187565637,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004953036102954202,
      "loss": 5.7067,
      "mean_token_accuracy": 0.15027424544095994,
      "num_tokens": 16041227.0,
      "step": 8690
    },
    {
      "entropy": 5.769024753570557,
      "epoch": 0.7305187985717286,
      "grad_norm": 1.5,
      "learning_rate": 0.0004952975217486852,
      "loss": 5.5608,
      "mean_token_accuracy": 0.15838102549314498,
      "num_tokens": 16049777.0,
      "step": 8695
    },
    {
      "entropy": 5.784084749221802,
      "epoch": 0.7309388783868935,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004952914292994928,
      "loss": 5.6425,
      "mean_token_accuracy": 0.15665655136108397,
      "num_tokens": 16059093.0,
      "step": 8700
    },
    {
      "entropy": 5.746081352233887,
      "epoch": 0.7313589582020584,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004952853329479514,
      "loss": 5.6752,
      "mean_token_accuracy": 0.1572483465075493,
      "num_tokens": 16068550.0,
      "step": 8705
    },
    {
      "entropy": 5.760213088989258,
      "epoch": 0.7317790380172233,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004952792326941686,
      "loss": 5.6979,
      "mean_token_accuracy": 0.15022268071770667,
      "num_tokens": 16078286.0,
      "step": 8710
    },
    {
      "entropy": 5.767646551132202,
      "epoch": 0.7321991178323881,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004952731285382527,
      "loss": 5.638,
      "mean_token_accuracy": 0.15017070174217223,
      "num_tokens": 16087560.0,
      "step": 8715
    },
    {
      "entropy": 5.730300951004028,
      "epoch": 0.732619197647553,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004952670204803118,
      "loss": 5.6012,
      "mean_token_accuracy": 0.16169655174016953,
      "num_tokens": 16097478.0,
      "step": 8720
    },
    {
      "entropy": 5.802074241638183,
      "epoch": 0.7330392774627179,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004952609085204539,
      "loss": 5.6932,
      "mean_token_accuracy": 0.16233148872852327,
      "num_tokens": 16106884.0,
      "step": 8725
    },
    {
      "entropy": 5.745241928100586,
      "epoch": 0.7334593572778828,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004952547926587876,
      "loss": 5.6056,
      "mean_token_accuracy": 0.14735328257083893,
      "num_tokens": 16115689.0,
      "step": 8730
    },
    {
      "entropy": 5.717710161209107,
      "epoch": 0.7338794370930477,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004952486728954209,
      "loss": 5.5406,
      "mean_token_accuracy": 0.1627890720963478,
      "num_tokens": 16125237.0,
      "step": 8735
    },
    {
      "entropy": 5.711509656906128,
      "epoch": 0.7342995169082126,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004952425492304624,
      "loss": 5.5612,
      "mean_token_accuracy": 0.1623712345957756,
      "num_tokens": 16133940.0,
      "step": 8740
    },
    {
      "entropy": 5.761761999130249,
      "epoch": 0.7347195967233774,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004952364216640207,
      "loss": 5.6536,
      "mean_token_accuracy": 0.1498600222170353,
      "num_tokens": 16143256.0,
      "step": 8745
    },
    {
      "entropy": 5.749999523162842,
      "epoch": 0.7351396765385423,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000495230290196204,
      "loss": 5.5377,
      "mean_token_accuracy": 0.1550111636519432,
      "num_tokens": 16153259.0,
      "step": 8750
    },
    {
      "entropy": 5.784195375442505,
      "epoch": 0.7355597563537072,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004952241548271212,
      "loss": 5.7846,
      "mean_token_accuracy": 0.14801348447799684,
      "num_tokens": 16162125.0,
      "step": 8755
    },
    {
      "entropy": 5.844126415252686,
      "epoch": 0.7359798361688721,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004952180155568809,
      "loss": 5.6959,
      "mean_token_accuracy": 0.15129489600658416,
      "num_tokens": 16171680.0,
      "step": 8760
    },
    {
      "entropy": 5.847824907302856,
      "epoch": 0.736399915984037,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004952118723855919,
      "loss": 5.7082,
      "mean_token_accuracy": 0.15476434975862502,
      "num_tokens": 16181559.0,
      "step": 8765
    },
    {
      "entropy": 5.748734664916992,
      "epoch": 0.7368199957992019,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004952057253133628,
      "loss": 5.6461,
      "mean_token_accuracy": 0.1511749282479286,
      "num_tokens": 16190611.0,
      "step": 8770
    },
    {
      "entropy": 5.809700012207031,
      "epoch": 0.7372400756143668,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004951995743403028,
      "loss": 5.6609,
      "mean_token_accuracy": 0.1494298830628395,
      "num_tokens": 16200156.0,
      "step": 8775
    },
    {
      "entropy": 5.723226308822632,
      "epoch": 0.7376601554295316,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004951934194665208,
      "loss": 5.626,
      "mean_token_accuracy": 0.14901441484689712,
      "num_tokens": 16209808.0,
      "step": 8780
    },
    {
      "entropy": 5.698558235168457,
      "epoch": 0.7380802352446965,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004951872606921257,
      "loss": 5.5831,
      "mean_token_accuracy": 0.1541206181049347,
      "num_tokens": 16219243.0,
      "step": 8785
    },
    {
      "entropy": 5.68091344833374,
      "epoch": 0.7385003150598614,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004951810980172265,
      "loss": 5.5962,
      "mean_token_accuracy": 0.1625403195619583,
      "num_tokens": 16228180.0,
      "step": 8790
    },
    {
      "entropy": 5.785898876190186,
      "epoch": 0.7389203948750263,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004951749314419327,
      "loss": 5.6209,
      "mean_token_accuracy": 0.1556001588702202,
      "num_tokens": 16237045.0,
      "step": 8795
    },
    {
      "entropy": 5.7458240509033205,
      "epoch": 0.7393404746901912,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004951687609663533,
      "loss": 5.5277,
      "mean_token_accuracy": 0.16299475580453873,
      "num_tokens": 16245307.0,
      "step": 8800
    },
    {
      "entropy": 5.7147376537323,
      "epoch": 0.739760554505356,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004951625865905977,
      "loss": 5.5793,
      "mean_token_accuracy": 0.1496004968881607,
      "num_tokens": 16255047.0,
      "step": 8805
    },
    {
      "entropy": 5.714106130599975,
      "epoch": 0.740180634320521,
      "grad_norm": 3.0,
      "learning_rate": 0.0004951564083147753,
      "loss": 5.6177,
      "mean_token_accuracy": 0.16076902598142623,
      "num_tokens": 16264969.0,
      "step": 8810
    },
    {
      "entropy": 5.778882122039795,
      "epoch": 0.7406007141356857,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004951502261389953,
      "loss": 5.7085,
      "mean_token_accuracy": 0.14696210622787476,
      "num_tokens": 16274757.0,
      "step": 8815
    },
    {
      "entropy": 5.745514392852783,
      "epoch": 0.7410207939508506,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004951440400633677,
      "loss": 5.6118,
      "mean_token_accuracy": 0.16788006722927093,
      "num_tokens": 16283409.0,
      "step": 8820
    },
    {
      "entropy": 5.71191258430481,
      "epoch": 0.7414408737660155,
      "grad_norm": 1.875,
      "learning_rate": 0.0004951378500880015,
      "loss": 5.5751,
      "mean_token_accuracy": 0.15628346055746078,
      "num_tokens": 16293206.0,
      "step": 8825
    },
    {
      "entropy": 5.776321268081665,
      "epoch": 0.7418609535811804,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004951316562130067,
      "loss": 5.6155,
      "mean_token_accuracy": 0.15167658925056457,
      "num_tokens": 16303121.0,
      "step": 8830
    },
    {
      "entropy": 5.771509647369385,
      "epoch": 0.7422810333963453,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000495125458438493,
      "loss": 5.5594,
      "mean_token_accuracy": 0.1614787310361862,
      "num_tokens": 16312710.0,
      "step": 8835
    },
    {
      "entropy": 5.833417177200317,
      "epoch": 0.7427011132115102,
      "grad_norm": 1.375,
      "learning_rate": 0.0004951192567645702,
      "loss": 5.768,
      "mean_token_accuracy": 0.14854381531476973,
      "num_tokens": 16322280.0,
      "step": 8840
    },
    {
      "entropy": 5.673296499252319,
      "epoch": 0.7431211930266751,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004951130511913481,
      "loss": 5.6098,
      "mean_token_accuracy": 0.1588355764746666,
      "num_tokens": 16331656.0,
      "step": 8845
    },
    {
      "entropy": 5.734766006469727,
      "epoch": 0.7435412728418399,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004951068417189366,
      "loss": 5.63,
      "mean_token_accuracy": 0.15699938535690308,
      "num_tokens": 16341074.0,
      "step": 8850
    },
    {
      "entropy": 5.8237183570861815,
      "epoch": 0.7439613526570048,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004951006283474457,
      "loss": 5.6113,
      "mean_token_accuracy": 0.15277699157595634,
      "num_tokens": 16350097.0,
      "step": 8855
    },
    {
      "entropy": 5.658589601516724,
      "epoch": 0.7443814324721697,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004950944110769856,
      "loss": 5.5342,
      "mean_token_accuracy": 0.15926240533590316,
      "num_tokens": 16359274.0,
      "step": 8860
    },
    {
      "entropy": 5.662064790725708,
      "epoch": 0.7448015122873346,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004950881899076663,
      "loss": 5.5024,
      "mean_token_accuracy": 0.17023404836654663,
      "num_tokens": 16368445.0,
      "step": 8865
    },
    {
      "entropy": 5.7490709781646725,
      "epoch": 0.7452215921024995,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004950819648395979,
      "loss": 5.6167,
      "mean_token_accuracy": 0.15595170110464096,
      "num_tokens": 16377689.0,
      "step": 8870
    },
    {
      "entropy": 5.68521409034729,
      "epoch": 0.7456416719176644,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000495075735872891,
      "loss": 5.5741,
      "mean_token_accuracy": 0.1552984483540058,
      "num_tokens": 16386713.0,
      "step": 8875
    },
    {
      "entropy": 5.772252321243286,
      "epoch": 0.7460617517328293,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004950695030076557,
      "loss": 5.5904,
      "mean_token_accuracy": 0.14960389882326125,
      "num_tokens": 16395390.0,
      "step": 8880
    },
    {
      "entropy": 5.771421718597412,
      "epoch": 0.7464818315479941,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004950632662440027,
      "loss": 5.6654,
      "mean_token_accuracy": 0.15664495676755905,
      "num_tokens": 16404531.0,
      "step": 8885
    },
    {
      "entropy": 5.691759538650513,
      "epoch": 0.746901911363159,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004950570255820419,
      "loss": 5.5474,
      "mean_token_accuracy": 0.15819469094276428,
      "num_tokens": 16413649.0,
      "step": 8890
    },
    {
      "entropy": 5.676243829727173,
      "epoch": 0.7473219911783239,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004950507810218843,
      "loss": 5.7051,
      "mean_token_accuracy": 0.14750183895230293,
      "num_tokens": 16423247.0,
      "step": 8895
    },
    {
      "entropy": 5.826502656936645,
      "epoch": 0.7477420709934888,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004950445325636405,
      "loss": 5.6103,
      "mean_token_accuracy": 0.15324292033910752,
      "num_tokens": 16432190.0,
      "step": 8900
    },
    {
      "entropy": 5.820600986480713,
      "epoch": 0.7481621508086537,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004950382802074211,
      "loss": 5.573,
      "mean_token_accuracy": 0.15902330875396728,
      "num_tokens": 16443091.0,
      "step": 8905
    },
    {
      "entropy": 5.668134021759033,
      "epoch": 0.7485822306238186,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004950320239533369,
      "loss": 5.6035,
      "mean_token_accuracy": 0.15703300386667252,
      "num_tokens": 16452077.0,
      "step": 8910
    },
    {
      "entropy": 5.829920911788941,
      "epoch": 0.7490023104389834,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004950257638014986,
      "loss": 5.7188,
      "mean_token_accuracy": 0.15111715570092202,
      "num_tokens": 16461893.0,
      "step": 8915
    },
    {
      "entropy": 5.807134437561035,
      "epoch": 0.7494223902541483,
      "grad_norm": 1.5,
      "learning_rate": 0.0004950194997520172,
      "loss": 5.5552,
      "mean_token_accuracy": 0.1532232567667961,
      "num_tokens": 16470904.0,
      "step": 8920
    },
    {
      "entropy": 5.714702939987182,
      "epoch": 0.7498424700693131,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004950132318050037,
      "loss": 5.6202,
      "mean_token_accuracy": 0.15520734190940857,
      "num_tokens": 16480130.0,
      "step": 8925
    },
    {
      "entropy": 5.704076814651489,
      "epoch": 0.750262549884478,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004950069599605691,
      "loss": 5.675,
      "mean_token_accuracy": 0.1532868653535843,
      "num_tokens": 16489485.0,
      "step": 8930
    },
    {
      "entropy": 5.757295370101929,
      "epoch": 0.750682629699643,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004950006842188245,
      "loss": 5.6303,
      "mean_token_accuracy": 0.15377950444817542,
      "num_tokens": 16498529.0,
      "step": 8935
    },
    {
      "entropy": 5.776675844192505,
      "epoch": 0.7511027095148078,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000494994404579881,
      "loss": 5.5551,
      "mean_token_accuracy": 0.15588496178388594,
      "num_tokens": 16508094.0,
      "step": 8940
    },
    {
      "entropy": 5.746360969543457,
      "epoch": 0.7515227893299727,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00049498812104385,
      "loss": 5.648,
      "mean_token_accuracy": 0.15265263319015504,
      "num_tokens": 16517620.0,
      "step": 8945
    },
    {
      "entropy": 5.644059801101685,
      "epoch": 0.7519428691451375,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004949818336108425,
      "loss": 5.6292,
      "mean_token_accuracy": 0.1522407725453377,
      "num_tokens": 16526720.0,
      "step": 8950
    },
    {
      "entropy": 5.75996880531311,
      "epoch": 0.7523629489603024,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004949755422809703,
      "loss": 5.6226,
      "mean_token_accuracy": 0.15166783481836318,
      "num_tokens": 16535979.0,
      "step": 8955
    },
    {
      "entropy": 5.7575671672821045,
      "epoch": 0.7527830287754673,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004949692470543446,
      "loss": 5.5063,
      "mean_token_accuracy": 0.16554583013057708,
      "num_tokens": 16544538.0,
      "step": 8960
    },
    {
      "entropy": 5.661995124816895,
      "epoch": 0.7532031085906322,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004949629479310769,
      "loss": 5.5843,
      "mean_token_accuracy": 0.15953092277050018,
      "num_tokens": 16553962.0,
      "step": 8965
    },
    {
      "entropy": 5.749283409118652,
      "epoch": 0.7536231884057971,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004949566449112788,
      "loss": 5.5071,
      "mean_token_accuracy": 0.1593552276492119,
      "num_tokens": 16562652.0,
      "step": 8970
    },
    {
      "entropy": 5.792291498184204,
      "epoch": 0.754043268220962,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004949503379950621,
      "loss": 5.5993,
      "mean_token_accuracy": 0.15694511979818343,
      "num_tokens": 16570887.0,
      "step": 8975
    },
    {
      "entropy": 5.760328102111816,
      "epoch": 0.7544633480361269,
      "grad_norm": 1.75,
      "learning_rate": 0.0004949440271825385,
      "loss": 5.7427,
      "mean_token_accuracy": 0.15050431489944457,
      "num_tokens": 16581469.0,
      "step": 8980
    },
    {
      "entropy": 5.7655247211456295,
      "epoch": 0.7548834278512917,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004949377124738196,
      "loss": 5.621,
      "mean_token_accuracy": 0.14822689667344094,
      "num_tokens": 16590213.0,
      "step": 8985
    },
    {
      "entropy": 5.710941362380981,
      "epoch": 0.7553035076664566,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004949313938690174,
      "loss": 5.6041,
      "mean_token_accuracy": 0.1535157397389412,
      "num_tokens": 16598384.0,
      "step": 8990
    },
    {
      "entropy": 5.6081578731536865,
      "epoch": 0.7557235874816215,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004949250713682438,
      "loss": 5.5748,
      "mean_token_accuracy": 0.15989954620599747,
      "num_tokens": 16607670.0,
      "step": 8995
    },
    {
      "entropy": 5.772069692611694,
      "epoch": 0.7561436672967864,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004949187449716107,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15062423944473266,
      "num_tokens": 16617560.0,
      "step": 9000
    },
    {
      "epoch": 0.7561436672967864,
      "eval_entropy": 5.61633398843715,
      "eval_loss": 5.63568115234375,
      "eval_mean_token_accuracy": 0.16088676016583953,
      "eval_num_tokens": 16617560.0,
      "eval_runtime": 27.2667,
      "eval_samples_per_second": 1370.39,
      "eval_steps_per_second": 171.308,
      "step": 9000
    },
    {
      "entropy": 5.758602285385132,
      "epoch": 0.7565637471119513,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004949124146792304,
      "loss": 5.5707,
      "mean_token_accuracy": 0.16094929501414298,
      "num_tokens": 16626038.0,
      "step": 9005
    },
    {
      "entropy": 5.69229474067688,
      "epoch": 0.7569838269271162,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004949060804912149,
      "loss": 5.5829,
      "mean_token_accuracy": 0.15550542622804642,
      "num_tokens": 16636490.0,
      "step": 9010
    },
    {
      "entropy": 5.739988565444946,
      "epoch": 0.7574039067422811,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004948997424076764,
      "loss": 5.5949,
      "mean_token_accuracy": 0.15632076263427735,
      "num_tokens": 16645369.0,
      "step": 9015
    },
    {
      "entropy": 5.8396875858306885,
      "epoch": 0.7578239865574459,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004948934004287272,
      "loss": 5.6646,
      "mean_token_accuracy": 0.15439294129610062,
      "num_tokens": 16654348.0,
      "step": 9020
    },
    {
      "entropy": 5.805532503128052,
      "epoch": 0.7582440663726108,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004948870545544796,
      "loss": 5.6685,
      "mean_token_accuracy": 0.14863917008042335,
      "num_tokens": 16664009.0,
      "step": 9025
    },
    {
      "entropy": 5.720396566390991,
      "epoch": 0.7586641461877757,
      "grad_norm": 1.453125,
      "learning_rate": 0.000494880704785046,
      "loss": 5.6789,
      "mean_token_accuracy": 0.15146776288747787,
      "num_tokens": 16674079.0,
      "step": 9030
    },
    {
      "entropy": 5.800774335861206,
      "epoch": 0.7590842260029406,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004948743511205392,
      "loss": 5.6275,
      "mean_token_accuracy": 0.1500827968120575,
      "num_tokens": 16683687.0,
      "step": 9035
    },
    {
      "entropy": 5.75395884513855,
      "epoch": 0.7595043058181055,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004948679935610712,
      "loss": 5.5145,
      "mean_token_accuracy": 0.16577970534563063,
      "num_tokens": 16693311.0,
      "step": 9040
    },
    {
      "entropy": 5.70733790397644,
      "epoch": 0.7599243856332704,
      "grad_norm": 1.390625,
      "learning_rate": 0.000494861632106755,
      "loss": 5.5713,
      "mean_token_accuracy": 0.15583091229200363,
      "num_tokens": 16702121.0,
      "step": 9045
    },
    {
      "entropy": 5.73174524307251,
      "epoch": 0.7603444654484351,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004948552667577033,
      "loss": 5.5963,
      "mean_token_accuracy": 0.15507323294878006,
      "num_tokens": 16711883.0,
      "step": 9050
    },
    {
      "entropy": 5.765106439590454,
      "epoch": 0.7607645452636,
      "grad_norm": 3.609375,
      "learning_rate": 0.0004948488975140286,
      "loss": 5.6685,
      "mean_token_accuracy": 0.14950548410415648,
      "num_tokens": 16721449.0,
      "step": 9055
    },
    {
      "entropy": 5.715044593811035,
      "epoch": 0.7611846250787649,
      "grad_norm": 1.484375,
      "learning_rate": 0.000494842524375844,
      "loss": 5.5945,
      "mean_token_accuracy": 0.153316530585289,
      "num_tokens": 16730068.0,
      "step": 9060
    },
    {
      "entropy": 5.686846685409546,
      "epoch": 0.7616047048939298,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004948361473432623,
      "loss": 5.6167,
      "mean_token_accuracy": 0.1580433189868927,
      "num_tokens": 16739970.0,
      "step": 9065
    },
    {
      "entropy": 5.780406522750854,
      "epoch": 0.7620247847090947,
      "grad_norm": 1.625,
      "learning_rate": 0.0004948297664163964,
      "loss": 5.674,
      "mean_token_accuracy": 0.14935452342033387,
      "num_tokens": 16749461.0,
      "step": 9070
    },
    {
      "entropy": 5.811807107925415,
      "epoch": 0.7624448645242596,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004948233815953593,
      "loss": 5.7538,
      "mean_token_accuracy": 0.15181455612182618,
      "num_tokens": 16758747.0,
      "step": 9075
    },
    {
      "entropy": 5.731409168243408,
      "epoch": 0.7628649443394245,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004948169928802643,
      "loss": 5.4593,
      "mean_token_accuracy": 0.16504789292812347,
      "num_tokens": 16767212.0,
      "step": 9080
    },
    {
      "entropy": 5.768598890304565,
      "epoch": 0.7632850241545893,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004948106002712245,
      "loss": 5.6169,
      "mean_token_accuracy": 0.15554029792547225,
      "num_tokens": 16776514.0,
      "step": 9085
    },
    {
      "entropy": 5.767139911651611,
      "epoch": 0.7637051039697542,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004948042037683529,
      "loss": 5.6008,
      "mean_token_accuracy": 0.15485090836882592,
      "num_tokens": 16786310.0,
      "step": 9090
    },
    {
      "entropy": 5.766057252883911,
      "epoch": 0.7641251837849191,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004947978033717632,
      "loss": 5.6302,
      "mean_token_accuracy": 0.1551863893866539,
      "num_tokens": 16795551.0,
      "step": 9095
    },
    {
      "entropy": 5.779314613342285,
      "epoch": 0.764545263600084,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004947913990815684,
      "loss": 5.6047,
      "mean_token_accuracy": 0.156571102142334,
      "num_tokens": 16805099.0,
      "step": 9100
    },
    {
      "entropy": 5.7028850555419925,
      "epoch": 0.7649653434152489,
      "grad_norm": 2.625,
      "learning_rate": 0.0004947849908978824,
      "loss": 5.6365,
      "mean_token_accuracy": 0.15258645862340928,
      "num_tokens": 16813963.0,
      "step": 9105
    },
    {
      "entropy": 5.773057985305786,
      "epoch": 0.7653854232304138,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004947785788208182,
      "loss": 5.6729,
      "mean_token_accuracy": 0.15363204330205918,
      "num_tokens": 16822814.0,
      "step": 9110
    },
    {
      "entropy": 5.817337226867676,
      "epoch": 0.7658055030455787,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004947721628504898,
      "loss": 5.7105,
      "mean_token_accuracy": 0.14875250086188316,
      "num_tokens": 16831906.0,
      "step": 9115
    },
    {
      "entropy": 5.712357425689698,
      "epoch": 0.7662255828607435,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004947657429870108,
      "loss": 5.5169,
      "mean_token_accuracy": 0.1595179632306099,
      "num_tokens": 16840050.0,
      "step": 9120
    },
    {
      "entropy": 5.650595998764038,
      "epoch": 0.7666456626759084,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004947593192304946,
      "loss": 5.5462,
      "mean_token_accuracy": 0.15449218600988388,
      "num_tokens": 16848404.0,
      "step": 9125
    },
    {
      "entropy": 5.7080615043640135,
      "epoch": 0.7670657424910733,
      "grad_norm": 1.625,
      "learning_rate": 0.0004947528915810554,
      "loss": 5.5407,
      "mean_token_accuracy": 0.15641333758831025,
      "num_tokens": 16856568.0,
      "step": 9130
    },
    {
      "entropy": 5.756184530258179,
      "epoch": 0.7674858223062382,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004947464600388066,
      "loss": 5.5755,
      "mean_token_accuracy": 0.16077920794487,
      "num_tokens": 16864936.0,
      "step": 9135
    },
    {
      "entropy": 5.886310577392578,
      "epoch": 0.7679059021214031,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004947400246038627,
      "loss": 5.7142,
      "mean_token_accuracy": 0.1526971772313118,
      "num_tokens": 16874504.0,
      "step": 9140
    },
    {
      "entropy": 5.670875883102417,
      "epoch": 0.768325981936568,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004947335852763374,
      "loss": 5.4664,
      "mean_token_accuracy": 0.15701207965612413,
      "num_tokens": 16883365.0,
      "step": 9145
    },
    {
      "entropy": 5.751158428192139,
      "epoch": 0.7687460617517329,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004947271420563447,
      "loss": 5.7208,
      "mean_token_accuracy": 0.14481201022863388,
      "num_tokens": 16892701.0,
      "step": 9150
    },
    {
      "entropy": 5.7310854434967045,
      "epoch": 0.7691661415668977,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004947206949439989,
      "loss": 5.5348,
      "mean_token_accuracy": 0.1528614804148674,
      "num_tokens": 16901864.0,
      "step": 9155
    },
    {
      "entropy": 5.725323820114136,
      "epoch": 0.7695862213820626,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000494714243939414,
      "loss": 5.5737,
      "mean_token_accuracy": 0.16332917958498,
      "num_tokens": 16910908.0,
      "step": 9160
    },
    {
      "entropy": 5.7280755043029785,
      "epoch": 0.7700063011972275,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004947077890427045,
      "loss": 5.6237,
      "mean_token_accuracy": 0.1593033030629158,
      "num_tokens": 16920299.0,
      "step": 9165
    },
    {
      "entropy": 5.845327138900757,
      "epoch": 0.7704263810123924,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004947013302539846,
      "loss": 5.7456,
      "mean_token_accuracy": 0.1439913384616375,
      "num_tokens": 16930027.0,
      "step": 9170
    },
    {
      "entropy": 5.800181293487549,
      "epoch": 0.7708464608275573,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004946948675733688,
      "loss": 5.6265,
      "mean_token_accuracy": 0.15197980552911758,
      "num_tokens": 16939387.0,
      "step": 9175
    },
    {
      "entropy": 5.726575946807861,
      "epoch": 0.7712665406427222,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004946884010009714,
      "loss": 5.5999,
      "mean_token_accuracy": 0.15767696648836135,
      "num_tokens": 16950024.0,
      "step": 9180
    },
    {
      "entropy": 5.690348434448242,
      "epoch": 0.771686620457887,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004946819305369073,
      "loss": 5.4956,
      "mean_token_accuracy": 0.16214724481105805,
      "num_tokens": 16958219.0,
      "step": 9185
    },
    {
      "entropy": 5.695686388015747,
      "epoch": 0.7721067002730518,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004946754561812909,
      "loss": 5.4751,
      "mean_token_accuracy": 0.16501737236976624,
      "num_tokens": 16966829.0,
      "step": 9190
    },
    {
      "entropy": 5.6394744396209715,
      "epoch": 0.7725267800882167,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004946689779342367,
      "loss": 5.5829,
      "mean_token_accuracy": 0.157248455286026,
      "num_tokens": 16975585.0,
      "step": 9195
    },
    {
      "entropy": 5.688712549209595,
      "epoch": 0.7729468599033816,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004946624957958599,
      "loss": 5.5508,
      "mean_token_accuracy": 0.15958887487649917,
      "num_tokens": 16984848.0,
      "step": 9200
    },
    {
      "entropy": 5.699162340164184,
      "epoch": 0.7733669397185465,
      "grad_norm": 1.578125,
      "learning_rate": 0.000494656009766275,
      "loss": 5.5601,
      "mean_token_accuracy": 0.16677376627922058,
      "num_tokens": 16993179.0,
      "step": 9205
    },
    {
      "entropy": 5.676146554946899,
      "epoch": 0.7737870195337114,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000494649519845597,
      "loss": 5.58,
      "mean_token_accuracy": 0.15755597352981568,
      "num_tokens": 17002563.0,
      "step": 9210
    },
    {
      "entropy": 5.762381267547608,
      "epoch": 0.7742070993488763,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004946430260339409,
      "loss": 5.6151,
      "mean_token_accuracy": 0.15177177786827087,
      "num_tokens": 17011805.0,
      "step": 9215
    },
    {
      "entropy": 5.730970144271851,
      "epoch": 0.7746271791640411,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004946365283314216,
      "loss": 5.5539,
      "mean_token_accuracy": 0.15288405269384384,
      "num_tokens": 17020398.0,
      "step": 9220
    },
    {
      "entropy": 5.67476453781128,
      "epoch": 0.775047258979206,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004946300267381545,
      "loss": 5.5585,
      "mean_token_accuracy": 0.15727789849042892,
      "num_tokens": 17030805.0,
      "step": 9225
    },
    {
      "entropy": 5.746345138549804,
      "epoch": 0.7754673387943709,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004946235212542544,
      "loss": 5.5945,
      "mean_token_accuracy": 0.15904315412044526,
      "num_tokens": 17040164.0,
      "step": 9230
    },
    {
      "entropy": 5.746207189559937,
      "epoch": 0.7758874186095358,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004946170118798367,
      "loss": 5.6473,
      "mean_token_accuracy": 0.15356625765562057,
      "num_tokens": 17049519.0,
      "step": 9235
    },
    {
      "entropy": 5.743830728530884,
      "epoch": 0.7763074984247007,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004946104986150167,
      "loss": 5.5775,
      "mean_token_accuracy": 0.15959766507148743,
      "num_tokens": 17058042.0,
      "step": 9240
    },
    {
      "entropy": 5.7007852554321286,
      "epoch": 0.7767275782398656,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004946039814599099,
      "loss": 5.6015,
      "mean_token_accuracy": 0.1615740664303303,
      "num_tokens": 17067107.0,
      "step": 9245
    },
    {
      "entropy": 5.7166907787323,
      "epoch": 0.7771476580550305,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004945974604146316,
      "loss": 5.6825,
      "mean_token_accuracy": 0.16084015890955924,
      "num_tokens": 17076975.0,
      "step": 9250
    },
    {
      "entropy": 5.709249782562256,
      "epoch": 0.7775677378701953,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004945909354792974,
      "loss": 5.5367,
      "mean_token_accuracy": 0.1583693414926529,
      "num_tokens": 17086405.0,
      "step": 9255
    },
    {
      "entropy": 5.691678667068482,
      "epoch": 0.7779878176853602,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004945844066540229,
      "loss": 5.6114,
      "mean_token_accuracy": 0.15048627853393554,
      "num_tokens": 17095333.0,
      "step": 9260
    },
    {
      "entropy": 5.744761037826538,
      "epoch": 0.7784078975005251,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004945778739389236,
      "loss": 5.6402,
      "mean_token_accuracy": 0.15518757700920105,
      "num_tokens": 17103631.0,
      "step": 9265
    },
    {
      "entropy": 5.740716123580933,
      "epoch": 0.77882797731569,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004945713373341152,
      "loss": 5.5545,
      "mean_token_accuracy": 0.15527277439832687,
      "num_tokens": 17112612.0,
      "step": 9270
    },
    {
      "entropy": 5.7405472755432125,
      "epoch": 0.7792480571308549,
      "grad_norm": 1.625,
      "learning_rate": 0.0004945647968397139,
      "loss": 5.5985,
      "mean_token_accuracy": 0.15742275416851043,
      "num_tokens": 17121592.0,
      "step": 9275
    },
    {
      "entropy": 5.7078413486480715,
      "epoch": 0.7796681369460198,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004945582524558352,
      "loss": 5.6183,
      "mean_token_accuracy": 0.15746289044618605,
      "num_tokens": 17131003.0,
      "step": 9280
    },
    {
      "entropy": 5.772906827926636,
      "epoch": 0.7800882167611847,
      "grad_norm": 1.640625,
      "learning_rate": 0.000494551704182595,
      "loss": 5.621,
      "mean_token_accuracy": 0.15148474127054215,
      "num_tokens": 17140013.0,
      "step": 9285
    },
    {
      "entropy": 5.876817035675049,
      "epoch": 0.7805082965763495,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004945451520201095,
      "loss": 5.7745,
      "mean_token_accuracy": 0.144088314473629,
      "num_tokens": 17150406.0,
      "step": 9290
    },
    {
      "entropy": 5.749739122390747,
      "epoch": 0.7809283763915144,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004945385959684947,
      "loss": 5.6295,
      "mean_token_accuracy": 0.15528156161308287,
      "num_tokens": 17159757.0,
      "step": 9295
    },
    {
      "entropy": 5.772205591201782,
      "epoch": 0.7813484562066793,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004945320360278667,
      "loss": 5.6364,
      "mean_token_accuracy": 0.16099981665611268,
      "num_tokens": 17169317.0,
      "step": 9300
    },
    {
      "entropy": 5.811273384094238,
      "epoch": 0.7817685360218442,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004945254721983416,
      "loss": 5.6388,
      "mean_token_accuracy": 0.16424707993865012,
      "num_tokens": 17178410.0,
      "step": 9305
    },
    {
      "entropy": 5.776491260528564,
      "epoch": 0.782188615837009,
      "grad_norm": 2.1875,
      "learning_rate": 0.000494518904480036,
      "loss": 5.5526,
      "mean_token_accuracy": 0.15707804411649703,
      "num_tokens": 17186922.0,
      "step": 9310
    },
    {
      "entropy": 5.737505054473877,
      "epoch": 0.782608695652174,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004945123328730659,
      "loss": 5.651,
      "mean_token_accuracy": 0.15191663503646852,
      "num_tokens": 17197125.0,
      "step": 9315
    },
    {
      "entropy": 5.706150245666504,
      "epoch": 0.7830287754673388,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000494505757377548,
      "loss": 5.5702,
      "mean_token_accuracy": 0.15521910637617112,
      "num_tokens": 17206169.0,
      "step": 9320
    },
    {
      "entropy": 5.650849008560181,
      "epoch": 0.7834488552825036,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004944991779935985,
      "loss": 5.5078,
      "mean_token_accuracy": 0.15866184681653978,
      "num_tokens": 17214607.0,
      "step": 9325
    },
    {
      "entropy": 5.662286853790283,
      "epoch": 0.7838689350976685,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000494492594721334,
      "loss": 5.4998,
      "mean_token_accuracy": 0.1558862790465355,
      "num_tokens": 17223616.0,
      "step": 9330
    },
    {
      "entropy": 5.785139083862305,
      "epoch": 0.7842890149128334,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004944860075608715,
      "loss": 5.591,
      "mean_token_accuracy": 0.15421077311038972,
      "num_tokens": 17232729.0,
      "step": 9335
    },
    {
      "entropy": 5.724941205978394,
      "epoch": 0.7847090947279983,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004944794165123272,
      "loss": 5.6489,
      "mean_token_accuracy": 0.15867509245872496,
      "num_tokens": 17242128.0,
      "step": 9340
    },
    {
      "entropy": 5.691767406463623,
      "epoch": 0.7851291745431632,
      "grad_norm": 1.65625,
      "learning_rate": 0.000494472821575818,
      "loss": 5.538,
      "mean_token_accuracy": 0.15931878685951234,
      "num_tokens": 17250806.0,
      "step": 9345
    },
    {
      "entropy": 5.807684230804443,
      "epoch": 0.7855492543583281,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004944662227514609,
      "loss": 5.7718,
      "mean_token_accuracy": 0.14448002949357033,
      "num_tokens": 17260888.0,
      "step": 9350
    },
    {
      "entropy": 5.739693880081177,
      "epoch": 0.785969334173493,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004944596200393726,
      "loss": 5.5242,
      "mean_token_accuracy": 0.16088390797376634,
      "num_tokens": 17270387.0,
      "step": 9355
    },
    {
      "entropy": 5.755869579315186,
      "epoch": 0.7863894139886578,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004944530134396702,
      "loss": 5.5782,
      "mean_token_accuracy": 0.15905830040574073,
      "num_tokens": 17279866.0,
      "step": 9360
    },
    {
      "entropy": 5.681657218933106,
      "epoch": 0.7868094938038227,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004944464029524707,
      "loss": 5.5575,
      "mean_token_accuracy": 0.15929284542798997,
      "num_tokens": 17289233.0,
      "step": 9365
    },
    {
      "entropy": 5.764404249191284,
      "epoch": 0.7872295736189876,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000494439788577891,
      "loss": 5.6412,
      "mean_token_accuracy": 0.1553253263235092,
      "num_tokens": 17298705.0,
      "step": 9370
    },
    {
      "entropy": 5.760324287414551,
      "epoch": 0.7876496534341525,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004944331703160486,
      "loss": 5.6003,
      "mean_token_accuracy": 0.15822061598300935,
      "num_tokens": 17307793.0,
      "step": 9375
    },
    {
      "entropy": 5.7375633239746096,
      "epoch": 0.7880697332493174,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004944265481670605,
      "loss": 5.6821,
      "mean_token_accuracy": 0.14730169177055358,
      "num_tokens": 17318248.0,
      "step": 9380
    },
    {
      "entropy": 5.723534107208252,
      "epoch": 0.7884898130644823,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004944199221310441,
      "loss": 5.5923,
      "mean_token_accuracy": 0.15378608107566832,
      "num_tokens": 17327281.0,
      "step": 9385
    },
    {
      "entropy": 5.751460123062134,
      "epoch": 0.7889098928796471,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004944132922081168,
      "loss": 5.6041,
      "mean_token_accuracy": 0.16031598746776582,
      "num_tokens": 17336805.0,
      "step": 9390
    },
    {
      "entropy": 5.720217990875244,
      "epoch": 0.789329972694812,
      "grad_norm": 1.5,
      "learning_rate": 0.0004944066583983961,
      "loss": 5.5563,
      "mean_token_accuracy": 0.15304937809705735,
      "num_tokens": 17346024.0,
      "step": 9395
    },
    {
      "entropy": 5.72339243888855,
      "epoch": 0.7897500525099769,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004944000207019992,
      "loss": 5.638,
      "mean_token_accuracy": 0.15410090163350104,
      "num_tokens": 17355100.0,
      "step": 9400
    },
    {
      "entropy": 5.789851570129395,
      "epoch": 0.7901701323251418,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004943933791190441,
      "loss": 5.6855,
      "mean_token_accuracy": 0.14976395517587662,
      "num_tokens": 17364769.0,
      "step": 9405
    },
    {
      "entropy": 5.733008146286011,
      "epoch": 0.7905902121403067,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004943867336496482,
      "loss": 5.5309,
      "mean_token_accuracy": 0.16207973062992095,
      "num_tokens": 17374082.0,
      "step": 9410
    },
    {
      "entropy": 5.677785396575928,
      "epoch": 0.7910102919554716,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004943800842939293,
      "loss": 5.5769,
      "mean_token_accuracy": 0.15575399696826936,
      "num_tokens": 17383570.0,
      "step": 9415
    },
    {
      "entropy": 5.715662574768066,
      "epoch": 0.7914303717706365,
      "grad_norm": 1.453125,
      "learning_rate": 0.000494373431052005,
      "loss": 5.5744,
      "mean_token_accuracy": 0.15315101221203803,
      "num_tokens": 17392105.0,
      "step": 9420
    },
    {
      "entropy": 5.687941646575927,
      "epoch": 0.7918504515858013,
      "grad_norm": 1.5,
      "learning_rate": 0.0004943667739239935,
      "loss": 5.5394,
      "mean_token_accuracy": 0.16118402928113937,
      "num_tokens": 17401363.0,
      "step": 9425
    },
    {
      "entropy": 5.805943298339844,
      "epoch": 0.7922705314009661,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004943601129100125,
      "loss": 5.5689,
      "mean_token_accuracy": 0.1596407786011696,
      "num_tokens": 17411333.0,
      "step": 9430
    },
    {
      "entropy": 5.742655897140503,
      "epoch": 0.792690611216131,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004943534480101801,
      "loss": 5.6116,
      "mean_token_accuracy": 0.16205159723758697,
      "num_tokens": 17421162.0,
      "step": 9435
    },
    {
      "entropy": 5.675297021865845,
      "epoch": 0.793110691031296,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004943467792246142,
      "loss": 5.5643,
      "mean_token_accuracy": 0.16239771544933318,
      "num_tokens": 17430119.0,
      "step": 9440
    },
    {
      "entropy": 5.73559455871582,
      "epoch": 0.7935307708464608,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004943401065534332,
      "loss": 5.5718,
      "mean_token_accuracy": 0.1576070472598076,
      "num_tokens": 17439617.0,
      "step": 9445
    },
    {
      "entropy": 5.6667726039886475,
      "epoch": 0.7939508506616257,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004943334299967551,
      "loss": 5.6816,
      "mean_token_accuracy": 0.15144198015332222,
      "num_tokens": 17448720.0,
      "step": 9450
    },
    {
      "entropy": 5.692353296279907,
      "epoch": 0.7943709304767906,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004943267495546982,
      "loss": 5.5709,
      "mean_token_accuracy": 0.16103459149599075,
      "num_tokens": 17457458.0,
      "step": 9455
    },
    {
      "entropy": 5.826778411865234,
      "epoch": 0.7947910102919554,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004943200652273809,
      "loss": 5.5988,
      "mean_token_accuracy": 0.1572289600968361,
      "num_tokens": 17467095.0,
      "step": 9460
    },
    {
      "entropy": 5.734735774993896,
      "epoch": 0.7952110901071203,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004943133770149216,
      "loss": 5.6124,
      "mean_token_accuracy": 0.14904361963272095,
      "num_tokens": 17476247.0,
      "step": 9465
    },
    {
      "entropy": 5.758768129348755,
      "epoch": 0.7956311699222852,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004943066849174386,
      "loss": 5.6185,
      "mean_token_accuracy": 0.16090165376663207,
      "num_tokens": 17486352.0,
      "step": 9470
    },
    {
      "entropy": 5.764822721481323,
      "epoch": 0.7960512497374501,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004942999889350508,
      "loss": 5.6045,
      "mean_token_accuracy": 0.1577667087316513,
      "num_tokens": 17495633.0,
      "step": 9475
    },
    {
      "entropy": 5.76590781211853,
      "epoch": 0.796471329552615,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004942932890678765,
      "loss": 5.6398,
      "mean_token_accuracy": 0.1567259430885315,
      "num_tokens": 17504325.0,
      "step": 9480
    },
    {
      "entropy": 5.784614133834839,
      "epoch": 0.7968914093677799,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004942865853160346,
      "loss": 5.6395,
      "mean_token_accuracy": 0.15519623905420304,
      "num_tokens": 17513265.0,
      "step": 9485
    },
    {
      "entropy": 5.779874706268311,
      "epoch": 0.7973114891829448,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004942798776796436,
      "loss": 5.6635,
      "mean_token_accuracy": 0.1501373194158077,
      "num_tokens": 17522939.0,
      "step": 9490
    },
    {
      "entropy": 5.813601112365722,
      "epoch": 0.7977315689981096,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004942731661588226,
      "loss": 5.6776,
      "mean_token_accuracy": 0.14708788096904754,
      "num_tokens": 17532250.0,
      "step": 9495
    },
    {
      "entropy": 5.8093610286712645,
      "epoch": 0.7981516488132745,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004942664507536904,
      "loss": 5.6768,
      "mean_token_accuracy": 0.15758057236671447,
      "num_tokens": 17541368.0,
      "step": 9500
    },
    {
      "entropy": 5.709500455856324,
      "epoch": 0.7985717286284394,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004942597314643659,
      "loss": 5.6119,
      "mean_token_accuracy": 0.15914746522903442,
      "num_tokens": 17550871.0,
      "step": 9505
    },
    {
      "entropy": 5.7264715194702145,
      "epoch": 0.7989918084436043,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004942530082909681,
      "loss": 5.5592,
      "mean_token_accuracy": 0.16579595506191253,
      "num_tokens": 17559683.0,
      "step": 9510
    },
    {
      "entropy": 5.739404821395874,
      "epoch": 0.7994118882587692,
      "grad_norm": 1.375,
      "learning_rate": 0.0004942462812336163,
      "loss": 5.5622,
      "mean_token_accuracy": 0.15668560117483138,
      "num_tokens": 17568877.0,
      "step": 9515
    },
    {
      "entropy": 5.809852504730225,
      "epoch": 0.7998319680739341,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004942395502924293,
      "loss": 5.6985,
      "mean_token_accuracy": 0.14932294711470603,
      "num_tokens": 17578202.0,
      "step": 9520
    },
    {
      "entropy": 5.7818114280700685,
      "epoch": 0.800252047889099,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004942328154675268,
      "loss": 5.5476,
      "mean_token_accuracy": 0.1580620989203453,
      "num_tokens": 17587342.0,
      "step": 9525
    },
    {
      "entropy": 5.704915952682495,
      "epoch": 0.8006721277042638,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004942260767590277,
      "loss": 5.4071,
      "mean_token_accuracy": 0.16135389506816863,
      "num_tokens": 17595671.0,
      "step": 9530
    },
    {
      "entropy": 5.718443107604981,
      "epoch": 0.8010922075194287,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004942193341670516,
      "loss": 5.7412,
      "mean_token_accuracy": 0.14880940914154053,
      "num_tokens": 17605649.0,
      "step": 9535
    },
    {
      "entropy": 5.7878273010253904,
      "epoch": 0.8015122873345936,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004942125876917178,
      "loss": 5.6307,
      "mean_token_accuracy": 0.15066928714513778,
      "num_tokens": 17615286.0,
      "step": 9540
    },
    {
      "entropy": 5.6786088943481445,
      "epoch": 0.8019323671497585,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000494205837333146,
      "loss": 5.5965,
      "mean_token_accuracy": 0.15368861109018325,
      "num_tokens": 17624583.0,
      "step": 9545
    },
    {
      "entropy": 5.73230299949646,
      "epoch": 0.8023524469649234,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004941990830914557,
      "loss": 5.5956,
      "mean_token_accuracy": 0.15637702941894532,
      "num_tokens": 17633894.0,
      "step": 9550
    },
    {
      "entropy": 5.777722024917603,
      "epoch": 0.8027725267800883,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004941923249667663,
      "loss": 5.6906,
      "mean_token_accuracy": 0.1498961329460144,
      "num_tokens": 17643172.0,
      "step": 9555
    },
    {
      "entropy": 5.776974630355835,
      "epoch": 0.803192606595253,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004941855629591979,
      "loss": 5.564,
      "mean_token_accuracy": 0.15488137155771256,
      "num_tokens": 17651901.0,
      "step": 9560
    },
    {
      "entropy": 5.695595645904541,
      "epoch": 0.8036126864104179,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004941787970688701,
      "loss": 5.5599,
      "mean_token_accuracy": 0.1548061341047287,
      "num_tokens": 17660806.0,
      "step": 9565
    },
    {
      "entropy": 5.745298910140991,
      "epoch": 0.8040327662255828,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004941720272959027,
      "loss": 5.6197,
      "mean_token_accuracy": 0.16593484580516815,
      "num_tokens": 17669157.0,
      "step": 9570
    },
    {
      "entropy": 5.67219820022583,
      "epoch": 0.8044528460407477,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004941652536404157,
      "loss": 5.5339,
      "mean_token_accuracy": 0.15697258710861206,
      "num_tokens": 17678664.0,
      "step": 9575
    },
    {
      "entropy": 5.7390752792358395,
      "epoch": 0.8048729258559126,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004941584761025291,
      "loss": 5.5754,
      "mean_token_accuracy": 0.15572021156549454,
      "num_tokens": 17688252.0,
      "step": 9580
    },
    {
      "entropy": 5.716245555877686,
      "epoch": 0.8052930056710775,
      "grad_norm": 1.703125,
      "learning_rate": 0.000494151694682363,
      "loss": 5.6001,
      "mean_token_accuracy": 0.15716407224535942,
      "num_tokens": 17696473.0,
      "step": 9585
    },
    {
      "entropy": 5.7132292747497555,
      "epoch": 0.8057130854862424,
      "grad_norm": 1.625,
      "learning_rate": 0.0004941449093800374,
      "loss": 5.6374,
      "mean_token_accuracy": 0.15638218745589255,
      "num_tokens": 17706177.0,
      "step": 9590
    },
    {
      "entropy": 5.664974069595337,
      "epoch": 0.8061331653014072,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004941381201956726,
      "loss": 5.4683,
      "mean_token_accuracy": 0.16399573683738708,
      "num_tokens": 17715355.0,
      "step": 9595
    },
    {
      "entropy": 5.719002342224121,
      "epoch": 0.8065532451165721,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004941313271293889,
      "loss": 5.5711,
      "mean_token_accuracy": 0.16259212642908097,
      "num_tokens": 17724345.0,
      "step": 9600
    },
    {
      "entropy": 5.7297220706939695,
      "epoch": 0.806973324931737,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004941245301813065,
      "loss": 5.4994,
      "mean_token_accuracy": 0.16674552410840987,
      "num_tokens": 17732805.0,
      "step": 9605
    },
    {
      "entropy": 5.628268814086914,
      "epoch": 0.8073934047469019,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004941177293515459,
      "loss": 5.5412,
      "mean_token_accuracy": 0.16236957609653474,
      "num_tokens": 17741963.0,
      "step": 9610
    },
    {
      "entropy": 5.633747529983521,
      "epoch": 0.8078134845620668,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004941109246402275,
      "loss": 5.5727,
      "mean_token_accuracy": 0.1555377036333084,
      "num_tokens": 17751858.0,
      "step": 9615
    },
    {
      "entropy": 5.845067882537842,
      "epoch": 0.8082335643772317,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004941041160474721,
      "loss": 5.6876,
      "mean_token_accuracy": 0.15293847322463988,
      "num_tokens": 17761152.0,
      "step": 9620
    },
    {
      "entropy": 5.818953990936279,
      "epoch": 0.8086536441923966,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004940973035733999,
      "loss": 5.6166,
      "mean_token_accuracy": 0.15496550351381302,
      "num_tokens": 17770493.0,
      "step": 9625
    },
    {
      "entropy": 5.828579998016357,
      "epoch": 0.8090737240075614,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004940904872181318,
      "loss": 5.6185,
      "mean_token_accuracy": 0.15005985349416734,
      "num_tokens": 17779871.0,
      "step": 9630
    },
    {
      "entropy": 5.764964485168457,
      "epoch": 0.8094938038227263,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004940836669817887,
      "loss": 5.6368,
      "mean_token_accuracy": 0.14861269071698188,
      "num_tokens": 17788606.0,
      "step": 9635
    },
    {
      "entropy": 5.634247922897339,
      "epoch": 0.8099138836378912,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004940768428644911,
      "loss": 5.5661,
      "mean_token_accuracy": 0.1579072043299675,
      "num_tokens": 17797458.0,
      "step": 9640
    },
    {
      "entropy": 5.667587184906006,
      "epoch": 0.8103339634530561,
      "grad_norm": 1.375,
      "learning_rate": 0.0004940700148663601,
      "loss": 5.5228,
      "mean_token_accuracy": 0.15832778215408325,
      "num_tokens": 17806902.0,
      "step": 9645
    },
    {
      "entropy": 5.784042167663574,
      "epoch": 0.810754043268221,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004940631829875165,
      "loss": 5.6538,
      "mean_token_accuracy": 0.15309607237577438,
      "num_tokens": 17816374.0,
      "step": 9650
    },
    {
      "entropy": 5.747311544418335,
      "epoch": 0.8111741230833859,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004940563472280815,
      "loss": 5.6394,
      "mean_token_accuracy": 0.15840844437479973,
      "num_tokens": 17825267.0,
      "step": 9655
    },
    {
      "entropy": 5.6849054336547855,
      "epoch": 0.8115942028985508,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004940495075881761,
      "loss": 5.5501,
      "mean_token_accuracy": 0.15928401798009872,
      "num_tokens": 17834027.0,
      "step": 9660
    },
    {
      "entropy": 5.710701990127563,
      "epoch": 0.8120142827137156,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004940426640679214,
      "loss": 5.5574,
      "mean_token_accuracy": 0.15307731851935386,
      "num_tokens": 17843587.0,
      "step": 9665
    },
    {
      "entropy": 5.759457540512085,
      "epoch": 0.8124343625288805,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004940358166674388,
      "loss": 5.5651,
      "mean_token_accuracy": 0.1602780595421791,
      "num_tokens": 17852284.0,
      "step": 9670
    },
    {
      "entropy": 5.7291789054870605,
      "epoch": 0.8128544423440454,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004940289653868494,
      "loss": 5.5959,
      "mean_token_accuracy": 0.1548108696937561,
      "num_tokens": 17860896.0,
      "step": 9675
    },
    {
      "entropy": 5.592970943450927,
      "epoch": 0.8132745221592103,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004940221102262747,
      "loss": 5.5689,
      "mean_token_accuracy": 0.15153344422578813,
      "num_tokens": 17870796.0,
      "step": 9680
    },
    {
      "entropy": 5.760674333572387,
      "epoch": 0.8136946019743752,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004940152511858361,
      "loss": 5.6389,
      "mean_token_accuracy": 0.15776861757040023,
      "num_tokens": 17880016.0,
      "step": 9685
    },
    {
      "entropy": 5.840538072586059,
      "epoch": 0.81411468178954,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004940083882656551,
      "loss": 5.6842,
      "mean_token_accuracy": 0.1527305245399475,
      "num_tokens": 17889348.0,
      "step": 9690
    },
    {
      "entropy": 5.7093421459198,
      "epoch": 0.814534761604705,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004940015214658532,
      "loss": 5.5527,
      "mean_token_accuracy": 0.16001000851392747,
      "num_tokens": 17898392.0,
      "step": 9695
    },
    {
      "entropy": 5.737645149230957,
      "epoch": 0.8149548414198697,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004939946507865522,
      "loss": 5.644,
      "mean_token_accuracy": 0.16057325303554534,
      "num_tokens": 17907141.0,
      "step": 9700
    },
    {
      "entropy": 5.655606842041015,
      "epoch": 0.8153749212350346,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004939877762278737,
      "loss": 5.5158,
      "mean_token_accuracy": 0.16194361746311187,
      "num_tokens": 17915792.0,
      "step": 9705
    },
    {
      "entropy": 5.797022104263306,
      "epoch": 0.8157950010501995,
      "grad_norm": 1.5,
      "learning_rate": 0.0004939808977899396,
      "loss": 5.6857,
      "mean_token_accuracy": 0.14906044080853462,
      "num_tokens": 17925603.0,
      "step": 9710
    },
    {
      "entropy": 5.802898216247558,
      "epoch": 0.8162150808653644,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004939740154728716,
      "loss": 5.6278,
      "mean_token_accuracy": 0.15891481786966324,
      "num_tokens": 17934436.0,
      "step": 9715
    },
    {
      "entropy": 5.697312355041504,
      "epoch": 0.8166351606805293,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004939671292767915,
      "loss": 5.5631,
      "mean_token_accuracy": 0.16518739312887193,
      "num_tokens": 17942969.0,
      "step": 9720
    },
    {
      "entropy": 5.805132532119751,
      "epoch": 0.8170552404956942,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004939602392018216,
      "loss": 5.6504,
      "mean_token_accuracy": 0.15329954028129578,
      "num_tokens": 17952053.0,
      "step": 9725
    },
    {
      "entropy": 5.761464977264405,
      "epoch": 0.817475320310859,
      "grad_norm": 1.375,
      "learning_rate": 0.0004939533452480839,
      "loss": 5.6279,
      "mean_token_accuracy": 0.15541652143001555,
      "num_tokens": 17960707.0,
      "step": 9730
    },
    {
      "entropy": 5.756668901443481,
      "epoch": 0.8178954001260239,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004939464474157003,
      "loss": 5.7156,
      "mean_token_accuracy": 0.1402080774307251,
      "num_tokens": 17971035.0,
      "step": 9735
    },
    {
      "entropy": 5.764472818374633,
      "epoch": 0.8183154799411888,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004939395457047932,
      "loss": 5.5814,
      "mean_token_accuracy": 0.15014344453811646,
      "num_tokens": 17980656.0,
      "step": 9740
    },
    {
      "entropy": 5.8357971668243405,
      "epoch": 0.8187355597563537,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004939326401154847,
      "loss": 5.6189,
      "mean_token_accuracy": 0.15061766505241395,
      "num_tokens": 17990977.0,
      "step": 9745
    },
    {
      "entropy": 5.659277296066284,
      "epoch": 0.8191556395715186,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004939257306478973,
      "loss": 5.6026,
      "mean_token_accuracy": 0.15962347537279128,
      "num_tokens": 18000186.0,
      "step": 9750
    },
    {
      "entropy": 5.66544737815857,
      "epoch": 0.8195757193866835,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004939188173021532,
      "loss": 5.584,
      "mean_token_accuracy": 0.15631651431322097,
      "num_tokens": 18010269.0,
      "step": 9755
    },
    {
      "entropy": 5.734535169601441,
      "epoch": 0.8199957992018484,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004939119000783751,
      "loss": 5.5143,
      "mean_token_accuracy": 0.1653580456972122,
      "num_tokens": 18018461.0,
      "step": 9760
    },
    {
      "entropy": 5.650555562973023,
      "epoch": 0.8204158790170132,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004939049789766855,
      "loss": 5.5355,
      "mean_token_accuracy": 0.15736581087112428,
      "num_tokens": 18027173.0,
      "step": 9765
    },
    {
      "entropy": 5.6493651390075685,
      "epoch": 0.8208359588321781,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004938980539972068,
      "loss": 5.6419,
      "mean_token_accuracy": 0.1530544400215149,
      "num_tokens": 18036791.0,
      "step": 9770
    },
    {
      "entropy": 5.711198854446411,
      "epoch": 0.821256038647343,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004938911251400617,
      "loss": 5.5821,
      "mean_token_accuracy": 0.16347371712327002,
      "num_tokens": 18046908.0,
      "step": 9775
    },
    {
      "entropy": 5.675604200363159,
      "epoch": 0.8216761184625079,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004938841924053731,
      "loss": 5.4868,
      "mean_token_accuracy": 0.16802890747785568,
      "num_tokens": 18055825.0,
      "step": 9780
    },
    {
      "entropy": 5.74740252494812,
      "epoch": 0.8220961982776728,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004938772557932637,
      "loss": 5.6547,
      "mean_token_accuracy": 0.14992306679487227,
      "num_tokens": 18065334.0,
      "step": 9785
    },
    {
      "entropy": 5.804665422439575,
      "epoch": 0.8225162780928377,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004938703153038565,
      "loss": 5.5424,
      "mean_token_accuracy": 0.16053025126457215,
      "num_tokens": 18073999.0,
      "step": 9790
    },
    {
      "entropy": 5.638562202453613,
      "epoch": 0.8229363579080026,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004938633709372744,
      "loss": 5.5667,
      "mean_token_accuracy": 0.1526872143149376,
      "num_tokens": 18083665.0,
      "step": 9795
    },
    {
      "entropy": 5.719447278976441,
      "epoch": 0.8233564377231674,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004938564226936403,
      "loss": 5.5635,
      "mean_token_accuracy": 0.1585657551884651,
      "num_tokens": 18092501.0,
      "step": 9800
    },
    {
      "entropy": 5.707524585723877,
      "epoch": 0.8237765175383323,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004938494705730773,
      "loss": 5.5628,
      "mean_token_accuracy": 0.15374434292316436,
      "num_tokens": 18101320.0,
      "step": 9805
    },
    {
      "entropy": 5.691710615158081,
      "epoch": 0.8241965973534972,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004938425145757087,
      "loss": 5.5892,
      "mean_token_accuracy": 0.15482329502701758,
      "num_tokens": 18110190.0,
      "step": 9810
    },
    {
      "entropy": 5.7161225318908695,
      "epoch": 0.824616677168662,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004938355547016577,
      "loss": 5.578,
      "mean_token_accuracy": 0.15880000442266465,
      "num_tokens": 18119301.0,
      "step": 9815
    },
    {
      "entropy": 5.83074254989624,
      "epoch": 0.825036756983827,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004938285909510474,
      "loss": 5.6304,
      "mean_token_accuracy": 0.15105096995830536,
      "num_tokens": 18128959.0,
      "step": 9820
    },
    {
      "entropy": 5.730278205871582,
      "epoch": 0.8254568367989918,
      "grad_norm": 1.625,
      "learning_rate": 0.0004938216233240014,
      "loss": 5.5971,
      "mean_token_accuracy": 0.15822493731975557,
      "num_tokens": 18138156.0,
      "step": 9825
    },
    {
      "entropy": 5.770270156860351,
      "epoch": 0.8258769166141567,
      "grad_norm": 1.484375,
      "learning_rate": 0.000493814651820643,
      "loss": 5.6304,
      "mean_token_accuracy": 0.14718272238969804,
      "num_tokens": 18147244.0,
      "step": 9830
    },
    {
      "entropy": 5.809439325332642,
      "epoch": 0.8262969964293215,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004938076764410956,
      "loss": 5.65,
      "mean_token_accuracy": 0.15126846134662628,
      "num_tokens": 18156040.0,
      "step": 9835
    },
    {
      "entropy": 5.774372339248657,
      "epoch": 0.8267170762444864,
      "grad_norm": 1.796875,
      "learning_rate": 0.000493800697185483,
      "loss": 5.5587,
      "mean_token_accuracy": 0.15274341106414796,
      "num_tokens": 18165210.0,
      "step": 9840
    },
    {
      "entropy": 5.755823945999145,
      "epoch": 0.8271371560596513,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004937937140539288,
      "loss": 5.6189,
      "mean_token_accuracy": 0.15705254673957825,
      "num_tokens": 18174841.0,
      "step": 9845
    },
    {
      "entropy": 5.673149156570434,
      "epoch": 0.8275572358748162,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004937867270465564,
      "loss": 5.5197,
      "mean_token_accuracy": 0.15668955445289612,
      "num_tokens": 18184112.0,
      "step": 9850
    },
    {
      "entropy": 5.792433404922486,
      "epoch": 0.8279773156899811,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004937797361634899,
      "loss": 5.7046,
      "mean_token_accuracy": 0.15178792774677277,
      "num_tokens": 18193564.0,
      "step": 9855
    },
    {
      "entropy": 5.670192384719849,
      "epoch": 0.828397395505146,
      "grad_norm": 1.46875,
      "learning_rate": 0.000493772741404853,
      "loss": 5.4317,
      "mean_token_accuracy": 0.16718422323465348,
      "num_tokens": 18202836.0,
      "step": 9860
    },
    {
      "entropy": 5.7452069282531735,
      "epoch": 0.8288174753203108,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004937657427707698,
      "loss": 5.5625,
      "mean_token_accuracy": 0.1660207688808441,
      "num_tokens": 18212098.0,
      "step": 9865
    },
    {
      "entropy": 5.751721334457398,
      "epoch": 0.8292375551354757,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004937587402613639,
      "loss": 5.5846,
      "mean_token_accuracy": 0.1579491063952446,
      "num_tokens": 18221541.0,
      "step": 9870
    },
    {
      "entropy": 5.679269027709961,
      "epoch": 0.8296576349506406,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004937517338767597,
      "loss": 5.5911,
      "mean_token_accuracy": 0.15127961188554764,
      "num_tokens": 18231015.0,
      "step": 9875
    },
    {
      "entropy": 5.7905172348022464,
      "epoch": 0.8300777147658055,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004937447236170811,
      "loss": 5.5991,
      "mean_token_accuracy": 0.15796486884355546,
      "num_tokens": 18239729.0,
      "step": 9880
    },
    {
      "entropy": 5.79022216796875,
      "epoch": 0.8304977945809704,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004937377094824523,
      "loss": 5.6634,
      "mean_token_accuracy": 0.15445734858512877,
      "num_tokens": 18249773.0,
      "step": 9885
    },
    {
      "entropy": 5.790219640731811,
      "epoch": 0.8309178743961353,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004937306914729977,
      "loss": 5.6113,
      "mean_token_accuracy": 0.15426182597875596,
      "num_tokens": 18259179.0,
      "step": 9890
    },
    {
      "entropy": 5.654812574386597,
      "epoch": 0.8313379542113002,
      "grad_norm": 1.375,
      "learning_rate": 0.0004937236695888416,
      "loss": 5.4941,
      "mean_token_accuracy": 0.16756488233804703,
      "num_tokens": 18268164.0,
      "step": 9895
    },
    {
      "entropy": 5.7460182189941404,
      "epoch": 0.831758034026465,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004937166438301082,
      "loss": 5.6649,
      "mean_token_accuracy": 0.15427334159612655,
      "num_tokens": 18276259.0,
      "step": 9900
    },
    {
      "entropy": 5.753793668746948,
      "epoch": 0.8321781138416299,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004937096141969221,
      "loss": 5.6431,
      "mean_token_accuracy": 0.15369066298007966,
      "num_tokens": 18285729.0,
      "step": 9905
    },
    {
      "entropy": 5.808768320083618,
      "epoch": 0.8325981936567948,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004937025806894077,
      "loss": 5.7907,
      "mean_token_accuracy": 0.1430658794939518,
      "num_tokens": 18295873.0,
      "step": 9910
    },
    {
      "entropy": 5.81347017288208,
      "epoch": 0.8330182734719597,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004936955433076899,
      "loss": 5.6261,
      "mean_token_accuracy": 0.16088502556085588,
      "num_tokens": 18305135.0,
      "step": 9915
    },
    {
      "entropy": 5.805698680877685,
      "epoch": 0.8334383532871246,
      "grad_norm": 1.75,
      "learning_rate": 0.000493688502051893,
      "loss": 5.6659,
      "mean_token_accuracy": 0.15734423473477363,
      "num_tokens": 18314251.0,
      "step": 9920
    },
    {
      "entropy": 5.717606830596924,
      "epoch": 0.8338584331022895,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004936814569221421,
      "loss": 5.5011,
      "mean_token_accuracy": 0.1689703121781349,
      "num_tokens": 18322863.0,
      "step": 9925
    },
    {
      "entropy": 5.666426181793213,
      "epoch": 0.8342785129174544,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004936744079185616,
      "loss": 5.5171,
      "mean_token_accuracy": 0.1564257636666298,
      "num_tokens": 18332129.0,
      "step": 9930
    },
    {
      "entropy": 5.784151458740235,
      "epoch": 0.8346985927326191,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004936673550412767,
      "loss": 5.6047,
      "mean_token_accuracy": 0.1550414428114891,
      "num_tokens": 18341457.0,
      "step": 9935
    },
    {
      "entropy": 5.771088027954102,
      "epoch": 0.835118672547784,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000493660298290412,
      "loss": 5.5876,
      "mean_token_accuracy": 0.1493722081184387,
      "num_tokens": 18351397.0,
      "step": 9940
    },
    {
      "entropy": 5.720294952392578,
      "epoch": 0.8355387523629489,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004936532376660929,
      "loss": 5.5374,
      "mean_token_accuracy": 0.1618410602211952,
      "num_tokens": 18360005.0,
      "step": 9945
    },
    {
      "entropy": 5.776373720169067,
      "epoch": 0.8359588321781138,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004936461731684442,
      "loss": 5.6228,
      "mean_token_accuracy": 0.15494077503681183,
      "num_tokens": 18369707.0,
      "step": 9950
    },
    {
      "entropy": 5.7964976787567135,
      "epoch": 0.8363789119932787,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004936391047975912,
      "loss": 5.7554,
      "mean_token_accuracy": 0.15112185776233672,
      "num_tokens": 18379514.0,
      "step": 9955
    },
    {
      "entropy": 5.773888969421387,
      "epoch": 0.8367989918084436,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004936320325536589,
      "loss": 5.4665,
      "mean_token_accuracy": 0.16086758822202682,
      "num_tokens": 18388854.0,
      "step": 9960
    },
    {
      "entropy": 5.768847894668579,
      "epoch": 0.8372190716236085,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004936249564367729,
      "loss": 5.6474,
      "mean_token_accuracy": 0.15842091292142868,
      "num_tokens": 18397806.0,
      "step": 9965
    },
    {
      "entropy": 5.585431623458862,
      "epoch": 0.8376391514387733,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004936178764470583,
      "loss": 5.4896,
      "mean_token_accuracy": 0.1574113130569458,
      "num_tokens": 18406645.0,
      "step": 9970
    },
    {
      "entropy": 5.645328092575073,
      "epoch": 0.8380592312539382,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004936107925846405,
      "loss": 5.5151,
      "mean_token_accuracy": 0.1597437858581543,
      "num_tokens": 18415730.0,
      "step": 9975
    },
    {
      "entropy": 5.784190368652344,
      "epoch": 0.8384793110691031,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004936037048496452,
      "loss": 5.6177,
      "mean_token_accuracy": 0.1599542900919914,
      "num_tokens": 18424638.0,
      "step": 9980
    },
    {
      "entropy": 5.734972286224365,
      "epoch": 0.838899390884268,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004935966132421977,
      "loss": 5.6581,
      "mean_token_accuracy": 0.1531389981508255,
      "num_tokens": 18434090.0,
      "step": 9985
    },
    {
      "entropy": 5.6199877738952635,
      "epoch": 0.8393194706994329,
      "grad_norm": 1.625,
      "learning_rate": 0.0004935895177624239,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16159379929304124,
      "num_tokens": 18442965.0,
      "step": 9990
    },
    {
      "entropy": 5.781462097167969,
      "epoch": 0.8397395505145978,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004935824184104493,
      "loss": 5.5283,
      "mean_token_accuracy": 0.16171049624681472,
      "num_tokens": 18451553.0,
      "step": 9995
    },
    {
      "entropy": 5.757485055923462,
      "epoch": 0.8401596303297627,
      "grad_norm": 1.5,
      "learning_rate": 0.0004935753151863997,
      "loss": 5.5918,
      "mean_token_accuracy": 0.1540190264582634,
      "num_tokens": 18461325.0,
      "step": 10000
    },
    {
      "entropy": 5.7339729309082035,
      "epoch": 0.8405797101449275,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004935682080904009,
      "loss": 5.5942,
      "mean_token_accuracy": 0.16424052268266678,
      "num_tokens": 18469977.0,
      "step": 10005
    },
    {
      "entropy": 5.734982681274414,
      "epoch": 0.8409997899600924,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004935610971225789,
      "loss": 5.5382,
      "mean_token_accuracy": 0.15915794372558595,
      "num_tokens": 18479534.0,
      "step": 10010
    },
    {
      "entropy": 5.69389705657959,
      "epoch": 0.8414198697752573,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004935539822830597,
      "loss": 5.6621,
      "mean_token_accuracy": 0.15215058922767638,
      "num_tokens": 18488800.0,
      "step": 10015
    },
    {
      "entropy": 5.745443725585938,
      "epoch": 0.8418399495904222,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000493546863571969,
      "loss": 5.6216,
      "mean_token_accuracy": 0.15481941550970077,
      "num_tokens": 18498083.0,
      "step": 10020
    },
    {
      "entropy": 5.7482555389404295,
      "epoch": 0.8422600294055871,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004935397409894333,
      "loss": 5.5776,
      "mean_token_accuracy": 0.14453350156545638,
      "num_tokens": 18508265.0,
      "step": 10025
    },
    {
      "entropy": 5.769057273864746,
      "epoch": 0.842680109220752,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004935326145355787,
      "loss": 5.6014,
      "mean_token_accuracy": 0.15536364465951918,
      "num_tokens": 18517283.0,
      "step": 10030
    },
    {
      "entropy": 5.758950281143188,
      "epoch": 0.8431001890359168,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004935254842105311,
      "loss": 5.6254,
      "mean_token_accuracy": 0.15739115327596664,
      "num_tokens": 18526482.0,
      "step": 10035
    },
    {
      "entropy": 5.663691616058349,
      "epoch": 0.8435202688510817,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004935183500144173,
      "loss": 5.4506,
      "mean_token_accuracy": 0.1704530283808708,
      "num_tokens": 18536150.0,
      "step": 10040
    },
    {
      "entropy": 5.828098583221435,
      "epoch": 0.8439403486662466,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004935112119473634,
      "loss": 5.6655,
      "mean_token_accuracy": 0.14966737627983093,
      "num_tokens": 18545168.0,
      "step": 10045
    },
    {
      "entropy": 5.721202039718628,
      "epoch": 0.8443604284814115,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004935040700094959,
      "loss": 5.5768,
      "mean_token_accuracy": 0.15967776775360107,
      "num_tokens": 18553363.0,
      "step": 10050
    },
    {
      "entropy": 5.646053218841553,
      "epoch": 0.8447805082965764,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004934969242009412,
      "loss": 5.5193,
      "mean_token_accuracy": 0.16305817514657975,
      "num_tokens": 18562546.0,
      "step": 10055
    },
    {
      "entropy": 5.729463911056518,
      "epoch": 0.8452005881117413,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004934897745218262,
      "loss": 5.6211,
      "mean_token_accuracy": 0.15121982246637344,
      "num_tokens": 18572149.0,
      "step": 10060
    },
    {
      "entropy": 5.745990800857544,
      "epoch": 0.8456206679269062,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004934826209722772,
      "loss": 5.4757,
      "mean_token_accuracy": 0.16152824461460114,
      "num_tokens": 18580842.0,
      "step": 10065
    },
    {
      "entropy": 5.741084957122803,
      "epoch": 0.8460407477420709,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004934754635524211,
      "loss": 5.5795,
      "mean_token_accuracy": 0.16007215827703475,
      "num_tokens": 18589765.0,
      "step": 10070
    },
    {
      "entropy": 5.711794137954712,
      "epoch": 0.8464608275572358,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004934683022623847,
      "loss": 5.6083,
      "mean_token_accuracy": 0.15259937196969986,
      "num_tokens": 18599532.0,
      "step": 10075
    },
    {
      "entropy": 5.65915470123291,
      "epoch": 0.8468809073724007,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004934611371022947,
      "loss": 5.5002,
      "mean_token_accuracy": 0.1630965918302536,
      "num_tokens": 18608438.0,
      "step": 10080
    },
    {
      "entropy": 5.7712366580963135,
      "epoch": 0.8473009871875656,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004934539680722783,
      "loss": 5.6625,
      "mean_token_accuracy": 0.15168794989585876,
      "num_tokens": 18617313.0,
      "step": 10085
    },
    {
      "entropy": 5.7000326156616214,
      "epoch": 0.8477210670027305,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004934467951724622,
      "loss": 5.4855,
      "mean_token_accuracy": 0.1603113070130348,
      "num_tokens": 18625880.0,
      "step": 10090
    },
    {
      "entropy": 5.690435123443604,
      "epoch": 0.8481411468178954,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004934396184029737,
      "loss": 5.5692,
      "mean_token_accuracy": 0.15285091400146483,
      "num_tokens": 18635727.0,
      "step": 10095
    },
    {
      "entropy": 5.747970628738403,
      "epoch": 0.8485612266330603,
      "grad_norm": 1.375,
      "learning_rate": 0.0004934324377639398,
      "loss": 5.621,
      "mean_token_accuracy": 0.15370510891079903,
      "num_tokens": 18645619.0,
      "step": 10100
    },
    {
      "entropy": 5.742952156066894,
      "epoch": 0.8489813064482251,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004934252532554878,
      "loss": 5.5197,
      "mean_token_accuracy": 0.1536563068628311,
      "num_tokens": 18654901.0,
      "step": 10105
    },
    {
      "entropy": 5.799785947799682,
      "epoch": 0.84940138626339,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004934180648777449,
      "loss": 5.7802,
      "mean_token_accuracy": 0.15087857395410537,
      "num_tokens": 18664523.0,
      "step": 10110
    },
    {
      "entropy": 5.763763666152954,
      "epoch": 0.8498214660785549,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004934108726308384,
      "loss": 5.6,
      "mean_token_accuracy": 0.15230715721845628,
      "num_tokens": 18673685.0,
      "step": 10115
    },
    {
      "entropy": 5.713348913192749,
      "epoch": 0.8502415458937198,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004934036765148958,
      "loss": 5.5767,
      "mean_token_accuracy": 0.15448796078562738,
      "num_tokens": 18682889.0,
      "step": 10120
    },
    {
      "entropy": 5.75453748703003,
      "epoch": 0.8506616257088847,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004933964765300446,
      "loss": 5.6178,
      "mean_token_accuracy": 0.1548921898007393,
      "num_tokens": 18692978.0,
      "step": 10125
    },
    {
      "entropy": 5.711153602600097,
      "epoch": 0.8510817055240496,
      "grad_norm": 1.390625,
      "learning_rate": 0.000493389272676412,
      "loss": 5.5298,
      "mean_token_accuracy": 0.15920393615961076,
      "num_tokens": 18701846.0,
      "step": 10130
    },
    {
      "entropy": 5.671429061889649,
      "epoch": 0.8515017853392145,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004933820649541262,
      "loss": 5.5761,
      "mean_token_accuracy": 0.16145771741867065,
      "num_tokens": 18711492.0,
      "step": 10135
    },
    {
      "entropy": 5.657542657852173,
      "epoch": 0.8519218651543793,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004933748533633145,
      "loss": 5.494,
      "mean_token_accuracy": 0.16596206575632094,
      "num_tokens": 18720407.0,
      "step": 10140
    },
    {
      "entropy": 5.707644414901734,
      "epoch": 0.8523419449695442,
      "grad_norm": 1.5,
      "learning_rate": 0.0004933676379041045,
      "loss": 5.5525,
      "mean_token_accuracy": 0.16160011291503906,
      "num_tokens": 18729968.0,
      "step": 10145
    },
    {
      "entropy": 5.740623044967651,
      "epoch": 0.8527620247847091,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004933604185766245,
      "loss": 5.6485,
      "mean_token_accuracy": 0.1480335585772991,
      "num_tokens": 18739525.0,
      "step": 10150
    },
    {
      "entropy": 5.733673620223999,
      "epoch": 0.853182104599874,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004933531953810019,
      "loss": 5.5518,
      "mean_token_accuracy": 0.16111933290958405,
      "num_tokens": 18749087.0,
      "step": 10155
    },
    {
      "entropy": 5.77870192527771,
      "epoch": 0.8536021844150389,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004933459683173652,
      "loss": 5.5768,
      "mean_token_accuracy": 0.1587308406829834,
      "num_tokens": 18758174.0,
      "step": 10160
    },
    {
      "entropy": 5.752344226837158,
      "epoch": 0.8540222642302038,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004933387373858418,
      "loss": 5.6233,
      "mean_token_accuracy": 0.15292344242334366,
      "num_tokens": 18767679.0,
      "step": 10165
    },
    {
      "entropy": 5.755201578140259,
      "epoch": 0.8544423440453687,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004933315025865602,
      "loss": 5.5476,
      "mean_token_accuracy": 0.15560670644044877,
      "num_tokens": 18776749.0,
      "step": 10170
    },
    {
      "entropy": 5.797035074234008,
      "epoch": 0.8548624238605335,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004933242639196485,
      "loss": 5.7157,
      "mean_token_accuracy": 0.14426579549908639,
      "num_tokens": 18786313.0,
      "step": 10175
    },
    {
      "entropy": 5.774004793167114,
      "epoch": 0.8552825036756984,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004933170213852348,
      "loss": 5.6112,
      "mean_token_accuracy": 0.1511244609951973,
      "num_tokens": 18795340.0,
      "step": 10180
    },
    {
      "entropy": 5.71902289390564,
      "epoch": 0.8557025834908633,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004933097749834476,
      "loss": 5.5333,
      "mean_token_accuracy": 0.1552624970674515,
      "num_tokens": 18804114.0,
      "step": 10185
    },
    {
      "entropy": 5.686375379562378,
      "epoch": 0.8561226633060282,
      "grad_norm": 1.625,
      "learning_rate": 0.000493302524714415,
      "loss": 5.5489,
      "mean_token_accuracy": 0.15422181487083436,
      "num_tokens": 18813797.0,
      "step": 10190
    },
    {
      "entropy": 5.722941493988037,
      "epoch": 0.856542743121193,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004932952705782657,
      "loss": 5.5966,
      "mean_token_accuracy": 0.15367913916707038,
      "num_tokens": 18822410.0,
      "step": 10195
    },
    {
      "entropy": 5.683314323425293,
      "epoch": 0.856962822936358,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000493288012575128,
      "loss": 5.5436,
      "mean_token_accuracy": 0.1596749320626259,
      "num_tokens": 18832091.0,
      "step": 10200
    },
    {
      "entropy": 5.7361071586608885,
      "epoch": 0.8573829027515227,
      "grad_norm": 1.375,
      "learning_rate": 0.0004932807507051307,
      "loss": 5.5526,
      "mean_token_accuracy": 0.15038529485464097,
      "num_tokens": 18841298.0,
      "step": 10205
    },
    {
      "entropy": 5.656687879562378,
      "epoch": 0.8578029825666876,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004932734849684022,
      "loss": 5.5347,
      "mean_token_accuracy": 0.1560280829668045,
      "num_tokens": 18849683.0,
      "step": 10210
    },
    {
      "entropy": 5.711640548706055,
      "epoch": 0.8582230623818525,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004932662153650712,
      "loss": 5.4793,
      "mean_token_accuracy": 0.15597545802593232,
      "num_tokens": 18858832.0,
      "step": 10215
    },
    {
      "entropy": 5.655262565612793,
      "epoch": 0.8586431421970174,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004932589418952668,
      "loss": 5.5174,
      "mean_token_accuracy": 0.1572958916425705,
      "num_tokens": 18867652.0,
      "step": 10220
    },
    {
      "entropy": 5.753128433227539,
      "epoch": 0.8590632220121823,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004932516645591175,
      "loss": 5.5962,
      "mean_token_accuracy": 0.15835005044937134,
      "num_tokens": 18877282.0,
      "step": 10225
    },
    {
      "entropy": 5.794670820236206,
      "epoch": 0.8594833018273472,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004932443833567524,
      "loss": 5.731,
      "mean_token_accuracy": 0.15675377175211908,
      "num_tokens": 18886565.0,
      "step": 10230
    },
    {
      "entropy": 5.803075695037842,
      "epoch": 0.8599033816425121,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004932370982883003,
      "loss": 5.6298,
      "mean_token_accuracy": 0.15467827767133713,
      "num_tokens": 18896440.0,
      "step": 10235
    },
    {
      "entropy": 5.814779329299927,
      "epoch": 0.8603234614576769,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004932298093538905,
      "loss": 5.6525,
      "mean_token_accuracy": 0.15493873208761216,
      "num_tokens": 18906246.0,
      "step": 10240
    },
    {
      "entropy": 5.695867204666138,
      "epoch": 0.8607435412728418,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000493222516553652,
      "loss": 5.5642,
      "mean_token_accuracy": 0.1503233715891838,
      "num_tokens": 18915108.0,
      "step": 10245
    },
    {
      "entropy": 5.709557580947876,
      "epoch": 0.8611636210880067,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004932152198877139,
      "loss": 5.5607,
      "mean_token_accuracy": 0.16035114824771882,
      "num_tokens": 18923664.0,
      "step": 10250
    },
    {
      "entropy": 5.74764666557312,
      "epoch": 0.8615837009031716,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004932079193562057,
      "loss": 5.6686,
      "mean_token_accuracy": 0.1497301295399666,
      "num_tokens": 18933496.0,
      "step": 10255
    },
    {
      "entropy": 5.679797077178955,
      "epoch": 0.8620037807183365,
      "grad_norm": 1.625,
      "learning_rate": 0.0004932006149592564,
      "loss": 5.5357,
      "mean_token_accuracy": 0.1623591274023056,
      "num_tokens": 18942222.0,
      "step": 10260
    },
    {
      "entropy": 5.8157978534698485,
      "epoch": 0.8624238605335014,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004931933066969957,
      "loss": 5.5669,
      "mean_token_accuracy": 0.15557005107402802,
      "num_tokens": 18952057.0,
      "step": 10265
    },
    {
      "entropy": 5.7592529296875,
      "epoch": 0.8628439403486663,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004931859945695528,
      "loss": 5.6168,
      "mean_token_accuracy": 0.1521659165620804,
      "num_tokens": 18961664.0,
      "step": 10270
    },
    {
      "entropy": 5.632269430160522,
      "epoch": 0.8632640201638311,
      "grad_norm": 1.75,
      "learning_rate": 0.0004931786785770575,
      "loss": 5.3931,
      "mean_token_accuracy": 0.1726113513112068,
      "num_tokens": 18969900.0,
      "step": 10275
    },
    {
      "entropy": 5.746176385879517,
      "epoch": 0.863684099978996,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004931713587196392,
      "loss": 5.6827,
      "mean_token_accuracy": 0.15060218945145606,
      "num_tokens": 18979286.0,
      "step": 10280
    },
    {
      "entropy": 5.798988389968872,
      "epoch": 0.8641041797941609,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004931640349974275,
      "loss": 5.5713,
      "mean_token_accuracy": 0.15988062992691993,
      "num_tokens": 18987553.0,
      "step": 10285
    },
    {
      "entropy": 5.792628335952759,
      "epoch": 0.8645242596093258,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004931567074105524,
      "loss": 5.658,
      "mean_token_accuracy": 0.15038609951734544,
      "num_tokens": 18996354.0,
      "step": 10290
    },
    {
      "entropy": 5.680066680908203,
      "epoch": 0.8649443394244907,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004931493759591435,
      "loss": 5.5531,
      "mean_token_accuracy": 0.15763965249061584,
      "num_tokens": 19005150.0,
      "step": 10295
    },
    {
      "entropy": 5.721858263015747,
      "epoch": 0.8653644192396556,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004931420406433308,
      "loss": 5.5557,
      "mean_token_accuracy": 0.1564570590853691,
      "num_tokens": 19014572.0,
      "step": 10300
    },
    {
      "entropy": 5.70249376296997,
      "epoch": 0.8657844990548205,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000493134701463244,
      "loss": 5.4212,
      "mean_token_accuracy": 0.1627628058195114,
      "num_tokens": 19023462.0,
      "step": 10305
    },
    {
      "entropy": 5.612477827072143,
      "epoch": 0.8662045788699853,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004931273584190135,
      "loss": 5.5076,
      "mean_token_accuracy": 0.15907565802335738,
      "num_tokens": 19032460.0,
      "step": 10310
    },
    {
      "entropy": 5.660311269760132,
      "epoch": 0.8666246586851502,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004931200115107691,
      "loss": 5.5494,
      "mean_token_accuracy": 0.16178966760635377,
      "num_tokens": 19041734.0,
      "step": 10315
    },
    {
      "entropy": 5.7060261249542235,
      "epoch": 0.867044738500315,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000493112660738641,
      "loss": 5.5361,
      "mean_token_accuracy": 0.15414502024650573,
      "num_tokens": 19050867.0,
      "step": 10320
    },
    {
      "entropy": 5.722083282470703,
      "epoch": 0.86746481831548,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004931053061027594,
      "loss": 5.5354,
      "mean_token_accuracy": 0.156645068526268,
      "num_tokens": 19060518.0,
      "step": 10325
    },
    {
      "entropy": 5.697010898590088,
      "epoch": 0.8678848981306448,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004930979476032546,
      "loss": 5.5213,
      "mean_token_accuracy": 0.16217473298311233,
      "num_tokens": 19069588.0,
      "step": 10330
    },
    {
      "entropy": 5.727666282653809,
      "epoch": 0.8683049779458097,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000493090585240257,
      "loss": 5.5772,
      "mean_token_accuracy": 0.14466141611337663,
      "num_tokens": 19079060.0,
      "step": 10335
    },
    {
      "entropy": 5.64642915725708,
      "epoch": 0.8687250577609746,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004930832190138969,
      "loss": 5.4841,
      "mean_token_accuracy": 0.15521228462457656,
      "num_tokens": 19087721.0,
      "step": 10340
    },
    {
      "entropy": 5.668660879135132,
      "epoch": 0.8691451375761394,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000493075848924305,
      "loss": 5.5447,
      "mean_token_accuracy": 0.15732968896627425,
      "num_tokens": 19096800.0,
      "step": 10345
    },
    {
      "entropy": 5.728336191177368,
      "epoch": 0.8695652173913043,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004930684749716117,
      "loss": 5.5913,
      "mean_token_accuracy": 0.15400783568620682,
      "num_tokens": 19106774.0,
      "step": 10350
    },
    {
      "entropy": 5.756087446212769,
      "epoch": 0.8699852972064692,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004930610971559476,
      "loss": 5.5439,
      "mean_token_accuracy": 0.1576864629983902,
      "num_tokens": 19116413.0,
      "step": 10355
    },
    {
      "entropy": 5.752991104125977,
      "epoch": 0.8704053770216341,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004930537154774436,
      "loss": 5.5522,
      "mean_token_accuracy": 0.15667675733566283,
      "num_tokens": 19125363.0,
      "step": 10360
    },
    {
      "entropy": 5.729405260086059,
      "epoch": 0.870825456836799,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004930463299362302,
      "loss": 5.6603,
      "mean_token_accuracy": 0.1461710177361965,
      "num_tokens": 19135461.0,
      "step": 10365
    },
    {
      "entropy": 5.793264532089234,
      "epoch": 0.8712455366519639,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004930389405324383,
      "loss": 5.5182,
      "mean_token_accuracy": 0.16262105852365494,
      "num_tokens": 19144085.0,
      "step": 10370
    },
    {
      "entropy": 5.733283281326294,
      "epoch": 0.8716656164671287,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004930315472661987,
      "loss": 5.5368,
      "mean_token_accuracy": 0.16369505524635314,
      "num_tokens": 19153291.0,
      "step": 10375
    },
    {
      "entropy": 5.709996700286865,
      "epoch": 0.8720856962822936,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004930241501376428,
      "loss": 5.5524,
      "mean_token_accuracy": 0.1532418116927147,
      "num_tokens": 19163514.0,
      "step": 10380
    },
    {
      "entropy": 5.618998575210571,
      "epoch": 0.8725057760974585,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004930167491469013,
      "loss": 5.447,
      "mean_token_accuracy": 0.16484934836626053,
      "num_tokens": 19172103.0,
      "step": 10385
    },
    {
      "entropy": 5.679058408737182,
      "epoch": 0.8729258559126234,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004930093442941053,
      "loss": 5.5135,
      "mean_token_accuracy": 0.15935876667499543,
      "num_tokens": 19180893.0,
      "step": 10390
    },
    {
      "entropy": 5.665105485916138,
      "epoch": 0.8733459357277883,
      "grad_norm": 1.625,
      "learning_rate": 0.0004930019355793858,
      "loss": 5.4451,
      "mean_token_accuracy": 0.16085383147001267,
      "num_tokens": 19190495.0,
      "step": 10395
    },
    {
      "entropy": 5.698199129104614,
      "epoch": 0.8737660155429532,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004929945230028746,
      "loss": 5.544,
      "mean_token_accuracy": 0.1626562222838402,
      "num_tokens": 19198988.0,
      "step": 10400
    },
    {
      "entropy": 5.6585792064666744,
      "epoch": 0.8741860953581181,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004929871065647024,
      "loss": 5.4557,
      "mean_token_accuracy": 0.16211153268814088,
      "num_tokens": 19208014.0,
      "step": 10405
    },
    {
      "entropy": 5.7339530944824215,
      "epoch": 0.8746061751732829,
      "grad_norm": 1.625,
      "learning_rate": 0.0004929796862650011,
      "loss": 5.6421,
      "mean_token_accuracy": 0.1563207045197487,
      "num_tokens": 19218220.0,
      "step": 10410
    },
    {
      "entropy": 5.693988180160522,
      "epoch": 0.8750262549884478,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004929722621039018,
      "loss": 5.5251,
      "mean_token_accuracy": 0.1552248328924179,
      "num_tokens": 19227176.0,
      "step": 10415
    },
    {
      "entropy": 5.644034004211425,
      "epoch": 0.8754463348036127,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004929648340815362,
      "loss": 5.5651,
      "mean_token_accuracy": 0.15514406561851501,
      "num_tokens": 19236085.0,
      "step": 10420
    },
    {
      "entropy": 5.694051218032837,
      "epoch": 0.8758664146187776,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004929574021980355,
      "loss": 5.6118,
      "mean_token_accuracy": 0.15097474902868271,
      "num_tokens": 19246671.0,
      "step": 10425
    },
    {
      "entropy": 5.706706523895264,
      "epoch": 0.8762864944339425,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004929499664535319,
      "loss": 5.5254,
      "mean_token_accuracy": 0.1542436182498932,
      "num_tokens": 19256321.0,
      "step": 10430
    },
    {
      "entropy": 5.661406517028809,
      "epoch": 0.8767065742491074,
      "grad_norm": 2.125,
      "learning_rate": 0.0004929425268481569,
      "loss": 5.4778,
      "mean_token_accuracy": 0.16220592856407165,
      "num_tokens": 19265518.0,
      "step": 10435
    },
    {
      "entropy": 5.673083209991455,
      "epoch": 0.8771266540642723,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004929350833820422,
      "loss": 5.4903,
      "mean_token_accuracy": 0.16280142813920975,
      "num_tokens": 19274120.0,
      "step": 10440
    },
    {
      "entropy": 5.7643969535827635,
      "epoch": 0.877546733879437,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004929276360553197,
      "loss": 5.5457,
      "mean_token_accuracy": 0.16031394377350808,
      "num_tokens": 19284377.0,
      "step": 10445
    },
    {
      "entropy": 5.61663851737976,
      "epoch": 0.8779668136946019,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004929201848681213,
      "loss": 5.4383,
      "mean_token_accuracy": 0.15702698826789857,
      "num_tokens": 19293326.0,
      "step": 10450
    },
    {
      "entropy": 5.622338199615479,
      "epoch": 0.8783868935097668,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004929127298205792,
      "loss": 5.4719,
      "mean_token_accuracy": 0.16489966809749604,
      "num_tokens": 19302086.0,
      "step": 10455
    },
    {
      "entropy": 5.729528093338013,
      "epoch": 0.8788069733249317,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004929052709128251,
      "loss": 5.5196,
      "mean_token_accuracy": 0.16125201433897018,
      "num_tokens": 19310124.0,
      "step": 10460
    },
    {
      "entropy": 5.654933595657349,
      "epoch": 0.8792270531400966,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004928978081449914,
      "loss": 5.5387,
      "mean_token_accuracy": 0.15293454825878144,
      "num_tokens": 19321269.0,
      "step": 10465
    },
    {
      "entropy": 5.675232601165772,
      "epoch": 0.8796471329552615,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004928903415172103,
      "loss": 5.5512,
      "mean_token_accuracy": 0.15898366868495942,
      "num_tokens": 19330390.0,
      "step": 10470
    },
    {
      "entropy": 5.7150938510894775,
      "epoch": 0.8800672127704264,
      "grad_norm": 1.9609375,
      "learning_rate": 0.000492882871029614,
      "loss": 5.5357,
      "mean_token_accuracy": 0.16283403784036637,
      "num_tokens": 19339457.0,
      "step": 10475
    },
    {
      "entropy": 5.688582849502564,
      "epoch": 0.8804872925855912,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004928753966823348,
      "loss": 5.6011,
      "mean_token_accuracy": 0.15648485273122786,
      "num_tokens": 19348710.0,
      "step": 10480
    },
    {
      "entropy": 5.686913347244262,
      "epoch": 0.8809073724007561,
      "grad_norm": 1.5,
      "learning_rate": 0.0004928679184755051,
      "loss": 5.6302,
      "mean_token_accuracy": 0.15994774252176286,
      "num_tokens": 19357215.0,
      "step": 10485
    },
    {
      "entropy": 5.690790605545044,
      "epoch": 0.881327452215921,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004928604364092574,
      "loss": 5.5702,
      "mean_token_accuracy": 0.16070739924907684,
      "num_tokens": 19366043.0,
      "step": 10490
    },
    {
      "entropy": 5.751299858093262,
      "epoch": 0.8817475320310859,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004928529504837243,
      "loss": 5.6335,
      "mean_token_accuracy": 0.15246885567903518,
      "num_tokens": 19375468.0,
      "step": 10495
    },
    {
      "entropy": 5.7539137840271,
      "epoch": 0.8821676118462508,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004928454606990383,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16103250831365584,
      "num_tokens": 19384467.0,
      "step": 10500
    },
    {
      "entropy": 5.682570075988769,
      "epoch": 0.8825876916614157,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004928379670553322,
      "loss": 5.5632,
      "mean_token_accuracy": 0.1597358301281929,
      "num_tokens": 19393618.0,
      "step": 10505
    },
    {
      "entropy": 5.729544639587402,
      "epoch": 0.8830077714765806,
      "grad_norm": 1.625,
      "learning_rate": 0.0004928304695527387,
      "loss": 5.609,
      "mean_token_accuracy": 0.15516545474529267,
      "num_tokens": 19402921.0,
      "step": 10510
    },
    {
      "entropy": 5.759986829757691,
      "epoch": 0.8834278512917454,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004928229681913905,
      "loss": 5.5892,
      "mean_token_accuracy": 0.15613035261631011,
      "num_tokens": 19412048.0,
      "step": 10515
    },
    {
      "entropy": 5.765500164031982,
      "epoch": 0.8838479311069103,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004928154629714207,
      "loss": 5.5768,
      "mean_token_accuracy": 0.15744251161813735,
      "num_tokens": 19420993.0,
      "step": 10520
    },
    {
      "entropy": 5.685095119476318,
      "epoch": 0.8842680109220752,
      "grad_norm": 1.671875,
      "learning_rate": 0.000492807953892962,
      "loss": 5.5523,
      "mean_token_accuracy": 0.15811701267957687,
      "num_tokens": 19430145.0,
      "step": 10525
    },
    {
      "entropy": 5.702225112915039,
      "epoch": 0.8846880907372401,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004928004409561476,
      "loss": 5.4636,
      "mean_token_accuracy": 0.1651538133621216,
      "num_tokens": 19438918.0,
      "step": 10530
    },
    {
      "entropy": 5.668575286865234,
      "epoch": 0.885108170552405,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004927929241611106,
      "loss": 5.518,
      "mean_token_accuracy": 0.16605824753642082,
      "num_tokens": 19448490.0,
      "step": 10535
    },
    {
      "entropy": 5.707067775726318,
      "epoch": 0.8855282503675699,
      "grad_norm": 2.140625,
      "learning_rate": 0.000492785403507984,
      "loss": 5.5652,
      "mean_token_accuracy": 0.15584489703178406,
      "num_tokens": 19457098.0,
      "step": 10540
    },
    {
      "entropy": 5.697981834411621,
      "epoch": 0.8859483301827347,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004927778789969012,
      "loss": 5.5594,
      "mean_token_accuracy": 0.1538078561425209,
      "num_tokens": 19466419.0,
      "step": 10545
    },
    {
      "entropy": 5.703227710723877,
      "epoch": 0.8863684099978996,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004927703506279955,
      "loss": 5.6216,
      "mean_token_accuracy": 0.15174467712640763,
      "num_tokens": 19475882.0,
      "step": 10550
    },
    {
      "entropy": 5.798790454864502,
      "epoch": 0.8867884898130645,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004927628184014,
      "loss": 5.6533,
      "mean_token_accuracy": 0.1536669135093689,
      "num_tokens": 19485917.0,
      "step": 10555
    },
    {
      "entropy": 5.729129076004028,
      "epoch": 0.8872085696282294,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004927552823172483,
      "loss": 5.5768,
      "mean_token_accuracy": 0.15445511490106584,
      "num_tokens": 19494984.0,
      "step": 10560
    },
    {
      "entropy": 5.739677906036377,
      "epoch": 0.8876286494433943,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000492747742375674,
      "loss": 5.5166,
      "mean_token_accuracy": 0.16236048340797424,
      "num_tokens": 19504087.0,
      "step": 10565
    },
    {
      "entropy": 5.720370864868164,
      "epoch": 0.8880487292585592,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004927401985768106,
      "loss": 5.5661,
      "mean_token_accuracy": 0.16260765492916107,
      "num_tokens": 19512880.0,
      "step": 10570
    },
    {
      "entropy": 5.652678537368774,
      "epoch": 0.888468809073724,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004927326509207915,
      "loss": 5.5515,
      "mean_token_accuracy": 0.1601003259420395,
      "num_tokens": 19521723.0,
      "step": 10575
    },
    {
      "entropy": 5.738314771652222,
      "epoch": 0.8888888888888888,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004927250994077508,
      "loss": 5.6127,
      "mean_token_accuracy": 0.15562157332897186,
      "num_tokens": 19531352.0,
      "step": 10580
    },
    {
      "entropy": 5.794875717163086,
      "epoch": 0.8893089687040537,
      "grad_norm": 2.359375,
      "learning_rate": 0.000492717544037822,
      "loss": 5.7321,
      "mean_token_accuracy": 0.15739693194627763,
      "num_tokens": 19540943.0,
      "step": 10585
    },
    {
      "entropy": 5.683047342300415,
      "epoch": 0.8897290485192186,
      "grad_norm": 1.9609375,
      "learning_rate": 0.000492709984811139,
      "loss": 5.4835,
      "mean_token_accuracy": 0.16288544684648515,
      "num_tokens": 19550527.0,
      "step": 10590
    },
    {
      "entropy": 5.669373846054077,
      "epoch": 0.8901491283343835,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004927024217278358,
      "loss": 5.4758,
      "mean_token_accuracy": 0.16461390554904937,
      "num_tokens": 19559746.0,
      "step": 10595
    },
    {
      "entropy": 5.729438066482544,
      "epoch": 0.8905692081495484,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004926948547880462,
      "loss": 5.636,
      "mean_token_accuracy": 0.1517734244465828,
      "num_tokens": 19569286.0,
      "step": 10600
    },
    {
      "entropy": 5.695087623596192,
      "epoch": 0.8909892879647133,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004926872839919044,
      "loss": 5.5164,
      "mean_token_accuracy": 0.15850782245397568,
      "num_tokens": 19578245.0,
      "step": 10605
    },
    {
      "entropy": 5.637440395355225,
      "epoch": 0.8914093677798782,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004926797093395446,
      "loss": 5.4863,
      "mean_token_accuracy": 0.16317880302667617,
      "num_tokens": 19587244.0,
      "step": 10610
    },
    {
      "entropy": 5.670547389984131,
      "epoch": 0.891829447595043,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004926721308311006,
      "loss": 5.5628,
      "mean_token_accuracy": 0.16251896917819977,
      "num_tokens": 19596932.0,
      "step": 10615
    },
    {
      "entropy": 5.772653007507325,
      "epoch": 0.8922495274102079,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004926645484667069,
      "loss": 5.6799,
      "mean_token_accuracy": 0.15312184244394303,
      "num_tokens": 19606256.0,
      "step": 10620
    },
    {
      "entropy": 5.808226013183594,
      "epoch": 0.8926696072253728,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004926569622464979,
      "loss": 5.6835,
      "mean_token_accuracy": 0.1561840444803238,
      "num_tokens": 19615726.0,
      "step": 10625
    },
    {
      "entropy": 5.75749716758728,
      "epoch": 0.8930896870405377,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004926493721706079,
      "loss": 5.5446,
      "mean_token_accuracy": 0.15656498074531555,
      "num_tokens": 19624037.0,
      "step": 10630
    },
    {
      "entropy": 5.733023357391358,
      "epoch": 0.8935097668557026,
      "grad_norm": 1.875,
      "learning_rate": 0.0004926417782391713,
      "loss": 5.5234,
      "mean_token_accuracy": 0.16201919168233872,
      "num_tokens": 19632882.0,
      "step": 10635
    },
    {
      "entropy": 5.693700790405273,
      "epoch": 0.8939298466708675,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004926341804523227,
      "loss": 5.6482,
      "mean_token_accuracy": 0.15508821457624436,
      "num_tokens": 19642686.0,
      "step": 10640
    },
    {
      "entropy": 5.671689987182617,
      "epoch": 0.8943499264860324,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004926265788101966,
      "loss": 5.5437,
      "mean_token_accuracy": 0.1606654331088066,
      "num_tokens": 19651380.0,
      "step": 10645
    },
    {
      "entropy": 5.678468942642212,
      "epoch": 0.8947700063011972,
      "grad_norm": 2.0,
      "learning_rate": 0.0004926189733129278,
      "loss": 5.4821,
      "mean_token_accuracy": 0.159104885160923,
      "num_tokens": 19660136.0,
      "step": 10650
    },
    {
      "entropy": 5.646151065826416,
      "epoch": 0.8951900861163621,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004926113639606509,
      "loss": 5.525,
      "mean_token_accuracy": 0.16899669989943505,
      "num_tokens": 19669146.0,
      "step": 10655
    },
    {
      "entropy": 5.76297607421875,
      "epoch": 0.895610165931527,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004926037507535008,
      "loss": 5.6574,
      "mean_token_accuracy": 0.15518373548984526,
      "num_tokens": 19678627.0,
      "step": 10660
    },
    {
      "entropy": 5.736645460128784,
      "epoch": 0.8960302457466919,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004925961336916122,
      "loss": 5.5778,
      "mean_token_accuracy": 0.15911105573177337,
      "num_tokens": 19688033.0,
      "step": 10665
    },
    {
      "entropy": 5.739694452285766,
      "epoch": 0.8964503255618568,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004925885127751202,
      "loss": 5.5968,
      "mean_token_accuracy": 0.16282687336206436,
      "num_tokens": 19696523.0,
      "step": 10670
    },
    {
      "entropy": 5.778036117553711,
      "epoch": 0.8968704053770217,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004925808880041596,
      "loss": 5.5251,
      "mean_token_accuracy": 0.15527614951133728,
      "num_tokens": 19706339.0,
      "step": 10675
    },
    {
      "entropy": 5.7975170612335205,
      "epoch": 0.8972904851921865,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004925732593788658,
      "loss": 5.5636,
      "mean_token_accuracy": 0.15571669340133668,
      "num_tokens": 19714779.0,
      "step": 10680
    },
    {
      "entropy": 5.723025703430176,
      "epoch": 0.8977105650073514,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004925656268993737,
      "loss": 5.6257,
      "mean_token_accuracy": 0.15665646344423295,
      "num_tokens": 19723727.0,
      "step": 10685
    },
    {
      "entropy": 5.654568767547607,
      "epoch": 0.8981306448225163,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004925579905658185,
      "loss": 5.5947,
      "mean_token_accuracy": 0.15805498957633973,
      "num_tokens": 19732783.0,
      "step": 10690
    },
    {
      "entropy": 5.73252534866333,
      "epoch": 0.8985507246376812,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004925503503783355,
      "loss": 5.5656,
      "mean_token_accuracy": 0.15477960258722306,
      "num_tokens": 19741268.0,
      "step": 10695
    },
    {
      "entropy": 5.781156778335571,
      "epoch": 0.898970804452846,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004925427063370601,
      "loss": 5.489,
      "mean_token_accuracy": 0.1559618294239044,
      "num_tokens": 19751490.0,
      "step": 10700
    },
    {
      "entropy": 5.717162132263184,
      "epoch": 0.899390884268011,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004925350584421278,
      "loss": 5.5318,
      "mean_token_accuracy": 0.1574894294142723,
      "num_tokens": 19760487.0,
      "step": 10705
    },
    {
      "entropy": 5.703652477264404,
      "epoch": 0.8998109640831758,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004925274066936738,
      "loss": 5.4923,
      "mean_token_accuracy": 0.16476034224033356,
      "num_tokens": 19768984.0,
      "step": 10710
    },
    {
      "entropy": 5.730563688278198,
      "epoch": 0.9002310438983406,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004925197510918339,
      "loss": 5.5004,
      "mean_token_accuracy": 0.16111984401941298,
      "num_tokens": 19778335.0,
      "step": 10715
    },
    {
      "entropy": 5.710481452941894,
      "epoch": 0.9006511237135055,
      "grad_norm": 2.0,
      "learning_rate": 0.0004925120916367435,
      "loss": 5.6523,
      "mean_token_accuracy": 0.1486131727695465,
      "num_tokens": 19789082.0,
      "step": 10720
    },
    {
      "entropy": 5.637285423278809,
      "epoch": 0.9010712035286704,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004925044283285384,
      "loss": 5.353,
      "mean_token_accuracy": 0.1743251711130142,
      "num_tokens": 19797902.0,
      "step": 10725
    },
    {
      "entropy": 5.603871488571167,
      "epoch": 0.9014912833438353,
      "grad_norm": 1.5,
      "learning_rate": 0.0004924967611673544,
      "loss": 5.5226,
      "mean_token_accuracy": 0.16380647271871568,
      "num_tokens": 19806481.0,
      "step": 10730
    },
    {
      "entropy": 5.588102340698242,
      "epoch": 0.9019113631590002,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004924890901533273,
      "loss": 5.4262,
      "mean_token_accuracy": 0.172759747505188,
      "num_tokens": 19815226.0,
      "step": 10735
    },
    {
      "entropy": 5.783882665634155,
      "epoch": 0.9023314429741651,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004924814152865929,
      "loss": 5.6552,
      "mean_token_accuracy": 0.1513750970363617,
      "num_tokens": 19824577.0,
      "step": 10740
    },
    {
      "entropy": 5.754131412506103,
      "epoch": 0.90275152278933,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004924737365672873,
      "loss": 5.5549,
      "mean_token_accuracy": 0.1550630509853363,
      "num_tokens": 19832936.0,
      "step": 10745
    },
    {
      "entropy": 5.814233350753784,
      "epoch": 0.9031716026044948,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004924660539955463,
      "loss": 5.6924,
      "mean_token_accuracy": 0.1609632506966591,
      "num_tokens": 19841946.0,
      "step": 10750
    },
    {
      "entropy": 5.760442447662354,
      "epoch": 0.9035916824196597,
      "grad_norm": 2.25,
      "learning_rate": 0.0004924583675715063,
      "loss": 5.5839,
      "mean_token_accuracy": 0.15627699494361877,
      "num_tokens": 19851469.0,
      "step": 10755
    },
    {
      "entropy": 5.779312658309936,
      "epoch": 0.9040117622348246,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004924506772953031,
      "loss": 5.6631,
      "mean_token_accuracy": 0.15602973401546477,
      "num_tokens": 19860731.0,
      "step": 10760
    },
    {
      "entropy": 5.787699460983276,
      "epoch": 0.9044318420499895,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004924429831670733,
      "loss": 5.6568,
      "mean_token_accuracy": 0.15296549201011658,
      "num_tokens": 19869717.0,
      "step": 10765
    },
    {
      "entropy": 5.779390621185303,
      "epoch": 0.9048519218651544,
      "grad_norm": 1.40625,
      "learning_rate": 0.000492435285186953,
      "loss": 5.6022,
      "mean_token_accuracy": 0.16025329232215882,
      "num_tokens": 19879229.0,
      "step": 10770
    },
    {
      "entropy": 5.836106300354004,
      "epoch": 0.9052720016803193,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004924275833550785,
      "loss": 5.6011,
      "mean_token_accuracy": 0.15716341137886047,
      "num_tokens": 19888260.0,
      "step": 10775
    },
    {
      "entropy": 5.820453453063965,
      "epoch": 0.9056920814954842,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004924198776715865,
      "loss": 5.6123,
      "mean_token_accuracy": 0.16213833093643187,
      "num_tokens": 19897070.0,
      "step": 10780
    },
    {
      "entropy": 5.75965895652771,
      "epoch": 0.906112161310649,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004924121681366132,
      "loss": 5.603,
      "mean_token_accuracy": 0.15155648514628411,
      "num_tokens": 19907170.0,
      "step": 10785
    },
    {
      "entropy": 5.723458623886108,
      "epoch": 0.9065322411258139,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004924044547502951,
      "loss": 5.5437,
      "mean_token_accuracy": 0.15625743418931962,
      "num_tokens": 19917220.0,
      "step": 10790
    },
    {
      "entropy": 5.658956575393677,
      "epoch": 0.9069523209409788,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004923967375127692,
      "loss": 5.6019,
      "mean_token_accuracy": 0.1578077644109726,
      "num_tokens": 19926724.0,
      "step": 10795
    },
    {
      "entropy": 5.811443948745728,
      "epoch": 0.9073724007561437,
      "grad_norm": 2.5,
      "learning_rate": 0.000492389016424172,
      "loss": 5.7088,
      "mean_token_accuracy": 0.1508877694606781,
      "num_tokens": 19936429.0,
      "step": 10800
    },
    {
      "entropy": 5.7526062488555905,
      "epoch": 0.9077924805713086,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004923812914846404,
      "loss": 5.4788,
      "mean_token_accuracy": 0.1610741063952446,
      "num_tokens": 19945096.0,
      "step": 10805
    },
    {
      "entropy": 5.742060136795044,
      "epoch": 0.9082125603864735,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004923735626943111,
      "loss": 5.5539,
      "mean_token_accuracy": 0.16053722351789473,
      "num_tokens": 19953560.0,
      "step": 10810
    },
    {
      "entropy": 5.70956597328186,
      "epoch": 0.9086326402016384,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004923658300533211,
      "loss": 5.5411,
      "mean_token_accuracy": 0.15640968233346939,
      "num_tokens": 19962669.0,
      "step": 10815
    },
    {
      "entropy": 5.794610786437988,
      "epoch": 0.9090527200168032,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004923580935618073,
      "loss": 5.5757,
      "mean_token_accuracy": 0.1538527265191078,
      "num_tokens": 19971990.0,
      "step": 10820
    },
    {
      "entropy": 5.746137094497681,
      "epoch": 0.909472799831968,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004923503532199069,
      "loss": 5.5822,
      "mean_token_accuracy": 0.15799881666898727,
      "num_tokens": 19981850.0,
      "step": 10825
    },
    {
      "entropy": 5.759499168395996,
      "epoch": 0.909892879647133,
      "grad_norm": 1.75,
      "learning_rate": 0.0004923426090277567,
      "loss": 5.6109,
      "mean_token_accuracy": 0.15298285335302353,
      "num_tokens": 19991574.0,
      "step": 10830
    },
    {
      "entropy": 5.754549360275268,
      "epoch": 0.9103129594622978,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004923348609854943,
      "loss": 5.5906,
      "mean_token_accuracy": 0.15852705538272857,
      "num_tokens": 20001392.0,
      "step": 10835
    },
    {
      "entropy": 5.73292727470398,
      "epoch": 0.9107330392774627,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004923271090932566,
      "loss": 5.6202,
      "mean_token_accuracy": 0.1504620835185051,
      "num_tokens": 20011277.0,
      "step": 10840
    },
    {
      "entropy": 5.673085355758667,
      "epoch": 0.9111531190926276,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004923193533511812,
      "loss": 5.536,
      "mean_token_accuracy": 0.15325282141566277,
      "num_tokens": 20021171.0,
      "step": 10845
    },
    {
      "entropy": 5.827489376068115,
      "epoch": 0.9115731989077924,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004923115937594053,
      "loss": 5.61,
      "mean_token_accuracy": 0.15670632869005202,
      "num_tokens": 20030189.0,
      "step": 10850
    },
    {
      "entropy": 5.8075799465179445,
      "epoch": 0.9119932787229573,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004923038303180664,
      "loss": 5.5889,
      "mean_token_accuracy": 0.1611725553870201,
      "num_tokens": 20038287.0,
      "step": 10855
    },
    {
      "entropy": 5.653801584243775,
      "epoch": 0.9124133585381222,
      "grad_norm": 3.515625,
      "learning_rate": 0.000492296063027302,
      "loss": 5.5977,
      "mean_token_accuracy": 0.15539312362670898,
      "num_tokens": 20047653.0,
      "step": 10860
    },
    {
      "entropy": 5.692626237869263,
      "epoch": 0.9128334383532871,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004922882918872498,
      "loss": 5.5947,
      "mean_token_accuracy": 0.1545352503657341,
      "num_tokens": 20057415.0,
      "step": 10865
    },
    {
      "entropy": 5.82832670211792,
      "epoch": 0.913253518168452,
      "grad_norm": 1.625,
      "learning_rate": 0.0004922805168980475,
      "loss": 5.5944,
      "mean_token_accuracy": 0.15924015045166015,
      "num_tokens": 20065996.0,
      "step": 10870
    },
    {
      "entropy": 5.750165605545044,
      "epoch": 0.9136735979836169,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004922727380598326,
      "loss": 5.5405,
      "mean_token_accuracy": 0.15805808305740357,
      "num_tokens": 20075376.0,
      "step": 10875
    },
    {
      "entropy": 5.684902000427246,
      "epoch": 0.9140936777987818,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000492264955372743,
      "loss": 5.5865,
      "mean_token_accuracy": 0.14841223880648613,
      "num_tokens": 20084950.0,
      "step": 10880
    },
    {
      "entropy": 5.780635118484497,
      "epoch": 0.9145137576139466,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004922571688369165,
      "loss": 5.5569,
      "mean_token_accuracy": 0.16063966900110244,
      "num_tokens": 20094011.0,
      "step": 10885
    },
    {
      "entropy": 5.734232330322266,
      "epoch": 0.9149338374291115,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004922493784524914,
      "loss": 5.5363,
      "mean_token_accuracy": 0.163571584969759,
      "num_tokens": 20103037.0,
      "step": 10890
    },
    {
      "entropy": 5.757091379165649,
      "epoch": 0.9153539172442764,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004922415842196052,
      "loss": 5.6651,
      "mean_token_accuracy": 0.14362147375941275,
      "num_tokens": 20112727.0,
      "step": 10895
    },
    {
      "entropy": 5.6311070919036865,
      "epoch": 0.9157739970594413,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004922337861383963,
      "loss": 5.496,
      "mean_token_accuracy": 0.16367468982934952,
      "num_tokens": 20122341.0,
      "step": 10900
    },
    {
      "entropy": 5.730766153335571,
      "epoch": 0.9161940768746062,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004922259842090027,
      "loss": 5.4708,
      "mean_token_accuracy": 0.16051837801933289,
      "num_tokens": 20131354.0,
      "step": 10905
    },
    {
      "entropy": 5.721527194976806,
      "epoch": 0.9166141566897711,
      "grad_norm": 1.5,
      "learning_rate": 0.0004922181784315627,
      "loss": 5.5248,
      "mean_token_accuracy": 0.16043773144483567,
      "num_tokens": 20140440.0,
      "step": 10910
    },
    {
      "entropy": 5.667795944213867,
      "epoch": 0.917034236504936,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004922103688062145,
      "loss": 5.5163,
      "mean_token_accuracy": 0.16204050332307815,
      "num_tokens": 20149331.0,
      "step": 10915
    },
    {
      "entropy": 5.7060816287994385,
      "epoch": 0.9174543163201008,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004922025553330964,
      "loss": 5.4956,
      "mean_token_accuracy": 0.16326858401298522,
      "num_tokens": 20158566.0,
      "step": 10920
    },
    {
      "entropy": 5.75770206451416,
      "epoch": 0.9178743961352657,
      "grad_norm": 1.546875,
      "learning_rate": 0.000492194738012347,
      "loss": 5.5981,
      "mean_token_accuracy": 0.15985073149204254,
      "num_tokens": 20168339.0,
      "step": 10925
    },
    {
      "entropy": 5.830264472961426,
      "epoch": 0.9182944759504306,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004921869168441045,
      "loss": 5.6149,
      "mean_token_accuracy": 0.1499125950038433,
      "num_tokens": 20177967.0,
      "step": 10930
    },
    {
      "entropy": 5.692331171035766,
      "epoch": 0.9187145557655955,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004921790918285077,
      "loss": 5.6048,
      "mean_token_accuracy": 0.1557812660932541,
      "num_tokens": 20187279.0,
      "step": 10935
    },
    {
      "entropy": 5.700629711151123,
      "epoch": 0.9191346355807604,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004921712629656951,
      "loss": 5.7076,
      "mean_token_accuracy": 0.16659070774912835,
      "num_tokens": 20195324.0,
      "step": 10940
    },
    {
      "entropy": 5.841782522201538,
      "epoch": 0.9195547153959253,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004921634302558054,
      "loss": 5.6266,
      "mean_token_accuracy": 0.15811780244112014,
      "num_tokens": 20204985.0,
      "step": 10945
    },
    {
      "entropy": 5.764072608947754,
      "epoch": 0.9199747952110902,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004921555936989773,
      "loss": 5.6432,
      "mean_token_accuracy": 0.15192149430513383,
      "num_tokens": 20214553.0,
      "step": 10950
    },
    {
      "entropy": 5.758094644546508,
      "epoch": 0.9203948750262549,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004921477532953497,
      "loss": 5.5565,
      "mean_token_accuracy": 0.15732019394636154,
      "num_tokens": 20224118.0,
      "step": 10955
    },
    {
      "entropy": 5.722699308395386,
      "epoch": 0.9208149548414198,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004921399090450616,
      "loss": 5.4984,
      "mean_token_accuracy": 0.15792709290981294,
      "num_tokens": 20233719.0,
      "step": 10960
    },
    {
      "entropy": 5.747791290283203,
      "epoch": 0.9212350346565847,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004921320609482517,
      "loss": 5.6043,
      "mean_token_accuracy": 0.1583334594964981,
      "num_tokens": 20242311.0,
      "step": 10965
    },
    {
      "entropy": 5.80935845375061,
      "epoch": 0.9216551144717496,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004921242090050591,
      "loss": 5.6304,
      "mean_token_accuracy": 0.15593896806240082,
      "num_tokens": 20252998.0,
      "step": 10970
    },
    {
      "entropy": 5.825342416763306,
      "epoch": 0.9220751942869145,
      "grad_norm": 2.234375,
      "learning_rate": 0.000492116353215623,
      "loss": 5.673,
      "mean_token_accuracy": 0.15853351205587388,
      "num_tokens": 20262456.0,
      "step": 10975
    },
    {
      "entropy": 5.665419816970825,
      "epoch": 0.9224952741020794,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004921084935800825,
      "loss": 5.4387,
      "mean_token_accuracy": 0.16515873968601227,
      "num_tokens": 20271516.0,
      "step": 10980
    },
    {
      "entropy": 5.676522207260132,
      "epoch": 0.9229153539172443,
      "grad_norm": 1.875,
      "learning_rate": 0.0004921006300985768,
      "loss": 5.4974,
      "mean_token_accuracy": 0.16081881821155547,
      "num_tokens": 20280373.0,
      "step": 10985
    },
    {
      "entropy": 5.7067461013793945,
      "epoch": 0.9233354337324091,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004920927627712453,
      "loss": 5.4998,
      "mean_token_accuracy": 0.16111356168985366,
      "num_tokens": 20289426.0,
      "step": 10990
    },
    {
      "entropy": 5.799027299880981,
      "epoch": 0.923755513547574,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004920848915982273,
      "loss": 5.6429,
      "mean_token_accuracy": 0.15812461227178573,
      "num_tokens": 20298045.0,
      "step": 10995
    },
    {
      "entropy": 5.667883491516113,
      "epoch": 0.9241755933627389,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004920770165796622,
      "loss": 5.5306,
      "mean_token_accuracy": 0.15947688817977906,
      "num_tokens": 20307352.0,
      "step": 11000
    },
    {
      "entropy": 5.72949481010437,
      "epoch": 0.9245956731779038,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004920691377156895,
      "loss": 5.5577,
      "mean_token_accuracy": 0.15705136358737945,
      "num_tokens": 20316448.0,
      "step": 11005
    },
    {
      "entropy": 5.833560037612915,
      "epoch": 0.9250157529930687,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004920612550064488,
      "loss": 5.712,
      "mean_token_accuracy": 0.15252626091241836,
      "num_tokens": 20326440.0,
      "step": 11010
    },
    {
      "entropy": 5.694369506835938,
      "epoch": 0.9254358328082336,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004920533684520797,
      "loss": 5.4748,
      "mean_token_accuracy": 0.16088208854198455,
      "num_tokens": 20335447.0,
      "step": 11015
    },
    {
      "entropy": 5.68902063369751,
      "epoch": 0.9258559126233984,
      "grad_norm": 1.640625,
      "learning_rate": 0.000492045478052722,
      "loss": 5.6386,
      "mean_token_accuracy": 0.15640645027160643,
      "num_tokens": 20344523.0,
      "step": 11020
    },
    {
      "entropy": 5.749677991867065,
      "epoch": 0.9262759924385633,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004920375838085154,
      "loss": 5.5937,
      "mean_token_accuracy": 0.15958297103643418,
      "num_tokens": 20354267.0,
      "step": 11025
    },
    {
      "entropy": 5.802869653701782,
      "epoch": 0.9266960722537282,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004920296857195998,
      "loss": 5.6479,
      "mean_token_accuracy": 0.1565267562866211,
      "num_tokens": 20364137.0,
      "step": 11030
    },
    {
      "entropy": 5.738079071044922,
      "epoch": 0.9271161520688931,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000492021783786115,
      "loss": 5.5488,
      "mean_token_accuracy": 0.16330525726079942,
      "num_tokens": 20372583.0,
      "step": 11035
    },
    {
      "entropy": 5.696683788299561,
      "epoch": 0.927536231884058,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004920138780082011,
      "loss": 5.4984,
      "mean_token_accuracy": 0.1606284871697426,
      "num_tokens": 20382050.0,
      "step": 11040
    },
    {
      "entropy": 5.689135313034058,
      "epoch": 0.9279563116992229,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004920059683859981,
      "loss": 5.4654,
      "mean_token_accuracy": 0.16342781931161882,
      "num_tokens": 20391425.0,
      "step": 11045
    },
    {
      "entropy": 5.7500804424285885,
      "epoch": 0.9283763915143878,
      "grad_norm": 1.75,
      "learning_rate": 0.0004919980549196461,
      "loss": 5.6327,
      "mean_token_accuracy": 0.1570240467786789,
      "num_tokens": 20400559.0,
      "step": 11050
    },
    {
      "entropy": 5.770569467544556,
      "epoch": 0.9287964713295526,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004919901376092853,
      "loss": 5.5516,
      "mean_token_accuracy": 0.15826557129621505,
      "num_tokens": 20408985.0,
      "step": 11055
    },
    {
      "entropy": 5.722348403930664,
      "epoch": 0.9292165511447175,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004919822164550559,
      "loss": 5.6533,
      "mean_token_accuracy": 0.14333268254995346,
      "num_tokens": 20417855.0,
      "step": 11060
    },
    {
      "entropy": 5.745558738708496,
      "epoch": 0.9296366309598824,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004919742914570983,
      "loss": 5.61,
      "mean_token_accuracy": 0.1635189712047577,
      "num_tokens": 20426191.0,
      "step": 11065
    },
    {
      "entropy": 5.728816938400269,
      "epoch": 0.9300567107750473,
      "grad_norm": 1.484375,
      "learning_rate": 0.000491966362615553,
      "loss": 5.5954,
      "mean_token_accuracy": 0.14999182894825935,
      "num_tokens": 20435592.0,
      "step": 11070
    },
    {
      "entropy": 5.795909309387207,
      "epoch": 0.9304767905902122,
      "grad_norm": 1.671875,
      "learning_rate": 0.00049195842993056,
      "loss": 5.6229,
      "mean_token_accuracy": 0.15422592610120772,
      "num_tokens": 20445504.0,
      "step": 11075
    },
    {
      "entropy": 5.764322519302368,
      "epoch": 0.930896870405377,
      "grad_norm": 1.75,
      "learning_rate": 0.0004919504934022604,
      "loss": 5.542,
      "mean_token_accuracy": 0.15572449266910554,
      "num_tokens": 20455153.0,
      "step": 11080
    },
    {
      "entropy": 5.667031097412109,
      "epoch": 0.931316950220542,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004919425530307943,
      "loss": 5.5469,
      "mean_token_accuracy": 0.1548906832933426,
      "num_tokens": 20465101.0,
      "step": 11085
    },
    {
      "entropy": 5.717367219924927,
      "epoch": 0.9317370300357067,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004919346088163028,
      "loss": 5.6009,
      "mean_token_accuracy": 0.15931420922279357,
      "num_tokens": 20474700.0,
      "step": 11090
    },
    {
      "entropy": 5.774007034301758,
      "epoch": 0.9321571098508716,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004919266607589263,
      "loss": 5.6202,
      "mean_token_accuracy": 0.1513561874628067,
      "num_tokens": 20483945.0,
      "step": 11095
    },
    {
      "entropy": 5.737395286560059,
      "epoch": 0.9325771896660365,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004919187088588057,
      "loss": 5.598,
      "mean_token_accuracy": 0.15983821004629134,
      "num_tokens": 20493307.0,
      "step": 11100
    },
    {
      "entropy": 5.709204387664795,
      "epoch": 0.9329972694812014,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004919107531160819,
      "loss": 5.511,
      "mean_token_accuracy": 0.16708060055971147,
      "num_tokens": 20501889.0,
      "step": 11105
    },
    {
      "entropy": 5.6891406059265135,
      "epoch": 0.9334173492963663,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004919027935308957,
      "loss": 5.5612,
      "mean_token_accuracy": 0.16252549216151238,
      "num_tokens": 20510577.0,
      "step": 11110
    },
    {
      "entropy": 5.646807098388672,
      "epoch": 0.9338374291115312,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004918948301033884,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1584120899438858,
      "num_tokens": 20520025.0,
      "step": 11115
    },
    {
      "entropy": 5.815218305587768,
      "epoch": 0.9342575089266961,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004918868628337007,
      "loss": 5.5602,
      "mean_token_accuracy": 0.16129291653633118,
      "num_tokens": 20528989.0,
      "step": 11120
    },
    {
      "entropy": 5.745483779907227,
      "epoch": 0.9346775887418609,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004918788917219739,
      "loss": 5.5327,
      "mean_token_accuracy": 0.15650796443223952,
      "num_tokens": 20538328.0,
      "step": 11125
    },
    {
      "entropy": 5.694914293289185,
      "epoch": 0.9350976685570258,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004918709167683493,
      "loss": 5.6755,
      "mean_token_accuracy": 0.14997887313365937,
      "num_tokens": 20548069.0,
      "step": 11130
    },
    {
      "entropy": 5.676989889144897,
      "epoch": 0.9355177483721907,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004918629379729681,
      "loss": 5.4148,
      "mean_token_accuracy": 0.16828625798225402,
      "num_tokens": 20557128.0,
      "step": 11135
    },
    {
      "entropy": 5.675128316879272,
      "epoch": 0.9359378281873556,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004918549553359715,
      "loss": 5.5333,
      "mean_token_accuracy": 0.15687671005725862,
      "num_tokens": 20566352.0,
      "step": 11140
    },
    {
      "entropy": 5.729965591430664,
      "epoch": 0.9363579080025205,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004918469688575012,
      "loss": 5.5638,
      "mean_token_accuracy": 0.15703129321336745,
      "num_tokens": 20575814.0,
      "step": 11145
    },
    {
      "entropy": 5.715324926376343,
      "epoch": 0.9367779878176854,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004918389785376983,
      "loss": 5.4586,
      "mean_token_accuracy": 0.15755079239606856,
      "num_tokens": 20584715.0,
      "step": 11150
    },
    {
      "entropy": 5.646192789077759,
      "epoch": 0.9371980676328503,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004918309843767047,
      "loss": 5.5208,
      "mean_token_accuracy": 0.15936681032180786,
      "num_tokens": 20594630.0,
      "step": 11155
    },
    {
      "entropy": 5.648511266708374,
      "epoch": 0.9376181474480151,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004918229863746618,
      "loss": 5.5117,
      "mean_token_accuracy": 0.15583572238683702,
      "num_tokens": 20603653.0,
      "step": 11160
    },
    {
      "entropy": 5.783355236053467,
      "epoch": 0.93803822726318,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004918149845317114,
      "loss": 5.5747,
      "mean_token_accuracy": 0.15749593526124955,
      "num_tokens": 20612188.0,
      "step": 11165
    },
    {
      "entropy": 5.7411479473114015,
      "epoch": 0.9384583070783449,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004918069788479952,
      "loss": 5.5024,
      "mean_token_accuracy": 0.16424321234226227,
      "num_tokens": 20620933.0,
      "step": 11170
    },
    {
      "entropy": 5.719698190689087,
      "epoch": 0.9388783868935098,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004917989693236549,
      "loss": 5.5437,
      "mean_token_accuracy": 0.16058558821678162,
      "num_tokens": 20629919.0,
      "step": 11175
    },
    {
      "entropy": 5.704256439208985,
      "epoch": 0.9392984667086747,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004917909559588326,
      "loss": 5.5195,
      "mean_token_accuracy": 0.15566968470811843,
      "num_tokens": 20638475.0,
      "step": 11180
    },
    {
      "entropy": 5.814646291732788,
      "epoch": 0.9397185465238396,
      "grad_norm": 2.71875,
      "learning_rate": 0.00049178293875367,
      "loss": 5.6583,
      "mean_token_accuracy": 0.14793116524815558,
      "num_tokens": 20648105.0,
      "step": 11185
    },
    {
      "entropy": 5.706754207611084,
      "epoch": 0.9401386263390044,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004917749177083094,
      "loss": 5.5435,
      "mean_token_accuracy": 0.1529379442334175,
      "num_tokens": 20657527.0,
      "step": 11190
    },
    {
      "entropy": 5.738810300827026,
      "epoch": 0.9405587061541693,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004917668928228927,
      "loss": 5.5578,
      "mean_token_accuracy": 0.1584453448653221,
      "num_tokens": 20666375.0,
      "step": 11195
    },
    {
      "entropy": 5.731826686859131,
      "epoch": 0.9409787859693342,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004917588640975622,
      "loss": 5.4996,
      "mean_token_accuracy": 0.1631929226219654,
      "num_tokens": 20675350.0,
      "step": 11200
    },
    {
      "entropy": 5.603848695755005,
      "epoch": 0.941398865784499,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00049175083153246,
      "loss": 5.4319,
      "mean_token_accuracy": 0.15711961686611176,
      "num_tokens": 20684072.0,
      "step": 11205
    },
    {
      "entropy": 5.622369909286499,
      "epoch": 0.941818945599664,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004917427951277284,
      "loss": 5.5076,
      "mean_token_accuracy": 0.16261884421110154,
      "num_tokens": 20692989.0,
      "step": 11210
    },
    {
      "entropy": 5.732786989212036,
      "epoch": 0.9422390254148288,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004917347548835097,
      "loss": 5.4937,
      "mean_token_accuracy": 0.15980781465768815,
      "num_tokens": 20701269.0,
      "step": 11215
    },
    {
      "entropy": 5.7467756271362305,
      "epoch": 0.9426591052299937,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004917267107999466,
      "loss": 5.5982,
      "mean_token_accuracy": 0.153051495552063,
      "num_tokens": 20709739.0,
      "step": 11220
    },
    {
      "entropy": 5.66306734085083,
      "epoch": 0.9430791850451585,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004917186628771812,
      "loss": 5.536,
      "mean_token_accuracy": 0.16327179223299026,
      "num_tokens": 20718950.0,
      "step": 11225
    },
    {
      "entropy": 5.68591194152832,
      "epoch": 0.9434992648603234,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004917106111153565,
      "loss": 5.5229,
      "mean_token_accuracy": 0.1601232573390007,
      "num_tokens": 20729469.0,
      "step": 11230
    },
    {
      "entropy": 5.742890501022339,
      "epoch": 0.9439193446754883,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004917025555146148,
      "loss": 5.5378,
      "mean_token_accuracy": 0.16962186694145204,
      "num_tokens": 20738231.0,
      "step": 11235
    },
    {
      "entropy": 5.788369464874267,
      "epoch": 0.9443394244906532,
      "grad_norm": 1.71875,
      "learning_rate": 0.000491694496075099,
      "loss": 5.73,
      "mean_token_accuracy": 0.1434485659003258,
      "num_tokens": 20748578.0,
      "step": 11240
    },
    {
      "entropy": 5.8387247085571286,
      "epoch": 0.9447595043058181,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004916864327969517,
      "loss": 5.6702,
      "mean_token_accuracy": 0.14663874506950378,
      "num_tokens": 20759284.0,
      "step": 11245
    },
    {
      "entropy": 5.803573560714722,
      "epoch": 0.945179584120983,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004916783656803158,
      "loss": 5.6005,
      "mean_token_accuracy": 0.16080168783664703,
      "num_tokens": 20768186.0,
      "step": 11250
    },
    {
      "entropy": 5.677453899383545,
      "epoch": 0.9455996639361479,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004916702947253342,
      "loss": 5.4665,
      "mean_token_accuracy": 0.16224973052740096,
      "num_tokens": 20776711.0,
      "step": 11255
    },
    {
      "entropy": 5.703570175170898,
      "epoch": 0.9460197437513127,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004916622199321501,
      "loss": 5.5457,
      "mean_token_accuracy": 0.1602822169661522,
      "num_tokens": 20785154.0,
      "step": 11260
    },
    {
      "entropy": 5.761015748977661,
      "epoch": 0.9464398235664776,
      "grad_norm": 1.625,
      "learning_rate": 0.0004916541413009062,
      "loss": 5.4899,
      "mean_token_accuracy": 0.16483698636293412,
      "num_tokens": 20794114.0,
      "step": 11265
    },
    {
      "entropy": 5.797151041030884,
      "epoch": 0.9468599033816425,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004916460588317458,
      "loss": 5.5829,
      "mean_token_accuracy": 0.15465315878391267,
      "num_tokens": 20803892.0,
      "step": 11270
    },
    {
      "entropy": 5.611593103408813,
      "epoch": 0.9472799831968074,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004916379725248118,
      "loss": 5.4918,
      "mean_token_accuracy": 0.16398582309484483,
      "num_tokens": 20812892.0,
      "step": 11275
    },
    {
      "entropy": 5.729201555252075,
      "epoch": 0.9477000630119723,
      "grad_norm": 2.125,
      "learning_rate": 0.0004916298823802479,
      "loss": 5.548,
      "mean_token_accuracy": 0.1526999518275261,
      "num_tokens": 20821934.0,
      "step": 11280
    },
    {
      "entropy": 5.688755464553833,
      "epoch": 0.9481201428271372,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004916217883981971,
      "loss": 5.4658,
      "mean_token_accuracy": 0.16353494822978973,
      "num_tokens": 20830100.0,
      "step": 11285
    },
    {
      "entropy": 5.661793756484985,
      "epoch": 0.9485402226423021,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004916136905788029,
      "loss": 5.5466,
      "mean_token_accuracy": 0.15961572378873826,
      "num_tokens": 20839890.0,
      "step": 11290
    },
    {
      "entropy": 5.753994560241699,
      "epoch": 0.9489603024574669,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004916055889222087,
      "loss": 5.6685,
      "mean_token_accuracy": 0.14428120404481887,
      "num_tokens": 20848670.0,
      "step": 11295
    },
    {
      "entropy": 5.780421400070191,
      "epoch": 0.9493803822726318,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000491597483428558,
      "loss": 5.4958,
      "mean_token_accuracy": 0.170906001329422,
      "num_tokens": 20857291.0,
      "step": 11300
    },
    {
      "entropy": 5.632902812957764,
      "epoch": 0.9498004620877967,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004915893740979944,
      "loss": 5.4622,
      "mean_token_accuracy": 0.1642448052763939,
      "num_tokens": 20865341.0,
      "step": 11305
    },
    {
      "entropy": 5.743787240982056,
      "epoch": 0.9502205419029616,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004915812609306617,
      "loss": 5.6081,
      "mean_token_accuracy": 0.158392533659935,
      "num_tokens": 20875194.0,
      "step": 11310
    },
    {
      "entropy": 5.764333438873291,
      "epoch": 0.9506406217181265,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004915731439267034,
      "loss": 5.5164,
      "mean_token_accuracy": 0.15837468653917314,
      "num_tokens": 20884831.0,
      "step": 11315
    },
    {
      "entropy": 5.652561187744141,
      "epoch": 0.9510607015332914,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004915650230862634,
      "loss": 5.4019,
      "mean_token_accuracy": 0.16960276961326598,
      "num_tokens": 20893790.0,
      "step": 11320
    },
    {
      "entropy": 5.6419471263885494,
      "epoch": 0.9514807813484563,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004915568984094854,
      "loss": 5.5285,
      "mean_token_accuracy": 0.16155760288238524,
      "num_tokens": 20902175.0,
      "step": 11325
    },
    {
      "entropy": 5.830080270767212,
      "epoch": 0.951900861163621,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004915487698965136,
      "loss": 5.6617,
      "mean_token_accuracy": 0.15098460614681244,
      "num_tokens": 20911484.0,
      "step": 11330
    },
    {
      "entropy": 5.830371952056884,
      "epoch": 0.952320940978786,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004915406375474917,
      "loss": 5.6201,
      "mean_token_accuracy": 0.15232392996549607,
      "num_tokens": 20920916.0,
      "step": 11335
    },
    {
      "entropy": 5.751460123062134,
      "epoch": 0.9527410207939508,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000491532501362564,
      "loss": 5.6334,
      "mean_token_accuracy": 0.159017938375473,
      "num_tokens": 20930219.0,
      "step": 11340
    },
    {
      "entropy": 5.692816257476807,
      "epoch": 0.9531611006091157,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004915243613418745,
      "loss": 5.4478,
      "mean_token_accuracy": 0.16899611204862594,
      "num_tokens": 20939591.0,
      "step": 11345
    },
    {
      "entropy": 5.779568099975586,
      "epoch": 0.9535811804242806,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004915162174855675,
      "loss": 5.6211,
      "mean_token_accuracy": 0.1536381497979164,
      "num_tokens": 20950035.0,
      "step": 11350
    },
    {
      "entropy": 5.705796003341675,
      "epoch": 0.9540012602394455,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004915080697937872,
      "loss": 5.5265,
      "mean_token_accuracy": 0.15976119190454482,
      "num_tokens": 20959168.0,
      "step": 11355
    },
    {
      "entropy": 5.632742023468017,
      "epoch": 0.9544213400546103,
      "grad_norm": 1.875,
      "learning_rate": 0.0004914999182666779,
      "loss": 5.4551,
      "mean_token_accuracy": 0.1686936303973198,
      "num_tokens": 20967887.0,
      "step": 11360
    },
    {
      "entropy": 5.734130430221557,
      "epoch": 0.9548414198697752,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004914917629043839,
      "loss": 5.5737,
      "mean_token_accuracy": 0.1523936167359352,
      "num_tokens": 20977558.0,
      "step": 11365
    },
    {
      "entropy": 5.661308240890503,
      "epoch": 0.9552614996849401,
      "grad_norm": 1.75,
      "learning_rate": 0.00049148360370705,
      "loss": 5.521,
      "mean_token_accuracy": 0.16197700947523117,
      "num_tokens": 20986118.0,
      "step": 11370
    },
    {
      "entropy": 5.722346544265747,
      "epoch": 0.955681579500105,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004914754406748204,
      "loss": 5.4573,
      "mean_token_accuracy": 0.16239626556634904,
      "num_tokens": 20994623.0,
      "step": 11375
    },
    {
      "entropy": 5.779690885543824,
      "epoch": 0.9561016593152699,
      "grad_norm": 1.640625,
      "learning_rate": 0.00049146727380784,
      "loss": 5.637,
      "mean_token_accuracy": 0.1525883451104164,
      "num_tokens": 21004193.0,
      "step": 11380
    },
    {
      "entropy": 5.699955463409424,
      "epoch": 0.9565217391304348,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004914591031062531,
      "loss": 5.4637,
      "mean_token_accuracy": 0.16423846036195755,
      "num_tokens": 21013125.0,
      "step": 11385
    },
    {
      "entropy": 5.598379230499267,
      "epoch": 0.9569418189455997,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004914509285702048,
      "loss": 5.3926,
      "mean_token_accuracy": 0.17079196721315384,
      "num_tokens": 21021402.0,
      "step": 11390
    },
    {
      "entropy": 5.648781394958496,
      "epoch": 0.9573618987607645,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004914427501998397,
      "loss": 5.4695,
      "mean_token_accuracy": 0.16255177408456803,
      "num_tokens": 21029639.0,
      "step": 11395
    },
    {
      "entropy": 5.712826299667358,
      "epoch": 0.9577819785759294,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004914345679953027,
      "loss": 5.5082,
      "mean_token_accuracy": 0.1599772408604622,
      "num_tokens": 21037525.0,
      "step": 11400
    },
    {
      "entropy": 5.778018856048584,
      "epoch": 0.9582020583910943,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004914263819567388,
      "loss": 5.6048,
      "mean_token_accuracy": 0.14846348613500596,
      "num_tokens": 21047702.0,
      "step": 11405
    },
    {
      "entropy": 5.744657373428344,
      "epoch": 0.9586221382062592,
      "grad_norm": 1.4375,
      "learning_rate": 0.000491418192084293,
      "loss": 5.5155,
      "mean_token_accuracy": 0.16157382726669312,
      "num_tokens": 21056379.0,
      "step": 11410
    },
    {
      "entropy": 5.676009654998779,
      "epoch": 0.9590422180214241,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004914099983781104,
      "loss": 5.5156,
      "mean_token_accuracy": 0.16159558892250062,
      "num_tokens": 21065283.0,
      "step": 11415
    },
    {
      "entropy": 5.707050800323486,
      "epoch": 0.959462297836589,
      "grad_norm": 1.640625,
      "learning_rate": 0.000491401800838336,
      "loss": 5.6565,
      "mean_token_accuracy": 0.15205793231725692,
      "num_tokens": 21074938.0,
      "step": 11420
    },
    {
      "entropy": 5.712154531478882,
      "epoch": 0.9598823776517539,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004913935994651153,
      "loss": 5.4786,
      "mean_token_accuracy": 0.16634972542524337,
      "num_tokens": 21084729.0,
      "step": 11425
    },
    {
      "entropy": 5.6186995029449465,
      "epoch": 0.9603024574669187,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004913853942585932,
      "loss": 5.3877,
      "mean_token_accuracy": 0.16797681003808976,
      "num_tokens": 21093456.0,
      "step": 11430
    },
    {
      "entropy": 5.6789297580719,
      "epoch": 0.9607225372820836,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004913771852189155,
      "loss": 5.5291,
      "mean_token_accuracy": 0.16209679543972016,
      "num_tokens": 21102980.0,
      "step": 11435
    },
    {
      "entropy": 5.837604236602783,
      "epoch": 0.9611426170972485,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004913689723462271,
      "loss": 5.7492,
      "mean_token_accuracy": 0.16761390566825868,
      "num_tokens": 21112777.0,
      "step": 11440
    },
    {
      "entropy": 5.760186672210693,
      "epoch": 0.9615626969124134,
      "grad_norm": 1.765625,
      "learning_rate": 0.000491360755640674,
      "loss": 5.634,
      "mean_token_accuracy": 0.1518979102373123,
      "num_tokens": 21122139.0,
      "step": 11445
    },
    {
      "entropy": 5.685001182556152,
      "epoch": 0.9619827767275783,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004913525351024014,
      "loss": 5.5012,
      "mean_token_accuracy": 0.15827474147081375,
      "num_tokens": 21131425.0,
      "step": 11450
    },
    {
      "entropy": 5.645795059204102,
      "epoch": 0.9624028565427432,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004913443107315552,
      "loss": 5.4816,
      "mean_token_accuracy": 0.15899788290262223,
      "num_tokens": 21140784.0,
      "step": 11455
    },
    {
      "entropy": 5.737687778472901,
      "epoch": 0.962822936357908,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004913360825282807,
      "loss": 5.4913,
      "mean_token_accuracy": 0.16493094712495804,
      "num_tokens": 21150408.0,
      "step": 11460
    },
    {
      "entropy": 5.70665397644043,
      "epoch": 0.9632430161730728,
      "grad_norm": 1.8671875,
      "learning_rate": 0.000491327850492724,
      "loss": 5.5903,
      "mean_token_accuracy": 0.16218522936105728,
      "num_tokens": 21158915.0,
      "step": 11465
    },
    {
      "entropy": 5.5823643684387205,
      "epoch": 0.9636630959882377,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004913196146250309,
      "loss": 5.3863,
      "mean_token_accuracy": 0.16955134347081185,
      "num_tokens": 21167336.0,
      "step": 11470
    },
    {
      "entropy": 5.790083742141723,
      "epoch": 0.9640831758034026,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004913113749253472,
      "loss": 5.7179,
      "mean_token_accuracy": 0.15193437561392784,
      "num_tokens": 21177499.0,
      "step": 11475
    },
    {
      "entropy": 5.817345762252808,
      "epoch": 0.9645032556185675,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004913031313938188,
      "loss": 5.6051,
      "mean_token_accuracy": 0.15267798900604249,
      "num_tokens": 21186961.0,
      "step": 11480
    },
    {
      "entropy": 5.753744745254517,
      "epoch": 0.9649233354337324,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004912948840305919,
      "loss": 5.5022,
      "mean_token_accuracy": 0.1661657139658928,
      "num_tokens": 21196364.0,
      "step": 11485
    },
    {
      "entropy": 5.727255821228027,
      "epoch": 0.9653434152488973,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004912866328358125,
      "loss": 5.5777,
      "mean_token_accuracy": 0.15882964730262755,
      "num_tokens": 21206376.0,
      "step": 11490
    },
    {
      "entropy": 5.742443180084228,
      "epoch": 0.9657634950640621,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004912783778096266,
      "loss": 5.5585,
      "mean_token_accuracy": 0.16211859583854676,
      "num_tokens": 21215889.0,
      "step": 11495
    },
    {
      "entropy": 5.764708566665649,
      "epoch": 0.966183574879227,
      "grad_norm": 1.625,
      "learning_rate": 0.0004912701189521808,
      "loss": 5.5637,
      "mean_token_accuracy": 0.1641128808259964,
      "num_tokens": 21224959.0,
      "step": 11500
    },
    {
      "entropy": 5.76860499382019,
      "epoch": 0.9666036546943919,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004912618562636211,
      "loss": 5.6748,
      "mean_token_accuracy": 0.150497405230999,
      "num_tokens": 21234495.0,
      "step": 11505
    },
    {
      "entropy": 5.671495676040649,
      "epoch": 0.9670237345095568,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000491253589744094,
      "loss": 5.5133,
      "mean_token_accuracy": 0.1606043353676796,
      "num_tokens": 21244555.0,
      "step": 11510
    },
    {
      "entropy": 5.798943901062012,
      "epoch": 0.9674438143247217,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004912453193937459,
      "loss": 5.6487,
      "mean_token_accuracy": 0.15496133714914323,
      "num_tokens": 21254199.0,
      "step": 11515
    },
    {
      "entropy": 5.775474929809571,
      "epoch": 0.9678638941398866,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004912370452127234,
      "loss": 5.5574,
      "mean_token_accuracy": 0.15863036513328552,
      "num_tokens": 21262723.0,
      "step": 11520
    },
    {
      "entropy": 5.699438953399659,
      "epoch": 0.9682839739550515,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004912287672011728,
      "loss": 5.4811,
      "mean_token_accuracy": 0.16491955220699311,
      "num_tokens": 21271283.0,
      "step": 11525
    },
    {
      "entropy": 5.647049522399902,
      "epoch": 0.9687040537702163,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004912204853592411,
      "loss": 5.5223,
      "mean_token_accuracy": 0.1697501629590988,
      "num_tokens": 21279542.0,
      "step": 11530
    },
    {
      "entropy": 5.697355556488037,
      "epoch": 0.9691241335853812,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004912121996870748,
      "loss": 5.5099,
      "mean_token_accuracy": 0.1619595393538475,
      "num_tokens": 21288678.0,
      "step": 11535
    },
    {
      "entropy": 5.812397384643555,
      "epoch": 0.9695442134005461,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004912039101848207,
      "loss": 5.6565,
      "mean_token_accuracy": 0.15555005446076392,
      "num_tokens": 21298982.0,
      "step": 11540
    },
    {
      "entropy": 5.709640407562256,
      "epoch": 0.969964293215711,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004911956168526257,
      "loss": 5.5863,
      "mean_token_accuracy": 0.15883329659700393,
      "num_tokens": 21307663.0,
      "step": 11545
    },
    {
      "entropy": 5.733417558670044,
      "epoch": 0.9703843730308759,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004911873196906366,
      "loss": 5.5912,
      "mean_token_accuracy": 0.15337843149900438,
      "num_tokens": 21318004.0,
      "step": 11550
    },
    {
      "entropy": 5.624382829666137,
      "epoch": 0.9708044528460408,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004911790186990005,
      "loss": 5.4055,
      "mean_token_accuracy": 0.17055178582668304,
      "num_tokens": 21327373.0,
      "step": 11555
    },
    {
      "entropy": 5.6697979927062985,
      "epoch": 0.9712245326612057,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004911707138778643,
      "loss": 5.5004,
      "mean_token_accuracy": 0.15926228910684587,
      "num_tokens": 21335654.0,
      "step": 11560
    },
    {
      "entropy": 5.797804164886474,
      "epoch": 0.9716446124763705,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004911624052273754,
      "loss": 5.5656,
      "mean_token_accuracy": 0.15943690538406372,
      "num_tokens": 21344464.0,
      "step": 11565
    },
    {
      "entropy": 5.783263635635376,
      "epoch": 0.9720646922915354,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004911540927476807,
      "loss": 5.6574,
      "mean_token_accuracy": 0.1580268144607544,
      "num_tokens": 21354121.0,
      "step": 11570
    },
    {
      "entropy": 5.765107583999634,
      "epoch": 0.9724847721067003,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004911457764389275,
      "loss": 5.5917,
      "mean_token_accuracy": 0.16147245317697526,
      "num_tokens": 21363395.0,
      "step": 11575
    },
    {
      "entropy": 5.731990480422974,
      "epoch": 0.9729048519218652,
      "grad_norm": 2.0,
      "learning_rate": 0.0004911374563012633,
      "loss": 5.5543,
      "mean_token_accuracy": 0.15289353728294372,
      "num_tokens": 21372126.0,
      "step": 11580
    },
    {
      "entropy": 5.778444862365722,
      "epoch": 0.97332493173703,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004911291323348352,
      "loss": 5.6253,
      "mean_token_accuracy": 0.157344251871109,
      "num_tokens": 21380554.0,
      "step": 11585
    },
    {
      "entropy": 5.706854677200317,
      "epoch": 0.973745011552195,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004911208045397909,
      "loss": 5.514,
      "mean_token_accuracy": 0.15825875103473663,
      "num_tokens": 21389317.0,
      "step": 11590
    },
    {
      "entropy": 5.79919662475586,
      "epoch": 0.9741650913673598,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004911124729162778,
      "loss": 5.6453,
      "mean_token_accuracy": 0.1541647955775261,
      "num_tokens": 21398926.0,
      "step": 11595
    },
    {
      "entropy": 5.782274341583252,
      "epoch": 0.9745851711825246,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004911041374644435,
      "loss": 5.4558,
      "mean_token_accuracy": 0.16338740587234496,
      "num_tokens": 21406962.0,
      "step": 11600
    },
    {
      "entropy": 5.757547998428345,
      "epoch": 0.9750052509976895,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004910957981844357,
      "loss": 5.5533,
      "mean_token_accuracy": 0.16179322302341462,
      "num_tokens": 21415868.0,
      "step": 11605
    },
    {
      "entropy": 5.77811369895935,
      "epoch": 0.9754253308128544,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004910874550764022,
      "loss": 5.6645,
      "mean_token_accuracy": 0.15746580213308334,
      "num_tokens": 21424544.0,
      "step": 11610
    },
    {
      "entropy": 5.659853553771972,
      "epoch": 0.9758454106280193,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004910791081404907,
      "loss": 5.5353,
      "mean_token_accuracy": 0.16283178329467773,
      "num_tokens": 21433589.0,
      "step": 11615
    },
    {
      "entropy": 5.700931072235107,
      "epoch": 0.9762654904431842,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004910707573768489,
      "loss": 5.5844,
      "mean_token_accuracy": 0.15395784080028535,
      "num_tokens": 21442084.0,
      "step": 11620
    },
    {
      "entropy": 5.728076314926147,
      "epoch": 0.9766855702583491,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004910624027856251,
      "loss": 5.5069,
      "mean_token_accuracy": 0.16065045148134233,
      "num_tokens": 21450962.0,
      "step": 11625
    },
    {
      "entropy": 5.780987930297852,
      "epoch": 0.977105650073514,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004910540443669669,
      "loss": 5.6095,
      "mean_token_accuracy": 0.1500787116587162,
      "num_tokens": 21461322.0,
      "step": 11630
    },
    {
      "entropy": 5.753719711303711,
      "epoch": 0.9775257298886788,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004910456821210227,
      "loss": 5.5737,
      "mean_token_accuracy": 0.1601569026708603,
      "num_tokens": 21470800.0,
      "step": 11635
    },
    {
      "entropy": 5.7415358543396,
      "epoch": 0.9779458097038437,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004910373160479404,
      "loss": 5.4411,
      "mean_token_accuracy": 0.1655646875500679,
      "num_tokens": 21479707.0,
      "step": 11640
    },
    {
      "entropy": 5.748549699783325,
      "epoch": 0.9783658895190086,
      "grad_norm": 2.25,
      "learning_rate": 0.0004910289461478683,
      "loss": 5.6165,
      "mean_token_accuracy": 0.15316568911075593,
      "num_tokens": 21489469.0,
      "step": 11645
    },
    {
      "entropy": 5.751054954528809,
      "epoch": 0.9787859693341735,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004910205724209547,
      "loss": 5.5889,
      "mean_token_accuracy": 0.15403057783842086,
      "num_tokens": 21499226.0,
      "step": 11650
    },
    {
      "entropy": 5.619789409637451,
      "epoch": 0.9792060491493384,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004910121948673478,
      "loss": 5.4392,
      "mean_token_accuracy": 0.1629665359854698,
      "num_tokens": 21508129.0,
      "step": 11655
    },
    {
      "entropy": 5.699333000183105,
      "epoch": 0.9796261289645033,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004910038134871962,
      "loss": 5.4801,
      "mean_token_accuracy": 0.1643253058195114,
      "num_tokens": 21516293.0,
      "step": 11660
    },
    {
      "entropy": 5.8036095142364506,
      "epoch": 0.9800462087796681,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004909954282806482,
      "loss": 5.6387,
      "mean_token_accuracy": 0.16016865894198418,
      "num_tokens": 21525393.0,
      "step": 11665
    },
    {
      "entropy": 5.6630861282348635,
      "epoch": 0.980466288594833,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004909870392478524,
      "loss": 5.5018,
      "mean_token_accuracy": 0.16075061559677123,
      "num_tokens": 21534585.0,
      "step": 11670
    },
    {
      "entropy": 5.669094085693359,
      "epoch": 0.9808863684099979,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004909786463889575,
      "loss": 5.4321,
      "mean_token_accuracy": 0.1637320727109909,
      "num_tokens": 21542947.0,
      "step": 11675
    },
    {
      "entropy": 5.70667781829834,
      "epoch": 0.9813064482251628,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004909702497041121,
      "loss": 5.5403,
      "mean_token_accuracy": 0.16008226573467255,
      "num_tokens": 21552168.0,
      "step": 11680
    },
    {
      "entropy": 5.744578647613525,
      "epoch": 0.9817265280403277,
      "grad_norm": 2.125,
      "learning_rate": 0.0004909618491934648,
      "loss": 5.5414,
      "mean_token_accuracy": 0.16264388114213943,
      "num_tokens": 21562131.0,
      "step": 11685
    },
    {
      "entropy": 5.708233642578125,
      "epoch": 0.9821466078554926,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004909534448571647,
      "loss": 5.5015,
      "mean_token_accuracy": 0.16573767215013505,
      "num_tokens": 21571363.0,
      "step": 11690
    },
    {
      "entropy": 5.73539662361145,
      "epoch": 0.9825666876706575,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004909450366953604,
      "loss": 5.4867,
      "mean_token_accuracy": 0.16546985059976577,
      "num_tokens": 21580754.0,
      "step": 11695
    },
    {
      "entropy": 5.716777801513672,
      "epoch": 0.9829867674858223,
      "grad_norm": 1.625,
      "learning_rate": 0.000490936624708201,
      "loss": 5.5823,
      "mean_token_accuracy": 0.16040044873952866,
      "num_tokens": 21590053.0,
      "step": 11700
    },
    {
      "entropy": 5.717460250854492,
      "epoch": 0.9834068473009872,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004909282088958356,
      "loss": 5.5319,
      "mean_token_accuracy": 0.15609903037548065,
      "num_tokens": 21598681.0,
      "step": 11705
    },
    {
      "entropy": 5.778473806381226,
      "epoch": 0.983826927116152,
      "grad_norm": 2.625,
      "learning_rate": 0.000490919789258413,
      "loss": 5.5699,
      "mean_token_accuracy": 0.16667143106460572,
      "num_tokens": 21607465.0,
      "step": 11710
    },
    {
      "entropy": 5.76135516166687,
      "epoch": 0.984247006931317,
      "grad_norm": 2.125,
      "learning_rate": 0.0004909113657960826,
      "loss": 5.6466,
      "mean_token_accuracy": 0.1501710444688797,
      "num_tokens": 21617480.0,
      "step": 11715
    },
    {
      "entropy": 5.732358407974243,
      "epoch": 0.9846670867464818,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004909029385089935,
      "loss": 5.56,
      "mean_token_accuracy": 0.16049663126468658,
      "num_tokens": 21626434.0,
      "step": 11720
    },
    {
      "entropy": 5.741581773757934,
      "epoch": 0.9850871665616467,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000490894507397295,
      "loss": 5.5349,
      "mean_token_accuracy": 0.16318391263484955,
      "num_tokens": 21635627.0,
      "step": 11725
    },
    {
      "entropy": 5.715289211273193,
      "epoch": 0.9855072463768116,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004908860724611365,
      "loss": 5.538,
      "mean_token_accuracy": 0.1623338758945465,
      "num_tokens": 21644789.0,
      "step": 11730
    },
    {
      "entropy": 5.649227333068848,
      "epoch": 0.9859273261919764,
      "grad_norm": 2.125,
      "learning_rate": 0.0004908776337006675,
      "loss": 5.5411,
      "mean_token_accuracy": 0.15639186650514603,
      "num_tokens": 21653696.0,
      "step": 11735
    },
    {
      "entropy": 5.758316421508789,
      "epoch": 0.9863474060071413,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004908691911160373,
      "loss": 5.5441,
      "mean_token_accuracy": 0.14904894530773163,
      "num_tokens": 21664420.0,
      "step": 11740
    },
    {
      "entropy": 5.777101802825928,
      "epoch": 0.9867674858223062,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004908607447073954,
      "loss": 5.532,
      "mean_token_accuracy": 0.16310076788067818,
      "num_tokens": 21673716.0,
      "step": 11745
    },
    {
      "entropy": 5.710586738586426,
      "epoch": 0.9871875656374711,
      "grad_norm": 1.625,
      "learning_rate": 0.0004908522944748917,
      "loss": 5.5124,
      "mean_token_accuracy": 0.16769290566444398,
      "num_tokens": 21682860.0,
      "step": 11750
    },
    {
      "entropy": 5.605210161209106,
      "epoch": 0.987607645452636,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004908438404186758,
      "loss": 5.5075,
      "mean_token_accuracy": 0.17046434432268143,
      "num_tokens": 21691915.0,
      "step": 11755
    },
    {
      "entropy": 5.718367576599121,
      "epoch": 0.9880277252678009,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004908353825388973,
      "loss": 5.6478,
      "mean_token_accuracy": 0.1539776936173439,
      "num_tokens": 21701666.0,
      "step": 11760
    },
    {
      "entropy": 5.8222753524780275,
      "epoch": 0.9884478050829658,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004908269208357062,
      "loss": 5.5593,
      "mean_token_accuracy": 0.1612561821937561,
      "num_tokens": 21709267.0,
      "step": 11765
    },
    {
      "entropy": 5.731483888626099,
      "epoch": 0.9888678848981306,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004908184553092523,
      "loss": 5.4654,
      "mean_token_accuracy": 0.15959916561841964,
      "num_tokens": 21718117.0,
      "step": 11770
    },
    {
      "entropy": 5.744333124160766,
      "epoch": 0.9892879647132955,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004908099859596856,
      "loss": 5.5856,
      "mean_token_accuracy": 0.16114241033792495,
      "num_tokens": 21727952.0,
      "step": 11775
    },
    {
      "entropy": 5.728325080871582,
      "epoch": 0.9897080445284604,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004908015127871561,
      "loss": 5.4976,
      "mean_token_accuracy": 0.15698132812976837,
      "num_tokens": 21737878.0,
      "step": 11780
    },
    {
      "entropy": 5.6280214309692385,
      "epoch": 0.9901281243436253,
      "grad_norm": 2.734375,
      "learning_rate": 0.000490793035791814,
      "loss": 5.4182,
      "mean_token_accuracy": 0.16316151171922683,
      "num_tokens": 21747391.0,
      "step": 11785
    },
    {
      "entropy": 5.682892322540283,
      "epoch": 0.9905482041587902,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004907845549738093,
      "loss": 5.4602,
      "mean_token_accuracy": 0.1639404833316803,
      "num_tokens": 21756791.0,
      "step": 11790
    },
    {
      "entropy": 5.634412670135498,
      "epoch": 0.9909682839739551,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004907760703332923,
      "loss": 5.5007,
      "mean_token_accuracy": 0.162163844704628,
      "num_tokens": 21766020.0,
      "step": 11795
    },
    {
      "entropy": 5.730725479125977,
      "epoch": 0.99138836378912,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004907675818704134,
      "loss": 5.6054,
      "mean_token_accuracy": 0.15719361007213592,
      "num_tokens": 21775895.0,
      "step": 11800
    },
    {
      "entropy": 5.708724880218506,
      "epoch": 0.9918084436042848,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004907590895853228,
      "loss": 5.5163,
      "mean_token_accuracy": 0.16296806633472444,
      "num_tokens": 21784543.0,
      "step": 11805
    },
    {
      "entropy": 5.711108875274658,
      "epoch": 0.9922285234194497,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004907505934781712,
      "loss": 5.5718,
      "mean_token_accuracy": 0.1565998300909996,
      "num_tokens": 21793938.0,
      "step": 11810
    },
    {
      "entropy": 5.700426530838013,
      "epoch": 0.9926486032346146,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004907420935491087,
      "loss": 5.5456,
      "mean_token_accuracy": 0.15969295799732208,
      "num_tokens": 21803641.0,
      "step": 11815
    },
    {
      "entropy": 5.692903614044189,
      "epoch": 0.9930686830497795,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004907335897982862,
      "loss": 5.4649,
      "mean_token_accuracy": 0.1696073144674301,
      "num_tokens": 21812542.0,
      "step": 11820
    },
    {
      "entropy": 5.636066865921021,
      "epoch": 0.9934887628649444,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004907250822258543,
      "loss": 5.5573,
      "mean_token_accuracy": 0.157867032289505,
      "num_tokens": 21821847.0,
      "step": 11825
    },
    {
      "entropy": 5.794892740249634,
      "epoch": 0.9939088426801093,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004907165708319637,
      "loss": 5.5963,
      "mean_token_accuracy": 0.15924146473407746,
      "num_tokens": 21830799.0,
      "step": 11830
    },
    {
      "entropy": 5.723268651962281,
      "epoch": 0.994328922495274,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004907080556167651,
      "loss": 5.5417,
      "mean_token_accuracy": 0.16239422112703322,
      "num_tokens": 21840202.0,
      "step": 11835
    },
    {
      "entropy": 5.8009960651397705,
      "epoch": 0.994749002310439,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004906995365804093,
      "loss": 5.6415,
      "mean_token_accuracy": 0.15165836364030838,
      "num_tokens": 21849701.0,
      "step": 11840
    },
    {
      "entropy": 5.7475098133087155,
      "epoch": 0.9951690821256038,
      "grad_norm": 3.125,
      "learning_rate": 0.0004906910137230472,
      "loss": 5.5083,
      "mean_token_accuracy": 0.16084536910057068,
      "num_tokens": 21859191.0,
      "step": 11845
    },
    {
      "entropy": 5.686148691177368,
      "epoch": 0.9955891619407687,
      "grad_norm": 3.546875,
      "learning_rate": 0.00049068248704483,
      "loss": 5.5053,
      "mean_token_accuracy": 0.16064416766166686,
      "num_tokens": 21867944.0,
      "step": 11850
    },
    {
      "entropy": 5.62108964920044,
      "epoch": 0.9960092417559336,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004906739565459085,
      "loss": 5.5337,
      "mean_token_accuracy": 0.15536581873893737,
      "num_tokens": 21876368.0,
      "step": 11855
    },
    {
      "entropy": 5.824420833587647,
      "epoch": 0.9964293215710985,
      "grad_norm": 1.625,
      "learning_rate": 0.000490665422226434,
      "loss": 5.6125,
      "mean_token_accuracy": 0.15335456728935243,
      "num_tokens": 21885634.0,
      "step": 11860
    },
    {
      "entropy": 5.684914779663086,
      "epoch": 0.9968494013862634,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004906568840865576,
      "loss": 5.4319,
      "mean_token_accuracy": 0.16591923534870148,
      "num_tokens": 21894315.0,
      "step": 11865
    },
    {
      "entropy": 5.64821228981018,
      "epoch": 0.9972694812014282,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004906483421264305,
      "loss": 5.5444,
      "mean_token_accuracy": 0.16124298125505448,
      "num_tokens": 21903342.0,
      "step": 11870
    },
    {
      "entropy": 5.722136449813843,
      "epoch": 0.9976895610165931,
      "grad_norm": 2.0,
      "learning_rate": 0.000490639796346204,
      "loss": 5.6689,
      "mean_token_accuracy": 0.15262537002563475,
      "num_tokens": 21914158.0,
      "step": 11875
    },
    {
      "entropy": 5.826263189315796,
      "epoch": 0.998109640831758,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004906312467460297,
      "loss": 5.5359,
      "mean_token_accuracy": 0.16018486469984056,
      "num_tokens": 21922639.0,
      "step": 11880
    },
    {
      "entropy": 5.733368921279907,
      "epoch": 0.9985297206469229,
      "grad_norm": 1.875,
      "learning_rate": 0.0004906226933260588,
      "loss": 5.5305,
      "mean_token_accuracy": 0.1628845065832138,
      "num_tokens": 21931385.0,
      "step": 11885
    },
    {
      "entropy": 5.763300991058349,
      "epoch": 0.9989498004620878,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004906141360864429,
      "loss": 5.5486,
      "mean_token_accuracy": 0.1582605332136154,
      "num_tokens": 21940788.0,
      "step": 11890
    },
    {
      "entropy": 5.750997114181518,
      "epoch": 0.9993698802772527,
      "grad_norm": 2.125,
      "learning_rate": 0.0004906055750273336,
      "loss": 5.5633,
      "mean_token_accuracy": 0.15954309850931167,
      "num_tokens": 21950309.0,
      "step": 11895
    },
    {
      "entropy": 5.696226406097412,
      "epoch": 0.9997899600924176,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004905970101488826,
      "loss": 5.5533,
      "mean_token_accuracy": 0.16005461364984513,
      "num_tokens": 21959141.0,
      "step": 11900
    },
    {
      "entropy": 5.797101603613959,
      "epoch": 1.000168031926066,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004905884414512416,
      "loss": 5.5832,
      "mean_token_accuracy": 0.16606505546304914,
      "num_tokens": 21966665.0,
      "step": 11905
    },
    {
      "entropy": 5.731563329696655,
      "epoch": 1.0005881117412307,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004905798689345623,
      "loss": 5.5626,
      "mean_token_accuracy": 0.16171082034707068,
      "num_tokens": 21976728.0,
      "step": 11910
    },
    {
      "entropy": 5.629949522018433,
      "epoch": 1.0010081915563958,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004905712925989968,
      "loss": 5.3915,
      "mean_token_accuracy": 0.1608661249279976,
      "num_tokens": 21985915.0,
      "step": 11915
    },
    {
      "entropy": 5.7110429286956785,
      "epoch": 1.0014282713715605,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004905627124446967,
      "loss": 5.4499,
      "mean_token_accuracy": 0.1611791133880615,
      "num_tokens": 21995826.0,
      "step": 11920
    },
    {
      "entropy": 5.700401496887207,
      "epoch": 1.0018483511867255,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004905541284718142,
      "loss": 5.418,
      "mean_token_accuracy": 0.16362681239843369,
      "num_tokens": 22005299.0,
      "step": 11925
    },
    {
      "entropy": 5.667144775390625,
      "epoch": 1.0022684310018903,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004905455406805011,
      "loss": 5.4529,
      "mean_token_accuracy": 0.16137843877077102,
      "num_tokens": 22014499.0,
      "step": 11930
    },
    {
      "entropy": 5.7403564453125,
      "epoch": 1.0026885108170553,
      "grad_norm": 2.265625,
      "learning_rate": 0.00049053694907091,
      "loss": 5.6015,
      "mean_token_accuracy": 0.15019496381282807,
      "num_tokens": 22024531.0,
      "step": 11935
    },
    {
      "entropy": 5.714895439147949,
      "epoch": 1.0031085906322201,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004905283536431928,
      "loss": 5.5081,
      "mean_token_accuracy": 0.16284531652927398,
      "num_tokens": 22034036.0,
      "step": 11940
    },
    {
      "entropy": 5.638365077972412,
      "epoch": 1.003528670447385,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004905197543975017,
      "loss": 5.4115,
      "mean_token_accuracy": 0.1633414715528488,
      "num_tokens": 22042910.0,
      "step": 11945
    },
    {
      "entropy": 5.738863801956176,
      "epoch": 1.00394875026255,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004905111513339892,
      "loss": 5.5209,
      "mean_token_accuracy": 0.15980954617261886,
      "num_tokens": 22052242.0,
      "step": 11950
    },
    {
      "entropy": 5.7258447170257565,
      "epoch": 1.0043688300777147,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004905025444528076,
      "loss": 5.4832,
      "mean_token_accuracy": 0.15796915143728257,
      "num_tokens": 22061467.0,
      "step": 11955
    },
    {
      "entropy": 5.585642290115357,
      "epoch": 1.0047889098928797,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004904939337541093,
      "loss": 5.3351,
      "mean_token_accuracy": 0.16610788106918334,
      "num_tokens": 22070300.0,
      "step": 11960
    },
    {
      "entropy": 5.718946933746338,
      "epoch": 1.0052089897080445,
      "grad_norm": 2.125,
      "learning_rate": 0.0004904853192380472,
      "loss": 5.4909,
      "mean_token_accuracy": 0.16344053149223328,
      "num_tokens": 22078960.0,
      "step": 11965
    },
    {
      "entropy": 5.742123985290528,
      "epoch": 1.0056290695232095,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004904767009047733,
      "loss": 5.4348,
      "mean_token_accuracy": 0.1622706487774849,
      "num_tokens": 22088135.0,
      "step": 11970
    },
    {
      "entropy": 5.737835264205932,
      "epoch": 1.0060491493383743,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004904680787544408,
      "loss": 5.5541,
      "mean_token_accuracy": 0.15428006947040557,
      "num_tokens": 22098004.0,
      "step": 11975
    },
    {
      "entropy": 5.801382684707642,
      "epoch": 1.006469229153539,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004904594527872022,
      "loss": 5.5298,
      "mean_token_accuracy": 0.15679605156183243,
      "num_tokens": 22107680.0,
      "step": 11980
    },
    {
      "entropy": 5.745111274719238,
      "epoch": 1.006889308968704,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004904508230032103,
      "loss": 5.5311,
      "mean_token_accuracy": 0.16063058525323867,
      "num_tokens": 22118004.0,
      "step": 11985
    },
    {
      "entropy": 5.689800071716308,
      "epoch": 1.0073093887838689,
      "grad_norm": 2.125,
      "learning_rate": 0.000490442189402618,
      "loss": 5.4625,
      "mean_token_accuracy": 0.1709965944290161,
      "num_tokens": 22127825.0,
      "step": 11990
    },
    {
      "entropy": 5.684913110733032,
      "epoch": 1.007729468599034,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004904335519855783,
      "loss": 5.3906,
      "mean_token_accuracy": 0.1667776048183441,
      "num_tokens": 22136448.0,
      "step": 11995
    },
    {
      "entropy": 5.670077896118164,
      "epoch": 1.0081495484141987,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004904249107522442,
      "loss": 5.5095,
      "mean_token_accuracy": 0.15952253490686416,
      "num_tokens": 22146415.0,
      "step": 12000
    },
    {
      "epoch": 1.0081495484141987,
      "eval_entropy": 5.499641510231085,
      "eval_loss": 5.561922073364258,
      "eval_mean_token_accuracy": 0.16659502685389052,
      "eval_num_tokens": 22146415.0,
      "eval_runtime": 27.1007,
      "eval_samples_per_second": 1378.784,
      "eval_steps_per_second": 172.357,
      "step": 12000
    },
    {
      "entropy": 5.780040836334228,
      "epoch": 1.0085696282293637,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004904162657027685,
      "loss": 5.6277,
      "mean_token_accuracy": 0.16104299575090408,
      "num_tokens": 22156327.0,
      "step": 12005
    },
    {
      "entropy": 5.666675329208374,
      "epoch": 1.0089897080445285,
      "grad_norm": 1.5,
      "learning_rate": 0.0004904076168373049,
      "loss": 5.4484,
      "mean_token_accuracy": 0.1636186048388481,
      "num_tokens": 22165677.0,
      "step": 12010
    },
    {
      "entropy": 5.724397230148315,
      "epoch": 1.0094097878596933,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004903989641560061,
      "loss": 5.5786,
      "mean_token_accuracy": 0.16150926798582077,
      "num_tokens": 22175232.0,
      "step": 12015
    },
    {
      "entropy": 5.721542501449585,
      "epoch": 1.0098298676748583,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004903903076590256,
      "loss": 5.461,
      "mean_token_accuracy": 0.15944595038890838,
      "num_tokens": 22184026.0,
      "step": 12020
    },
    {
      "entropy": 5.613580513000488,
      "epoch": 1.010249947490023,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004903816473465167,
      "loss": 5.3558,
      "mean_token_accuracy": 0.1735146701335907,
      "num_tokens": 22192020.0,
      "step": 12025
    },
    {
      "entropy": 5.620361232757569,
      "epoch": 1.010670027305188,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004903729832186328,
      "loss": 5.3474,
      "mean_token_accuracy": 0.17106474488973616,
      "num_tokens": 22200060.0,
      "step": 12030
    },
    {
      "entropy": 5.619206380844116,
      "epoch": 1.0110901071203529,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004903643152755274,
      "loss": 5.3797,
      "mean_token_accuracy": 0.16311112642288209,
      "num_tokens": 22208625.0,
      "step": 12035
    },
    {
      "entropy": 5.636928939819336,
      "epoch": 1.0115101869355176,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004903556435173541,
      "loss": 5.3741,
      "mean_token_accuracy": 0.1699930876493454,
      "num_tokens": 22217781.0,
      "step": 12040
    },
    {
      "entropy": 5.725437021255493,
      "epoch": 1.0119302667506826,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004903469679442665,
      "loss": 5.5068,
      "mean_token_accuracy": 0.1630315825343132,
      "num_tokens": 22226432.0,
      "step": 12045
    },
    {
      "entropy": 5.628535366058349,
      "epoch": 1.0123503465658474,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004903382885564181,
      "loss": 5.4978,
      "mean_token_accuracy": 0.1623180940747261,
      "num_tokens": 22234811.0,
      "step": 12050
    },
    {
      "entropy": 5.606299209594726,
      "epoch": 1.0127704263810124,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000490329605353963,
      "loss": 5.3746,
      "mean_token_accuracy": 0.1714473158121109,
      "num_tokens": 22242808.0,
      "step": 12055
    },
    {
      "entropy": 5.720843267440796,
      "epoch": 1.0131905061961772,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004903209183370547,
      "loss": 5.4519,
      "mean_token_accuracy": 0.16782257854938507,
      "num_tokens": 22251371.0,
      "step": 12060
    },
    {
      "entropy": 5.802830362319947,
      "epoch": 1.0136105860113422,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004903122275058472,
      "loss": 5.5244,
      "mean_token_accuracy": 0.16609688699245453,
      "num_tokens": 22260868.0,
      "step": 12065
    },
    {
      "entropy": 5.656514883041382,
      "epoch": 1.014030665826507,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004903035328604944,
      "loss": 5.4333,
      "mean_token_accuracy": 0.16029565781354904,
      "num_tokens": 22270554.0,
      "step": 12070
    },
    {
      "entropy": 5.644852542877198,
      "epoch": 1.0144507456416718,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004902948344011506,
      "loss": 5.4394,
      "mean_token_accuracy": 0.1594587244093418,
      "num_tokens": 22279170.0,
      "step": 12075
    },
    {
      "entropy": 5.762671756744385,
      "epoch": 1.0148708254568368,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004902861321279694,
      "loss": 5.5814,
      "mean_token_accuracy": 0.15654578879475595,
      "num_tokens": 22288788.0,
      "step": 12080
    },
    {
      "entropy": 5.685765647888184,
      "epoch": 1.0152909052720016,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004902774260411055,
      "loss": 5.3707,
      "mean_token_accuracy": 0.1631255716085434,
      "num_tokens": 22297501.0,
      "step": 12085
    },
    {
      "entropy": 5.645720195770264,
      "epoch": 1.0157109850871666,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004902687161407126,
      "loss": 5.3224,
      "mean_token_accuracy": 0.17397422790527345,
      "num_tokens": 22306181.0,
      "step": 12090
    },
    {
      "entropy": 5.625262689590454,
      "epoch": 1.0161310649023314,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004902600024269454,
      "loss": 5.4623,
      "mean_token_accuracy": 0.16882133185863496,
      "num_tokens": 22315762.0,
      "step": 12095
    },
    {
      "entropy": 5.65147967338562,
      "epoch": 1.0165511447174964,
      "grad_norm": 1.921875,
      "learning_rate": 0.000490251284899958,
      "loss": 5.4041,
      "mean_token_accuracy": 0.17007135301828386,
      "num_tokens": 22325127.0,
      "step": 12100
    },
    {
      "entropy": 5.6629842758178714,
      "epoch": 1.0169712245326612,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000490242563559905,
      "loss": 5.4974,
      "mean_token_accuracy": 0.16227589547634125,
      "num_tokens": 22334038.0,
      "step": 12105
    },
    {
      "entropy": 5.6420886516571045,
      "epoch": 1.017391304347826,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004902338384069408,
      "loss": 5.3571,
      "mean_token_accuracy": 0.16831498742103576,
      "num_tokens": 22342658.0,
      "step": 12110
    },
    {
      "entropy": 5.773781061172485,
      "epoch": 1.017811384162991,
      "grad_norm": 1.640625,
      "learning_rate": 0.00049022510944122,
      "loss": 5.5329,
      "mean_token_accuracy": 0.1544503778219223,
      "num_tokens": 22352559.0,
      "step": 12115
    },
    {
      "entropy": 5.733228349685669,
      "epoch": 1.0182314639781558,
      "grad_norm": 1.875,
      "learning_rate": 0.0004902163766628972,
      "loss": 5.448,
      "mean_token_accuracy": 0.16392027586698532,
      "num_tokens": 22361455.0,
      "step": 12120
    },
    {
      "entropy": 5.713450241088867,
      "epoch": 1.0186515437933208,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004902076400721271,
      "loss": 5.4895,
      "mean_token_accuracy": 0.16010515689849852,
      "num_tokens": 22371163.0,
      "step": 12125
    },
    {
      "entropy": 5.771074008941651,
      "epoch": 1.0190716236084856,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004901988996690645,
      "loss": 5.4849,
      "mean_token_accuracy": 0.16219335049390793,
      "num_tokens": 22379975.0,
      "step": 12130
    },
    {
      "entropy": 5.794712877273559,
      "epoch": 1.0194917034236506,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004901901554538641,
      "loss": 5.5188,
      "mean_token_accuracy": 0.16042946726083757,
      "num_tokens": 22389657.0,
      "step": 12135
    },
    {
      "entropy": 5.588501882553101,
      "epoch": 1.0199117832388154,
      "grad_norm": 1.796875,
      "learning_rate": 0.000490181407426681,
      "loss": 5.341,
      "mean_token_accuracy": 0.17044000029563905,
      "num_tokens": 22398320.0,
      "step": 12140
    },
    {
      "entropy": 5.666534328460694,
      "epoch": 1.0203318630539802,
      "grad_norm": 2.0,
      "learning_rate": 0.0004901726555876701,
      "loss": 5.5372,
      "mean_token_accuracy": 0.15764785856008529,
      "num_tokens": 22406634.0,
      "step": 12145
    },
    {
      "entropy": 5.775462627410889,
      "epoch": 1.0207519428691452,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004901638999369862,
      "loss": 5.5891,
      "mean_token_accuracy": 0.1614912211894989,
      "num_tokens": 22415939.0,
      "step": 12150
    },
    {
      "entropy": 5.739730596542358,
      "epoch": 1.02117202268431,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004901551404747847,
      "loss": 5.5275,
      "mean_token_accuracy": 0.15698569566011428,
      "num_tokens": 22425256.0,
      "step": 12155
    },
    {
      "entropy": 5.706746006011963,
      "epoch": 1.021592102499475,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004901463772012209,
      "loss": 5.5684,
      "mean_token_accuracy": 0.15766305550932885,
      "num_tokens": 22434750.0,
      "step": 12160
    },
    {
      "entropy": 5.704181098937989,
      "epoch": 1.0220121823146397,
      "grad_norm": 2.125,
      "learning_rate": 0.0004901376101164495,
      "loss": 5.4751,
      "mean_token_accuracy": 0.15738504827022554,
      "num_tokens": 22443426.0,
      "step": 12165
    },
    {
      "entropy": 5.729747581481933,
      "epoch": 1.0224322621298048,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004901288392206263,
      "loss": 5.4715,
      "mean_token_accuracy": 0.15853249728679658,
      "num_tokens": 22452778.0,
      "step": 12170
    },
    {
      "entropy": 5.6724381923675535,
      "epoch": 1.0228523419449695,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004901200645139064,
      "loss": 5.4361,
      "mean_token_accuracy": 0.17019454389810562,
      "num_tokens": 22462864.0,
      "step": 12175
    },
    {
      "entropy": 5.674047183990479,
      "epoch": 1.0232724217601343,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004901112859964454,
      "loss": 5.4971,
      "mean_token_accuracy": 0.1562313809990883,
      "num_tokens": 22472849.0,
      "step": 12180
    },
    {
      "entropy": 5.676679944992065,
      "epoch": 1.0236925015752993,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004901025036683987,
      "loss": 5.4195,
      "mean_token_accuracy": 0.1597939297556877,
      "num_tokens": 22481693.0,
      "step": 12185
    },
    {
      "entropy": 5.6579413414001465,
      "epoch": 1.0241125813904641,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004900937175299219,
      "loss": 5.4204,
      "mean_token_accuracy": 0.16194331496953965,
      "num_tokens": 22490934.0,
      "step": 12190
    },
    {
      "entropy": 5.62444052696228,
      "epoch": 1.0245326612056291,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004900849275811707,
      "loss": 5.4767,
      "mean_token_accuracy": 0.15643872767686845,
      "num_tokens": 22500457.0,
      "step": 12195
    },
    {
      "entropy": 5.7097996234893795,
      "epoch": 1.024952741020794,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004900761338223007,
      "loss": 5.4013,
      "mean_token_accuracy": 0.16027526408433915,
      "num_tokens": 22509641.0,
      "step": 12200
    },
    {
      "entropy": 5.655760145187378,
      "epoch": 1.025372820835959,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004900673362534677,
      "loss": 5.347,
      "mean_token_accuracy": 0.17173382937908171,
      "num_tokens": 22518616.0,
      "step": 12205
    },
    {
      "entropy": 5.6903989791870115,
      "epoch": 1.0257929006511237,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004900585348748277,
      "loss": 5.4952,
      "mean_token_accuracy": 0.16641637086868286,
      "num_tokens": 22527599.0,
      "step": 12210
    },
    {
      "entropy": 5.6446558952331545,
      "epoch": 1.0262129804662885,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004900497296865365,
      "loss": 5.5004,
      "mean_token_accuracy": 0.1569361686706543,
      "num_tokens": 22537399.0,
      "step": 12215
    },
    {
      "entropy": 5.80555911064148,
      "epoch": 1.0266330602814535,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004900409206887499,
      "loss": 5.6797,
      "mean_token_accuracy": 0.15186302959918976,
      "num_tokens": 22546746.0,
      "step": 12220
    },
    {
      "entropy": 5.7677764892578125,
      "epoch": 1.0270531400966183,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004900321078816243,
      "loss": 5.4782,
      "mean_token_accuracy": 0.16736462563276291,
      "num_tokens": 22555735.0,
      "step": 12225
    },
    {
      "entropy": 5.750488138198852,
      "epoch": 1.0274732199117833,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004900232912653156,
      "loss": 5.4762,
      "mean_token_accuracy": 0.16161205470561982,
      "num_tokens": 22565010.0,
      "step": 12230
    },
    {
      "entropy": 5.669838380813599,
      "epoch": 1.027893299726948,
      "grad_norm": 2.5625,
      "learning_rate": 0.00049001447083998,
      "loss": 5.4535,
      "mean_token_accuracy": 0.1628490686416626,
      "num_tokens": 22573565.0,
      "step": 12235
    },
    {
      "entropy": 5.709480333328247,
      "epoch": 1.028313379542113,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004900056466057737,
      "loss": 5.4534,
      "mean_token_accuracy": 0.15907117128372192,
      "num_tokens": 22582549.0,
      "step": 12240
    },
    {
      "entropy": 5.680795192718506,
      "epoch": 1.028733459357278,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004899968185628531,
      "loss": 5.5212,
      "mean_token_accuracy": 0.1543809860944748,
      "num_tokens": 22592112.0,
      "step": 12245
    },
    {
      "entropy": 5.6241144180297855,
      "epoch": 1.0291535391724427,
      "grad_norm": 7.40625,
      "learning_rate": 0.0004899879867113746,
      "loss": 5.3522,
      "mean_token_accuracy": 0.1685326635837555,
      "num_tokens": 22600581.0,
      "step": 12250
    },
    {
      "entropy": 5.778968381881714,
      "epoch": 1.0295736189876077,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004899791510514945,
      "loss": 5.5624,
      "mean_token_accuracy": 0.1573510617017746,
      "num_tokens": 22610822.0,
      "step": 12255
    },
    {
      "entropy": 5.718343067169189,
      "epoch": 1.0299936988027725,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004899703115833696,
      "loss": 5.5635,
      "mean_token_accuracy": 0.15985433757305145,
      "num_tokens": 22619484.0,
      "step": 12260
    },
    {
      "entropy": 5.702852535247803,
      "epoch": 1.0304137786179375,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004899614683071563,
      "loss": 5.4026,
      "mean_token_accuracy": 0.16532386988401412,
      "num_tokens": 22629038.0,
      "step": 12265
    },
    {
      "entropy": 5.710755681991577,
      "epoch": 1.0308338584331023,
      "grad_norm": 2.0,
      "learning_rate": 0.0004899526212230112,
      "loss": 5.5015,
      "mean_token_accuracy": 0.15492028892040252,
      "num_tokens": 22638619.0,
      "step": 12270
    },
    {
      "entropy": 5.6497485637664795,
      "epoch": 1.0312539382482673,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004899437703310912,
      "loss": 5.4953,
      "mean_token_accuracy": 0.15784212797880173,
      "num_tokens": 22648065.0,
      "step": 12275
    },
    {
      "entropy": 5.785039758682251,
      "epoch": 1.031674018063432,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004899349156315529,
      "loss": 5.5548,
      "mean_token_accuracy": 0.15540689527988433,
      "num_tokens": 22658107.0,
      "step": 12280
    },
    {
      "entropy": 5.763218879699707,
      "epoch": 1.0320940978785969,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004899260571245533,
      "loss": 5.4364,
      "mean_token_accuracy": 0.15831174105405807,
      "num_tokens": 22667103.0,
      "step": 12285
    },
    {
      "entropy": 5.657338285446167,
      "epoch": 1.0325141776937619,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004899171948102492,
      "loss": 5.4215,
      "mean_token_accuracy": 0.1609957903623581,
      "num_tokens": 22676792.0,
      "step": 12290
    },
    {
      "entropy": 5.659443616867065,
      "epoch": 1.0329342575089266,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004899083286887977,
      "loss": 5.4341,
      "mean_token_accuracy": 0.16280112117528917,
      "num_tokens": 22685344.0,
      "step": 12295
    },
    {
      "entropy": 5.744201564788819,
      "epoch": 1.0333543373240917,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004898994587603559,
      "loss": 5.494,
      "mean_token_accuracy": 0.16114521920681,
      "num_tokens": 22694387.0,
      "step": 12300
    },
    {
      "entropy": 5.676422786712647,
      "epoch": 1.0337744171392564,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004898905850250807,
      "loss": 5.5338,
      "mean_token_accuracy": 0.15966002494096757,
      "num_tokens": 22704203.0,
      "step": 12305
    },
    {
      "entropy": 5.766281652450561,
      "epoch": 1.0341944969544214,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004898817074831295,
      "loss": 5.5864,
      "mean_token_accuracy": 0.15488215833902358,
      "num_tokens": 22713518.0,
      "step": 12310
    },
    {
      "entropy": 5.786115884780884,
      "epoch": 1.0346145767695862,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004898728261346595,
      "loss": 5.5937,
      "mean_token_accuracy": 0.15901960879564286,
      "num_tokens": 22722997.0,
      "step": 12315
    },
    {
      "entropy": 5.779094600677491,
      "epoch": 1.035034656584751,
      "grad_norm": 2.4375,
      "learning_rate": 0.000489863940979828,
      "loss": 5.517,
      "mean_token_accuracy": 0.15875154137611389,
      "num_tokens": 22732385.0,
      "step": 12320
    },
    {
      "entropy": 5.638829135894776,
      "epoch": 1.035454736399916,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004898550520187925,
      "loss": 5.3829,
      "mean_token_accuracy": 0.16449802070856095,
      "num_tokens": 22741148.0,
      "step": 12325
    },
    {
      "entropy": 5.625936222076416,
      "epoch": 1.0358748162150808,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004898461592517103,
      "loss": 5.3933,
      "mean_token_accuracy": 0.16470819115638732,
      "num_tokens": 22750239.0,
      "step": 12330
    },
    {
      "entropy": 5.745430994033813,
      "epoch": 1.0362948960302458,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004898372626787391,
      "loss": 5.5446,
      "mean_token_accuracy": 0.1539967328310013,
      "num_tokens": 22759290.0,
      "step": 12335
    },
    {
      "entropy": 5.816291427612304,
      "epoch": 1.0367149758454106,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004898283623000364,
      "loss": 5.5656,
      "mean_token_accuracy": 0.15133108496665953,
      "num_tokens": 22768450.0,
      "step": 12340
    },
    {
      "entropy": 5.727371311187744,
      "epoch": 1.0371350556605754,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004898194581157598,
      "loss": 5.4337,
      "mean_token_accuracy": 0.1574019357562065,
      "num_tokens": 22777711.0,
      "step": 12345
    },
    {
      "entropy": 5.715303945541382,
      "epoch": 1.0375551354757404,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004898105501260671,
      "loss": 5.5086,
      "mean_token_accuracy": 0.15961914360523224,
      "num_tokens": 22787153.0,
      "step": 12350
    },
    {
      "entropy": 5.750590991973877,
      "epoch": 1.0379752152909052,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004898016383311163,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16274025440216064,
      "num_tokens": 22797125.0,
      "step": 12355
    },
    {
      "entropy": 5.755704784393311,
      "epoch": 1.0383952951060702,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000489792722731065,
      "loss": 5.5023,
      "mean_token_accuracy": 0.16141920089721679,
      "num_tokens": 22806478.0,
      "step": 12360
    },
    {
      "entropy": 5.763215065002441,
      "epoch": 1.038815374921235,
      "grad_norm": 1.5,
      "learning_rate": 0.0004897838033260712,
      "loss": 5.5097,
      "mean_token_accuracy": 0.1506726086139679,
      "num_tokens": 22815375.0,
      "step": 12365
    },
    {
      "entropy": 5.749024534225464,
      "epoch": 1.0392354547364,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004897748801162929,
      "loss": 5.48,
      "mean_token_accuracy": 0.1599726364016533,
      "num_tokens": 22824401.0,
      "step": 12370
    },
    {
      "entropy": 5.738317537307739,
      "epoch": 1.0396555345515648,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004897659531018882,
      "loss": 5.5881,
      "mean_token_accuracy": 0.16056930348277093,
      "num_tokens": 22833933.0,
      "step": 12375
    },
    {
      "entropy": 5.668944311141968,
      "epoch": 1.0400756143667296,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004897570222830152,
      "loss": 5.466,
      "mean_token_accuracy": 0.15872760713100434,
      "num_tokens": 22843779.0,
      "step": 12380
    },
    {
      "entropy": 5.750452899932862,
      "epoch": 1.0404956941818946,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004897480876598322,
      "loss": 5.5429,
      "mean_token_accuracy": 0.15892354696989058,
      "num_tokens": 22852951.0,
      "step": 12385
    },
    {
      "entropy": 5.7894142150878904,
      "epoch": 1.0409157739970594,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004897391492324974,
      "loss": 5.5579,
      "mean_token_accuracy": 0.15414744317531587,
      "num_tokens": 22861398.0,
      "step": 12390
    },
    {
      "entropy": 5.685288190841675,
      "epoch": 1.0413358538122244,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004897302070011691,
      "loss": 5.443,
      "mean_token_accuracy": 0.16238410770893097,
      "num_tokens": 22870518.0,
      "step": 12395
    },
    {
      "entropy": 5.6724052906036375,
      "epoch": 1.0417559336273892,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004897212609660058,
      "loss": 5.5142,
      "mean_token_accuracy": 0.1540789134800434,
      "num_tokens": 22879389.0,
      "step": 12400
    },
    {
      "entropy": 5.71264681816101,
      "epoch": 1.0421760134425542,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004897123111271659,
      "loss": 5.5202,
      "mean_token_accuracy": 0.16525834202766418,
      "num_tokens": 22888977.0,
      "step": 12405
    },
    {
      "entropy": 5.797454452514648,
      "epoch": 1.042596093257719,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004897033574848079,
      "loss": 5.5219,
      "mean_token_accuracy": 0.1619536757469177,
      "num_tokens": 22898446.0,
      "step": 12410
    },
    {
      "entropy": 5.679445219039917,
      "epoch": 1.0430161730728837,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004896944000390907,
      "loss": 5.5193,
      "mean_token_accuracy": 0.160529488325119,
      "num_tokens": 22908044.0,
      "step": 12415
    },
    {
      "entropy": 5.76577320098877,
      "epoch": 1.0434362528880488,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004896854387901725,
      "loss": 5.5744,
      "mean_token_accuracy": 0.1569056034088135,
      "num_tokens": 22917330.0,
      "step": 12420
    },
    {
      "entropy": 5.791294860839844,
      "epoch": 1.0438563327032135,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004896764737382124,
      "loss": 5.5241,
      "mean_token_accuracy": 0.165054389834404,
      "num_tokens": 22927160.0,
      "step": 12425
    },
    {
      "entropy": 5.762517213821411,
      "epoch": 1.0442764125183785,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004896675048833691,
      "loss": 5.4934,
      "mean_token_accuracy": 0.15579424053430557,
      "num_tokens": 22936755.0,
      "step": 12430
    },
    {
      "entropy": 5.722726345062256,
      "epoch": 1.0446964923335433,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004896585322258014,
      "loss": 5.4676,
      "mean_token_accuracy": 0.16214157938957213,
      "num_tokens": 22945699.0,
      "step": 12435
    },
    {
      "entropy": 5.719200086593628,
      "epoch": 1.0451165721487083,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004896495557656685,
      "loss": 5.4392,
      "mean_token_accuracy": 0.16894277483224868,
      "num_tokens": 22954001.0,
      "step": 12440
    },
    {
      "entropy": 5.793335008621216,
      "epoch": 1.0455366519638731,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004896405755031293,
      "loss": 5.5349,
      "mean_token_accuracy": 0.15977277606725693,
      "num_tokens": 22963805.0,
      "step": 12445
    },
    {
      "entropy": 5.649032068252564,
      "epoch": 1.045956731779038,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004896315914383427,
      "loss": 5.488,
      "mean_token_accuracy": 0.15751570016145705,
      "num_tokens": 22973542.0,
      "step": 12450
    },
    {
      "entropy": 5.6373741149902346,
      "epoch": 1.046376811594203,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004896226035714679,
      "loss": 5.3747,
      "mean_token_accuracy": 0.1678253784775734,
      "num_tokens": 22982417.0,
      "step": 12455
    },
    {
      "entropy": 5.75918436050415,
      "epoch": 1.0467968914093677,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004896136119026642,
      "loss": 5.4927,
      "mean_token_accuracy": 0.1621846318244934,
      "num_tokens": 22992879.0,
      "step": 12460
    },
    {
      "entropy": 5.684741640090943,
      "epoch": 1.0472169712245327,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004896046164320911,
      "loss": 5.377,
      "mean_token_accuracy": 0.16672905832529067,
      "num_tokens": 23001344.0,
      "step": 12465
    },
    {
      "entropy": 5.665459394454956,
      "epoch": 1.0476370510396975,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004895956171599075,
      "loss": 5.4296,
      "mean_token_accuracy": 0.16854447424411773,
      "num_tokens": 23010007.0,
      "step": 12470
    },
    {
      "entropy": 5.713906764984131,
      "epoch": 1.0480571308548625,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004895866140862731,
      "loss": 5.5401,
      "mean_token_accuracy": 0.15899324417114258,
      "num_tokens": 23019120.0,
      "step": 12475
    },
    {
      "entropy": 5.713853216171264,
      "epoch": 1.0484772106700273,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004895776072113473,
      "loss": 5.5248,
      "mean_token_accuracy": 0.1668669253587723,
      "num_tokens": 23028562.0,
      "step": 12480
    },
    {
      "entropy": 5.7258385181427,
      "epoch": 1.048897290485192,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004895685965352898,
      "loss": 5.4638,
      "mean_token_accuracy": 0.15974617898464202,
      "num_tokens": 23037687.0,
      "step": 12485
    },
    {
      "entropy": 5.784030103683472,
      "epoch": 1.049317370300357,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004895595820582601,
      "loss": 5.4561,
      "mean_token_accuracy": 0.16309544295072556,
      "num_tokens": 23047475.0,
      "step": 12490
    },
    {
      "entropy": 5.745016288757324,
      "epoch": 1.0497374501155219,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004895505637804177,
      "loss": 5.5025,
      "mean_token_accuracy": 0.15667191743850709,
      "num_tokens": 23057475.0,
      "step": 12495
    },
    {
      "entropy": 5.627092552185059,
      "epoch": 1.050157529930687,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004895415417019227,
      "loss": 5.4526,
      "mean_token_accuracy": 0.1620897650718689,
      "num_tokens": 23066419.0,
      "step": 12500
    },
    {
      "entropy": 5.731131649017334,
      "epoch": 1.0505776097458517,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004895325158229346,
      "loss": 5.52,
      "mean_token_accuracy": 0.16126697659492492,
      "num_tokens": 23075516.0,
      "step": 12505
    },
    {
      "entropy": 5.699992942810058,
      "epoch": 1.0509976895610167,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004895234861436136,
      "loss": 5.4051,
      "mean_token_accuracy": 0.1671866923570633,
      "num_tokens": 23084132.0,
      "step": 12510
    },
    {
      "entropy": 5.773825979232788,
      "epoch": 1.0514177693761815,
      "grad_norm": 2.125,
      "learning_rate": 0.0004895144526641194,
      "loss": 5.4907,
      "mean_token_accuracy": 0.16415202766656875,
      "num_tokens": 23093958.0,
      "step": 12515
    },
    {
      "entropy": 5.825757265090942,
      "epoch": 1.0518378491913463,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004895054153846123,
      "loss": 5.5132,
      "mean_token_accuracy": 0.15762871354818345,
      "num_tokens": 23103524.0,
      "step": 12520
    },
    {
      "entropy": 5.695266485214233,
      "epoch": 1.0522579290065113,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004894963743052521,
      "loss": 5.4528,
      "mean_token_accuracy": 0.1564144730567932,
      "num_tokens": 23112445.0,
      "step": 12525
    },
    {
      "entropy": 5.752588367462158,
      "epoch": 1.052678008821676,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004894873294261991,
      "loss": 5.5023,
      "mean_token_accuracy": 0.15996319204568862,
      "num_tokens": 23121299.0,
      "step": 12530
    },
    {
      "entropy": 5.796125411987305,
      "epoch": 1.053098088636841,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004894782807476134,
      "loss": 5.515,
      "mean_token_accuracy": 0.15295109748840333,
      "num_tokens": 23130260.0,
      "step": 12535
    },
    {
      "entropy": 5.73946213722229,
      "epoch": 1.0535181684520059,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004894692282696555,
      "loss": 5.4445,
      "mean_token_accuracy": 0.16455678939819335,
      "num_tokens": 23139335.0,
      "step": 12540
    },
    {
      "entropy": 5.676587677001953,
      "epoch": 1.0539382482671709,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004894601719924857,
      "loss": 5.4467,
      "mean_token_accuracy": 0.16182892471551896,
      "num_tokens": 23149299.0,
      "step": 12545
    },
    {
      "entropy": 5.657404994964599,
      "epoch": 1.0543583280823356,
      "grad_norm": 1.625,
      "learning_rate": 0.0004894511119162644,
      "loss": 5.4024,
      "mean_token_accuracy": 0.16588439494371415,
      "num_tokens": 23158651.0,
      "step": 12550
    },
    {
      "entropy": 5.765262174606323,
      "epoch": 1.0547784078975004,
      "grad_norm": 3.09375,
      "learning_rate": 0.000489442048041152,
      "loss": 5.5059,
      "mean_token_accuracy": 0.15651110857725142,
      "num_tokens": 23167629.0,
      "step": 12555
    },
    {
      "entropy": 5.784561347961426,
      "epoch": 1.0551984877126654,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004894329803673092,
      "loss": 5.4831,
      "mean_token_accuracy": 0.15658986270427705,
      "num_tokens": 23177026.0,
      "step": 12560
    },
    {
      "entropy": 5.715599107742309,
      "epoch": 1.0556185675278302,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004894239088948964,
      "loss": 5.4563,
      "mean_token_accuracy": 0.16440708339214324,
      "num_tokens": 23185297.0,
      "step": 12565
    },
    {
      "entropy": 5.6783417701721195,
      "epoch": 1.0560386473429952,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004894148336240747,
      "loss": 5.4547,
      "mean_token_accuracy": 0.16344198435544968,
      "num_tokens": 23194804.0,
      "step": 12570
    },
    {
      "entropy": 5.733673143386841,
      "epoch": 1.05645872715816,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004894057545550045,
      "loss": 5.5114,
      "mean_token_accuracy": 0.15659749060869216,
      "num_tokens": 23205063.0,
      "step": 12575
    },
    {
      "entropy": 5.70140643119812,
      "epoch": 1.056878806973325,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004893966716878467,
      "loss": 5.4211,
      "mean_token_accuracy": 0.16358228921890258,
      "num_tokens": 23215038.0,
      "step": 12580
    },
    {
      "entropy": 5.809863042831421,
      "epoch": 1.0572988867884898,
      "grad_norm": 1.875,
      "learning_rate": 0.0004893875850227624,
      "loss": 5.5971,
      "mean_token_accuracy": 0.15614084005355836,
      "num_tokens": 23223530.0,
      "step": 12585
    },
    {
      "entropy": 5.7787669658660885,
      "epoch": 1.0577189666036546,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004893784945599124,
      "loss": 5.5362,
      "mean_token_accuracy": 0.15725200325250627,
      "num_tokens": 23232547.0,
      "step": 12590
    },
    {
      "entropy": 5.7005228996276855,
      "epoch": 1.0581390464188196,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004893694002994577,
      "loss": 5.5717,
      "mean_token_accuracy": 0.15885683745145798,
      "num_tokens": 23241305.0,
      "step": 12595
    },
    {
      "entropy": 5.873926401138306,
      "epoch": 1.0585591262339844,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004893603022415595,
      "loss": 5.5929,
      "mean_token_accuracy": 0.16115136668086052,
      "num_tokens": 23250708.0,
      "step": 12600
    },
    {
      "entropy": 5.781400442123413,
      "epoch": 1.0589792060491494,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004893512003863788,
      "loss": 5.4929,
      "mean_token_accuracy": 0.15788125842809678,
      "num_tokens": 23260161.0,
      "step": 12605
    },
    {
      "entropy": 5.643087482452392,
      "epoch": 1.0593992858643142,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004893420947340771,
      "loss": 5.3975,
      "mean_token_accuracy": 0.15917274206876755,
      "num_tokens": 23268932.0,
      "step": 12610
    },
    {
      "entropy": 5.656122493743896,
      "epoch": 1.0598193656794792,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004893329852848155,
      "loss": 5.49,
      "mean_token_accuracy": 0.1633685499429703,
      "num_tokens": 23277741.0,
      "step": 12615
    },
    {
      "entropy": 5.695137357711792,
      "epoch": 1.060239445494644,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004893238720387555,
      "loss": 5.4876,
      "mean_token_accuracy": 0.16543712615966796,
      "num_tokens": 23286982.0,
      "step": 12620
    },
    {
      "entropy": 5.719235515594482,
      "epoch": 1.0606595253098088,
      "grad_norm": 2.0,
      "learning_rate": 0.0004893147549960584,
      "loss": 5.4086,
      "mean_token_accuracy": 0.1632736824452877,
      "num_tokens": 23296902.0,
      "step": 12625
    },
    {
      "entropy": 5.667002820968628,
      "epoch": 1.0610796051249738,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004893056341568857,
      "loss": 5.4446,
      "mean_token_accuracy": 0.17050562798976898,
      "num_tokens": 23305443.0,
      "step": 12630
    },
    {
      "entropy": 5.652026033401489,
      "epoch": 1.0614996849401386,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004892965095213992,
      "loss": 5.4055,
      "mean_token_accuracy": 0.16462423503398896,
      "num_tokens": 23315420.0,
      "step": 12635
    },
    {
      "entropy": 5.749802541732788,
      "epoch": 1.0619197647553036,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004892873810897604,
      "loss": 5.52,
      "mean_token_accuracy": 0.15769467800855635,
      "num_tokens": 23324540.0,
      "step": 12640
    },
    {
      "entropy": 5.699535608291626,
      "epoch": 1.0623398445704684,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004892782488621308,
      "loss": 5.4591,
      "mean_token_accuracy": 0.16499506682157516,
      "num_tokens": 23334282.0,
      "step": 12645
    },
    {
      "entropy": 5.66744704246521,
      "epoch": 1.0627599243856332,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004892691128386725,
      "loss": 5.4481,
      "mean_token_accuracy": 0.16394151598215104,
      "num_tokens": 23342836.0,
      "step": 12650
    },
    {
      "entropy": 5.73470139503479,
      "epoch": 1.0631800042007982,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004892599730195471,
      "loss": 5.4398,
      "mean_token_accuracy": 0.16658534705638886,
      "num_tokens": 23351863.0,
      "step": 12655
    },
    {
      "entropy": 5.849531030654907,
      "epoch": 1.063600084015963,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004892508294049167,
      "loss": 5.587,
      "mean_token_accuracy": 0.16241029053926467,
      "num_tokens": 23361788.0,
      "step": 12660
    },
    {
      "entropy": 5.734504842758179,
      "epoch": 1.064020163831128,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004892416819949431,
      "loss": 5.417,
      "mean_token_accuracy": 0.1586165800690651,
      "num_tokens": 23370175.0,
      "step": 12665
    },
    {
      "entropy": 5.662772369384766,
      "epoch": 1.0644402436462927,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004892325307897886,
      "loss": 5.4679,
      "mean_token_accuracy": 0.16151935011148452,
      "num_tokens": 23378835.0,
      "step": 12670
    },
    {
      "entropy": 5.69964337348938,
      "epoch": 1.0648603234614578,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004892233757896149,
      "loss": 5.4882,
      "mean_token_accuracy": 0.16086159497499466,
      "num_tokens": 23389390.0,
      "step": 12675
    },
    {
      "entropy": 5.747312068939209,
      "epoch": 1.0652804032766225,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004892142169945845,
      "loss": 5.4696,
      "mean_token_accuracy": 0.16162890940904617,
      "num_tokens": 23398802.0,
      "step": 12680
    },
    {
      "entropy": 5.651406192779541,
      "epoch": 1.0657004830917876,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004892050544048596,
      "loss": 5.4437,
      "mean_token_accuracy": 0.1581580176949501,
      "num_tokens": 23407731.0,
      "step": 12685
    },
    {
      "entropy": 5.678383111953735,
      "epoch": 1.0661205629069523,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004891958880206024,
      "loss": 5.4943,
      "mean_token_accuracy": 0.1613358035683632,
      "num_tokens": 23417046.0,
      "step": 12690
    },
    {
      "entropy": 5.715891933441162,
      "epoch": 1.0665406427221171,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004891867178419753,
      "loss": 5.483,
      "mean_token_accuracy": 0.16457503885030747,
      "num_tokens": 23426107.0,
      "step": 12695
    },
    {
      "entropy": 5.764684009552002,
      "epoch": 1.0669607225372821,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004891775438691408,
      "loss": 5.529,
      "mean_token_accuracy": 0.1606599673628807,
      "num_tokens": 23435523.0,
      "step": 12700
    },
    {
      "entropy": 5.7297052383422855,
      "epoch": 1.067380802352447,
      "grad_norm": 1.875,
      "learning_rate": 0.0004891683661022615,
      "loss": 5.4656,
      "mean_token_accuracy": 0.16791018098592758,
      "num_tokens": 23444185.0,
      "step": 12705
    },
    {
      "entropy": 5.854583692550659,
      "epoch": 1.067800882167612,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004891591845414997,
      "loss": 5.6673,
      "mean_token_accuracy": 0.14531354904174804,
      "num_tokens": 23454100.0,
      "step": 12710
    },
    {
      "entropy": 5.826853799819946,
      "epoch": 1.0682209619827767,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004891499991870184,
      "loss": 5.551,
      "mean_token_accuracy": 0.1544210582971573,
      "num_tokens": 23463415.0,
      "step": 12715
    },
    {
      "entropy": 5.737698650360107,
      "epoch": 1.0686410417979415,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048914081003898,
      "loss": 5.4489,
      "mean_token_accuracy": 0.16085606068372726,
      "num_tokens": 23471515.0,
      "step": 12720
    },
    {
      "entropy": 5.716708326339722,
      "epoch": 1.0690611216131065,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004891316170975475,
      "loss": 5.519,
      "mean_token_accuracy": 0.15879474133253096,
      "num_tokens": 23481696.0,
      "step": 12725
    },
    {
      "entropy": 5.769877147674561,
      "epoch": 1.0694812014282713,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004891224203628836,
      "loss": 5.4707,
      "mean_token_accuracy": 0.16245131343603134,
      "num_tokens": 23490714.0,
      "step": 12730
    },
    {
      "entropy": 5.664735269546509,
      "epoch": 1.0699012812434363,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004891132198351514,
      "loss": 5.4382,
      "mean_token_accuracy": 0.16582657247781754,
      "num_tokens": 23500368.0,
      "step": 12735
    },
    {
      "entropy": 5.5692836284637455,
      "epoch": 1.070321361058601,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004891040155145137,
      "loss": 5.4033,
      "mean_token_accuracy": 0.16731729805469514,
      "num_tokens": 23508857.0,
      "step": 12740
    },
    {
      "entropy": 5.614154720306397,
      "epoch": 1.070741440873766,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004890948074011335,
      "loss": 5.3763,
      "mean_token_accuracy": 0.17136704474687575,
      "num_tokens": 23518128.0,
      "step": 12745
    },
    {
      "entropy": 5.760420560836792,
      "epoch": 1.071161520688931,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004890855954951741,
      "loss": 5.484,
      "mean_token_accuracy": 0.16657705008983612,
      "num_tokens": 23527292.0,
      "step": 12750
    },
    {
      "entropy": 5.741787719726562,
      "epoch": 1.0715816005040957,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004890763797967987,
      "loss": 5.4957,
      "mean_token_accuracy": 0.16227886378765105,
      "num_tokens": 23535694.0,
      "step": 12755
    },
    {
      "entropy": 5.71260814666748,
      "epoch": 1.0720016803192607,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004890671603061704,
      "loss": 5.4826,
      "mean_token_accuracy": 0.1630891129374504,
      "num_tokens": 23544766.0,
      "step": 12760
    },
    {
      "entropy": 5.747214603424072,
      "epoch": 1.0724217601344255,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004890579370234526,
      "loss": 5.4528,
      "mean_token_accuracy": 0.17010122984647752,
      "num_tokens": 23554037.0,
      "step": 12765
    },
    {
      "entropy": 5.735277700424194,
      "epoch": 1.0728418399495905,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004890487099488086,
      "loss": 5.5074,
      "mean_token_accuracy": 0.16082375198602678,
      "num_tokens": 23562282.0,
      "step": 12770
    },
    {
      "entropy": 5.7627918243408205,
      "epoch": 1.0732619197647553,
      "grad_norm": 2.03125,
      "learning_rate": 0.000489039479082402,
      "loss": 5.5698,
      "mean_token_accuracy": 0.15416947677731513,
      "num_tokens": 23571955.0,
      "step": 12775
    },
    {
      "entropy": 5.707983779907226,
      "epoch": 1.0736819995799203,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004890302444243962,
      "loss": 5.4593,
      "mean_token_accuracy": 0.15817068070173262,
      "num_tokens": 23580996.0,
      "step": 12780
    },
    {
      "entropy": 5.742614698410034,
      "epoch": 1.074102079395085,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004890210059749549,
      "loss": 5.5578,
      "mean_token_accuracy": 0.15325147658586502,
      "num_tokens": 23589618.0,
      "step": 12785
    },
    {
      "entropy": 5.707106637954712,
      "epoch": 1.0745221592102498,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004890117637342416,
      "loss": 5.3992,
      "mean_token_accuracy": 0.15842635929584503,
      "num_tokens": 23599574.0,
      "step": 12790
    },
    {
      "entropy": 5.699951601028443,
      "epoch": 1.0749422390254149,
      "grad_norm": 1.75,
      "learning_rate": 0.0004890025177024202,
      "loss": 5.4576,
      "mean_token_accuracy": 0.15711033195257187,
      "num_tokens": 23609205.0,
      "step": 12795
    },
    {
      "entropy": 5.704383707046508,
      "epoch": 1.0753623188405796,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004889932678796543,
      "loss": 5.477,
      "mean_token_accuracy": 0.1553438723087311,
      "num_tokens": 23617554.0,
      "step": 12800
    },
    {
      "entropy": 5.7429163455963135,
      "epoch": 1.0757823986557447,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004889840142661078,
      "loss": 5.5275,
      "mean_token_accuracy": 0.1620032712817192,
      "num_tokens": 23626757.0,
      "step": 12805
    },
    {
      "entropy": 5.784508323669433,
      "epoch": 1.0762024784709094,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004889747568619447,
      "loss": 5.4918,
      "mean_token_accuracy": 0.16077843606472014,
      "num_tokens": 23636111.0,
      "step": 12810
    },
    {
      "entropy": 5.714294719696045,
      "epoch": 1.0766225582860744,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004889654956673291,
      "loss": 5.4738,
      "mean_token_accuracy": 0.1631556496024132,
      "num_tokens": 23644579.0,
      "step": 12815
    },
    {
      "entropy": 5.718669843673706,
      "epoch": 1.0770426381012392,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004889562306824248,
      "loss": 5.3889,
      "mean_token_accuracy": 0.1642576903104782,
      "num_tokens": 23653263.0,
      "step": 12820
    },
    {
      "entropy": 5.622420167922973,
      "epoch": 1.077462717916404,
      "grad_norm": 2.734375,
      "learning_rate": 0.000488946961907396,
      "loss": 5.3746,
      "mean_token_accuracy": 0.17475783824920654,
      "num_tokens": 23662529.0,
      "step": 12825
    },
    {
      "entropy": 5.601463699340821,
      "epoch": 1.077882797731569,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004889376893424071,
      "loss": 5.4116,
      "mean_token_accuracy": 0.16859746128320693,
      "num_tokens": 23671491.0,
      "step": 12830
    },
    {
      "entropy": 5.712467527389526,
      "epoch": 1.0783028775467338,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004889284129876221,
      "loss": 5.3964,
      "mean_token_accuracy": 0.1634058475494385,
      "num_tokens": 23680121.0,
      "step": 12835
    },
    {
      "entropy": 5.710364437103271,
      "epoch": 1.0787229573618988,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004889191328432054,
      "loss": 5.4514,
      "mean_token_accuracy": 0.16312524527311326,
      "num_tokens": 23689008.0,
      "step": 12840
    },
    {
      "entropy": 5.679463958740234,
      "epoch": 1.0791430371770636,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004889098489093215,
      "loss": 5.4921,
      "mean_token_accuracy": 0.15698808282613755,
      "num_tokens": 23698551.0,
      "step": 12845
    },
    {
      "entropy": 5.807032060623169,
      "epoch": 1.0795631169922286,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004889005611861347,
      "loss": 5.653,
      "mean_token_accuracy": 0.15691937357187272,
      "num_tokens": 23707438.0,
      "step": 12850
    },
    {
      "entropy": 5.740315341949463,
      "epoch": 1.0799831968073934,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004888912696738096,
      "loss": 5.4873,
      "mean_token_accuracy": 0.1622299484908581,
      "num_tokens": 23715822.0,
      "step": 12855
    },
    {
      "entropy": 5.710565185546875,
      "epoch": 1.0804032766225582,
      "grad_norm": 2.125,
      "learning_rate": 0.0004888819743725108,
      "loss": 5.5124,
      "mean_token_accuracy": 0.1617675766348839,
      "num_tokens": 23725426.0,
      "step": 12860
    },
    {
      "entropy": 5.759817123413086,
      "epoch": 1.0808233564377232,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000488872675282403,
      "loss": 5.5061,
      "mean_token_accuracy": 0.15907542854547502,
      "num_tokens": 23735092.0,
      "step": 12865
    },
    {
      "entropy": 5.7521524906158445,
      "epoch": 1.081243436252888,
      "grad_norm": 2.125,
      "learning_rate": 0.0004888633724036509,
      "loss": 5.483,
      "mean_token_accuracy": 0.1602173462510109,
      "num_tokens": 23744255.0,
      "step": 12870
    },
    {
      "entropy": 5.685547685623169,
      "epoch": 1.081663516068053,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004888540657364192,
      "loss": 5.3441,
      "mean_token_accuracy": 0.16899178177118301,
      "num_tokens": 23752978.0,
      "step": 12875
    },
    {
      "entropy": 5.691501188278198,
      "epoch": 1.0820835958832178,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004888447552808729,
      "loss": 5.4208,
      "mean_token_accuracy": 0.16169283539056778,
      "num_tokens": 23761051.0,
      "step": 12880
    },
    {
      "entropy": 5.744138574600219,
      "epoch": 1.0825036756983828,
      "grad_norm": 1.75,
      "learning_rate": 0.0004888354410371768,
      "loss": 5.5421,
      "mean_token_accuracy": 0.15464980006217957,
      "num_tokens": 23770818.0,
      "step": 12885
    },
    {
      "entropy": 5.808643198013305,
      "epoch": 1.0829237555135476,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000488826123005496,
      "loss": 5.551,
      "mean_token_accuracy": 0.16078845858573915,
      "num_tokens": 23780597.0,
      "step": 12890
    },
    {
      "entropy": 5.6849569320678714,
      "epoch": 1.0833438353287124,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004888168011859957,
      "loss": 5.3939,
      "mean_token_accuracy": 0.16438076347112657,
      "num_tokens": 23790119.0,
      "step": 12895
    },
    {
      "entropy": 5.684030294418335,
      "epoch": 1.0837639151438774,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004888074755788407,
      "loss": 5.4628,
      "mean_token_accuracy": 0.16382815986871718,
      "num_tokens": 23798972.0,
      "step": 12900
    },
    {
      "entropy": 5.710223627090454,
      "epoch": 1.0841839949590422,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004887981461841963,
      "loss": 5.4488,
      "mean_token_accuracy": 0.16691740453243256,
      "num_tokens": 23808685.0,
      "step": 12905
    },
    {
      "entropy": 5.720619106292725,
      "epoch": 1.0846040747742072,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004887888130022279,
      "loss": 5.4419,
      "mean_token_accuracy": 0.16202156841754914,
      "num_tokens": 23817721.0,
      "step": 12910
    },
    {
      "entropy": 5.633600378036499,
      "epoch": 1.085024154589372,
      "grad_norm": 2.25,
      "learning_rate": 0.0004887794760331008,
      "loss": 5.4104,
      "mean_token_accuracy": 0.16871272772550583,
      "num_tokens": 23826892.0,
      "step": 12915
    },
    {
      "entropy": 5.680303621292114,
      "epoch": 1.085444234404537,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004887701352769804,
      "loss": 5.3538,
      "mean_token_accuracy": 0.17344372123479843,
      "num_tokens": 23835717.0,
      "step": 12920
    },
    {
      "entropy": 5.67311372756958,
      "epoch": 1.0858643142197018,
      "grad_norm": 1.921875,
      "learning_rate": 0.000488760790734032,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16426583826541902,
      "num_tokens": 23845814.0,
      "step": 12925
    },
    {
      "entropy": 5.720589590072632,
      "epoch": 1.0862843940348665,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004887514424044214,
      "loss": 5.4313,
      "mean_token_accuracy": 0.1544219210743904,
      "num_tokens": 23854779.0,
      "step": 12930
    },
    {
      "entropy": 5.624525451660157,
      "epoch": 1.0867044738500315,
      "grad_norm": 2.15625,
      "learning_rate": 0.000488742090288314,
      "loss": 5.48,
      "mean_token_accuracy": 0.15788208842277526,
      "num_tokens": 23863533.0,
      "step": 12935
    },
    {
      "entropy": 5.733237457275391,
      "epoch": 1.0871245536651963,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004887327343858755,
      "loss": 5.5062,
      "mean_token_accuracy": 0.1594018206000328,
      "num_tokens": 23872725.0,
      "step": 12940
    },
    {
      "entropy": 5.74036169052124,
      "epoch": 1.0875446334803613,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004887233746972717,
      "loss": 5.5062,
      "mean_token_accuracy": 0.15623962581157685,
      "num_tokens": 23881799.0,
      "step": 12945
    },
    {
      "entropy": 5.698176908493042,
      "epoch": 1.0879647132955261,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004887140112226684,
      "loss": 5.5244,
      "mean_token_accuracy": 0.16199513375759125,
      "num_tokens": 23890628.0,
      "step": 12950
    },
    {
      "entropy": 5.695692157745361,
      "epoch": 1.088384793110691,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004887046439622314,
      "loss": 5.5056,
      "mean_token_accuracy": 0.1679466962814331,
      "num_tokens": 23899968.0,
      "step": 12955
    },
    {
      "entropy": 5.761798763275147,
      "epoch": 1.088804872925856,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004886952729161267,
      "loss": 5.3557,
      "mean_token_accuracy": 0.16911853849887848,
      "num_tokens": 23908634.0,
      "step": 12960
    },
    {
      "entropy": 5.672391033172607,
      "epoch": 1.0892249527410207,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004886858980845202,
      "loss": 5.5356,
      "mean_token_accuracy": 0.1591207727789879,
      "num_tokens": 23917925.0,
      "step": 12965
    },
    {
      "entropy": 5.618830633163452,
      "epoch": 1.0896450325561857,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004886765194675782,
      "loss": 5.4423,
      "mean_token_accuracy": 0.16554049998521805,
      "num_tokens": 23927173.0,
      "step": 12970
    },
    {
      "entropy": 5.706417846679687,
      "epoch": 1.0900651123713505,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004886671370654665,
      "loss": 5.4005,
      "mean_token_accuracy": 0.16858862042427064,
      "num_tokens": 23936258.0,
      "step": 12975
    },
    {
      "entropy": 5.690302896499634,
      "epoch": 1.0904851921865155,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004886577508783516,
      "loss": 5.3668,
      "mean_token_accuracy": 0.16924734264612198,
      "num_tokens": 23944215.0,
      "step": 12980
    },
    {
      "entropy": 5.726575565338135,
      "epoch": 1.0909052720016803,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004886483609063997,
      "loss": 5.4322,
      "mean_token_accuracy": 0.16069534420967102,
      "num_tokens": 23953151.0,
      "step": 12985
    },
    {
      "entropy": 5.620859098434448,
      "epoch": 1.0913253518168453,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004886389671497769,
      "loss": 5.4298,
      "mean_token_accuracy": 0.17138218581676484,
      "num_tokens": 23962919.0,
      "step": 12990
    },
    {
      "entropy": 5.771068859100342,
      "epoch": 1.09174543163201,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048862956960865,
      "loss": 5.4751,
      "mean_token_accuracy": 0.15873296856880187,
      "num_tokens": 23971900.0,
      "step": 12995
    },
    {
      "entropy": 5.6600103855133055,
      "epoch": 1.0921655114471749,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004886201682831852,
      "loss": 5.4397,
      "mean_token_accuracy": 0.16507715582847596,
      "num_tokens": 23980945.0,
      "step": 13000
    },
    {
      "entropy": 5.621271753311158,
      "epoch": 1.09258559126234,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004886107631735491,
      "loss": 5.3981,
      "mean_token_accuracy": 0.1617731973528862,
      "num_tokens": 23990460.0,
      "step": 13005
    },
    {
      "entropy": 5.712856674194336,
      "epoch": 1.0930056710775047,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004886013542799083,
      "loss": 5.5658,
      "mean_token_accuracy": 0.1500063270330429,
      "num_tokens": 23999925.0,
      "step": 13010
    },
    {
      "entropy": 5.6885334014892575,
      "epoch": 1.0934257508926697,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004885919416024296,
      "loss": 5.4223,
      "mean_token_accuracy": 0.16228862702846528,
      "num_tokens": 24009039.0,
      "step": 13015
    },
    {
      "entropy": 5.796774768829346,
      "epoch": 1.0938458307078345,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004885825251412796,
      "loss": 5.4635,
      "mean_token_accuracy": 0.16517345756292343,
      "num_tokens": 24017725.0,
      "step": 13020
    },
    {
      "entropy": 5.7257390975952145,
      "epoch": 1.0942659105229993,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004885731048966252,
      "loss": 5.4809,
      "mean_token_accuracy": 0.1577977254986763,
      "num_tokens": 24027158.0,
      "step": 13025
    },
    {
      "entropy": 5.62857027053833,
      "epoch": 1.0946859903381643,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004885636808686331,
      "loss": 5.5101,
      "mean_token_accuracy": 0.16572368741035462,
      "num_tokens": 24037224.0,
      "step": 13030
    },
    {
      "entropy": 5.723979425430298,
      "epoch": 1.095106070153329,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004885542530574705,
      "loss": 5.486,
      "mean_token_accuracy": 0.16158615350723265,
      "num_tokens": 24046097.0,
      "step": 13035
    },
    {
      "entropy": 5.68514404296875,
      "epoch": 1.095526149968494,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004885448214633042,
      "loss": 5.3743,
      "mean_token_accuracy": 0.16373368501663207,
      "num_tokens": 24055270.0,
      "step": 13040
    },
    {
      "entropy": 5.738873386383057,
      "epoch": 1.0959462297836589,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004885353860863013,
      "loss": 5.5534,
      "mean_token_accuracy": 0.15277508795261383,
      "num_tokens": 24064995.0,
      "step": 13045
    },
    {
      "entropy": 5.781196355819702,
      "epoch": 1.0963663095988239,
      "grad_norm": 1.59375,
      "learning_rate": 0.000488525946926629,
      "loss": 5.5763,
      "mean_token_accuracy": 0.1572740152478218,
      "num_tokens": 24075523.0,
      "step": 13050
    },
    {
      "entropy": 5.707388401031494,
      "epoch": 1.0967863894139886,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004885165039844545,
      "loss": 5.4404,
      "mean_token_accuracy": 0.1659671738743782,
      "num_tokens": 24084933.0,
      "step": 13055
    },
    {
      "entropy": 5.674459743499756,
      "epoch": 1.0972064692291534,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004885070572599452,
      "loss": 5.4905,
      "mean_token_accuracy": 0.15537894517183304,
      "num_tokens": 24093964.0,
      "step": 13060
    },
    {
      "entropy": 5.693677806854248,
      "epoch": 1.0976265490443184,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004884976067532681,
      "loss": 5.4426,
      "mean_token_accuracy": 0.1547652930021286,
      "num_tokens": 24103951.0,
      "step": 13065
    },
    {
      "entropy": 5.671582841873169,
      "epoch": 1.0980466288594832,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000488488152464591,
      "loss": 5.5455,
      "mean_token_accuracy": 0.15263120234012603,
      "num_tokens": 24113392.0,
      "step": 13070
    },
    {
      "entropy": 5.696406650543213,
      "epoch": 1.0984667086746482,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004884786943940812,
      "loss": 5.399,
      "mean_token_accuracy": 0.16182387322187425,
      "num_tokens": 24123165.0,
      "step": 13075
    },
    {
      "entropy": 5.67077898979187,
      "epoch": 1.098886788489813,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004884692325419063,
      "loss": 5.4597,
      "mean_token_accuracy": 0.161738720536232,
      "num_tokens": 24132176.0,
      "step": 13080
    },
    {
      "entropy": 5.703751039505005,
      "epoch": 1.099306868304978,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004884597669082336,
      "loss": 5.5231,
      "mean_token_accuracy": 0.1586092695593834,
      "num_tokens": 24141737.0,
      "step": 13085
    },
    {
      "entropy": 5.698824644088745,
      "epoch": 1.0997269481201428,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004884502974932313,
      "loss": 5.4568,
      "mean_token_accuracy": 0.1605760633945465,
      "num_tokens": 24150477.0,
      "step": 13090
    },
    {
      "entropy": 5.7847676277160645,
      "epoch": 1.1001470279353076,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004884408242970668,
      "loss": 5.5412,
      "mean_token_accuracy": 0.1626332238316536,
      "num_tokens": 24158739.0,
      "step": 13095
    },
    {
      "entropy": 5.6294517993927,
      "epoch": 1.1005671077504726,
      "grad_norm": 1.625,
      "learning_rate": 0.0004884313473199081,
      "loss": 5.3848,
      "mean_token_accuracy": 0.16554880887269974,
      "num_tokens": 24167511.0,
      "step": 13100
    },
    {
      "entropy": 5.6367988109588625,
      "epoch": 1.1009871875656374,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004884218665619229,
      "loss": 5.3834,
      "mean_token_accuracy": 0.1640613242983818,
      "num_tokens": 24176413.0,
      "step": 13105
    },
    {
      "entropy": 5.660191440582276,
      "epoch": 1.1014072673808024,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004884123820232792,
      "loss": 5.3493,
      "mean_token_accuracy": 0.17108992785215377,
      "num_tokens": 24185135.0,
      "step": 13110
    },
    {
      "entropy": 5.679283857345581,
      "epoch": 1.1018273471959672,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004884028937041451,
      "loss": 5.4342,
      "mean_token_accuracy": 0.16592397689819335,
      "num_tokens": 24193273.0,
      "step": 13115
    },
    {
      "entropy": 5.7048530101776125,
      "epoch": 1.1022474270111322,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004883934016046886,
      "loss": 5.5063,
      "mean_token_accuracy": 0.15767362713813782,
      "num_tokens": 24202509.0,
      "step": 13120
    },
    {
      "entropy": 5.723241710662842,
      "epoch": 1.102667506826297,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000488383905725078,
      "loss": 5.4921,
      "mean_token_accuracy": 0.15816541612148285,
      "num_tokens": 24212644.0,
      "step": 13125
    },
    {
      "entropy": 5.697927713394165,
      "epoch": 1.1030875866414618,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004883744060654811,
      "loss": 5.3794,
      "mean_token_accuracy": 0.16595268547534942,
      "num_tokens": 24221838.0,
      "step": 13130
    },
    {
      "entropy": 5.6857527732849125,
      "epoch": 1.1035076664566268,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004883649026260667,
      "loss": 5.4619,
      "mean_token_accuracy": 0.16348368525505066,
      "num_tokens": 24230987.0,
      "step": 13135
    },
    {
      "entropy": 5.650682592391968,
      "epoch": 1.1039277462717916,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004883553954070028,
      "loss": 5.4118,
      "mean_token_accuracy": 0.1683964341878891,
      "num_tokens": 24240523.0,
      "step": 13140
    },
    {
      "entropy": 5.722428798675537,
      "epoch": 1.1043478260869566,
      "grad_norm": 1.796875,
      "learning_rate": 0.000488345884408458,
      "loss": 5.5057,
      "mean_token_accuracy": 0.16416788548231126,
      "num_tokens": 24249799.0,
      "step": 13145
    },
    {
      "entropy": 5.680908250808716,
      "epoch": 1.1047679059021214,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004883363696306007,
      "loss": 5.454,
      "mean_token_accuracy": 0.16249977350234984,
      "num_tokens": 24259361.0,
      "step": 13150
    },
    {
      "entropy": 5.735293912887573,
      "epoch": 1.1051879857172864,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004883268510735995,
      "loss": 5.4144,
      "mean_token_accuracy": 0.16268240362405778,
      "num_tokens": 24268010.0,
      "step": 13155
    },
    {
      "entropy": 5.6632569313049315,
      "epoch": 1.1056080655324512,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004883173287376229,
      "loss": 5.4696,
      "mean_token_accuracy": 0.15777850821614264,
      "num_tokens": 24277416.0,
      "step": 13160
    },
    {
      "entropy": 5.762572288513184,
      "epoch": 1.106028145347616,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004883078026228397,
      "loss": 5.5442,
      "mean_token_accuracy": 0.15905994772911072,
      "num_tokens": 24286185.0,
      "step": 13165
    },
    {
      "entropy": 5.7264659881591795,
      "epoch": 1.106448225162781,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004882982727294187,
      "loss": 5.4127,
      "mean_token_accuracy": 0.16001077443361283,
      "num_tokens": 24295382.0,
      "step": 13170
    },
    {
      "entropy": 5.637160968780518,
      "epoch": 1.1068683049779457,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004882887390575284,
      "loss": 5.4253,
      "mean_token_accuracy": 0.16418798714876176,
      "num_tokens": 24305197.0,
      "step": 13175
    },
    {
      "entropy": 5.688330411911011,
      "epoch": 1.1072883847931108,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004882792016073381,
      "loss": 5.5197,
      "mean_token_accuracy": 0.15734292566776276,
      "num_tokens": 24314149.0,
      "step": 13180
    },
    {
      "entropy": 5.752083873748779,
      "epoch": 1.1077084646082755,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048826966037901655,
      "loss": 5.452,
      "mean_token_accuracy": 0.16154199242591857,
      "num_tokens": 24323737.0,
      "step": 13185
    },
    {
      "entropy": 5.683094501495361,
      "epoch": 1.1081285444234406,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048826011537273276,
      "loss": 5.4217,
      "mean_token_accuracy": 0.164495849609375,
      "num_tokens": 24332853.0,
      "step": 13190
    },
    {
      "entropy": 5.667849111557007,
      "epoch": 1.1085486242386053,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004882505665886558,
      "loss": 5.5583,
      "mean_token_accuracy": 0.15878505557775496,
      "num_tokens": 24342632.0,
      "step": 13195
    },
    {
      "entropy": 5.685575389862061,
      "epoch": 1.1089687040537701,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048824101402695493,
      "loss": 5.3884,
      "mean_token_accuracy": 0.16723015904426575,
      "num_tokens": 24351659.0,
      "step": 13200
    },
    {
      "entropy": 5.627767610549927,
      "epoch": 1.1093887838689351,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004882314576877993,
      "loss": 5.4213,
      "mean_token_accuracy": 0.1608801230788231,
      "num_tokens": 24360938.0,
      "step": 13205
    },
    {
      "entropy": 5.6773035526275635,
      "epoch": 1.1098088636841,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004882218975713581,
      "loss": 5.4965,
      "mean_token_accuracy": 0.16553150117397308,
      "num_tokens": 24369603.0,
      "step": 13210
    },
    {
      "entropy": 5.66359806060791,
      "epoch": 1.110228943499265,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004882123336778009,
      "loss": 5.4316,
      "mean_token_accuracy": 0.16241261065006257,
      "num_tokens": 24377605.0,
      "step": 13215
    },
    {
      "entropy": 5.703319883346557,
      "epoch": 1.1106490233144297,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004882027660072969,
      "loss": 5.4694,
      "mean_token_accuracy": 0.15938784927129745,
      "num_tokens": 24386930.0,
      "step": 13220
    },
    {
      "entropy": 5.699520874023437,
      "epoch": 1.1110691031295947,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004881931945600157,
      "loss": 5.4329,
      "mean_token_accuracy": 0.16960556656122208,
      "num_tokens": 24396473.0,
      "step": 13225
    },
    {
      "entropy": 5.74881911277771,
      "epoch": 1.1114891829447595,
      "grad_norm": 1.75,
      "learning_rate": 0.0004881836193361269,
      "loss": 5.5287,
      "mean_token_accuracy": 0.16469872146844863,
      "num_tokens": 24405461.0,
      "step": 13230
    },
    {
      "entropy": 5.740612363815307,
      "epoch": 1.1119092627599243,
      "grad_norm": 1.625,
      "learning_rate": 0.0004881740403358,
      "loss": 5.459,
      "mean_token_accuracy": 0.1663040205836296,
      "num_tokens": 24414138.0,
      "step": 13235
    },
    {
      "entropy": 5.664918518066406,
      "epoch": 1.1123293425750893,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048816445755920474,
      "loss": 5.4601,
      "mean_token_accuracy": 0.15880708545446395,
      "num_tokens": 24423386.0,
      "step": 13240
    },
    {
      "entropy": 5.675559234619141,
      "epoch": 1.112749422390254,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004881548710065109,
      "loss": 5.4695,
      "mean_token_accuracy": 0.15784148275852203,
      "num_tokens": 24433637.0,
      "step": 13245
    },
    {
      "entropy": 5.733191776275635,
      "epoch": 1.113169502205419,
      "grad_norm": 1.625,
      "learning_rate": 0.0004881452806778883,
      "loss": 5.5209,
      "mean_token_accuracy": 0.16069816052913666,
      "num_tokens": 24443677.0,
      "step": 13250
    },
    {
      "entropy": 5.709711790084839,
      "epoch": 1.113589582020584,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048813568657350676,
      "loss": 5.4059,
      "mean_token_accuracy": 0.16754952371120452,
      "num_tokens": 24452317.0,
      "step": 13255
    },
    {
      "entropy": 5.711081409454346,
      "epoch": 1.1140096618357487,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004881260886935363,
      "loss": 5.4314,
      "mean_token_accuracy": 0.1628634065389633,
      "num_tokens": 24460626.0,
      "step": 13260
    },
    {
      "entropy": 5.741526031494141,
      "epoch": 1.1144297416509137,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048811648703814693,
      "loss": 5.5194,
      "mean_token_accuracy": 0.15109701529145242,
      "num_tokens": 24469583.0,
      "step": 13265
    },
    {
      "entropy": 5.68180022239685,
      "epoch": 1.1148498214660785,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004881068816075087,
      "loss": 5.4656,
      "mean_token_accuracy": 0.16133692264556884,
      "num_tokens": 24478811.0,
      "step": 13270
    },
    {
      "entropy": 5.655600738525391,
      "epoch": 1.1152699012812435,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048809727240179193,
      "loss": 5.4917,
      "mean_token_accuracy": 0.1590549409389496,
      "num_tokens": 24487818.0,
      "step": 13275
    },
    {
      "entropy": 5.6665290832519535,
      "epoch": 1.1156899810964083,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004880876594211665,
      "loss": 5.4686,
      "mean_token_accuracy": 0.15892062336206436,
      "num_tokens": 24497087.0,
      "step": 13280
    },
    {
      "entropy": 5.743152189254761,
      "epoch": 1.1161100609115733,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048807804266580304,
      "loss": 5.4271,
      "mean_token_accuracy": 0.1575818032026291,
      "num_tokens": 24505347.0,
      "step": 13285
    },
    {
      "entropy": 5.763370513916016,
      "epoch": 1.116530140726738,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004880684221358717,
      "loss": 5.4393,
      "mean_token_accuracy": 0.15934379994869233,
      "num_tokens": 24514732.0,
      "step": 13290
    },
    {
      "entropy": 5.707498073577881,
      "epoch": 1.116950220541903,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048805879783154305,
      "loss": 5.4771,
      "mean_token_accuracy": 0.16071149557828904,
      "num_tokens": 24523295.0,
      "step": 13295
    },
    {
      "entropy": 5.655390310287475,
      "epoch": 1.1173703003570679,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048804916975298744,
      "loss": 5.3769,
      "mean_token_accuracy": 0.16449277400970458,
      "num_tokens": 24532415.0,
      "step": 13300
    },
    {
      "entropy": 5.74219388961792,
      "epoch": 1.1177903801722326,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004880395379003755,
      "loss": 5.5072,
      "mean_token_accuracy": 0.16419040709733962,
      "num_tokens": 24541856.0,
      "step": 13305
    },
    {
      "entropy": 5.645684480667114,
      "epoch": 1.1182104599873977,
      "grad_norm": 1.625,
      "learning_rate": 0.00048802990227387797,
      "loss": 5.5095,
      "mean_token_accuracy": 0.1570297934114933,
      "num_tokens": 24550982.0,
      "step": 13310
    },
    {
      "entropy": 5.759794473648071,
      "epoch": 1.1186305398025624,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048802026287366525,
      "loss": 5.583,
      "mean_token_accuracy": 0.1525949463248253,
      "num_tokens": 24561176.0,
      "step": 13315
    },
    {
      "entropy": 5.707194566726685,
      "epoch": 1.1190506196177274,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048801061969990834,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16643868386745453,
      "num_tokens": 24570741.0,
      "step": 13320
    },
    {
      "entropy": 5.665846681594848,
      "epoch": 1.1194706994328922,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048800097275277795,
      "loss": 5.4593,
      "mean_token_accuracy": 0.16555610448122024,
      "num_tokens": 24580175.0,
      "step": 13325
    },
    {
      "entropy": 5.731688928604126,
      "epoch": 1.119890779248057,
      "grad_norm": 1.53125,
      "learning_rate": 0.000487991322032445,
      "loss": 5.4515,
      "mean_token_accuracy": 0.16520103812217712,
      "num_tokens": 24588754.0,
      "step": 13330
    },
    {
      "entropy": 5.82857723236084,
      "epoch": 1.120310859063222,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004879816675390805,
      "loss": 5.6301,
      "mean_token_accuracy": 0.15790451541543007,
      "num_tokens": 24599429.0,
      "step": 13335
    },
    {
      "entropy": 5.654234075546265,
      "epoch": 1.1207309388783868,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048797200927285547,
      "loss": 5.3745,
      "mean_token_accuracy": 0.1670805871486664,
      "num_tokens": 24608767.0,
      "step": 13340
    },
    {
      "entropy": 5.682362413406372,
      "epoch": 1.1211510186935518,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004879623472339409,
      "loss": 5.5402,
      "mean_token_accuracy": 0.16218066960573196,
      "num_tokens": 24618232.0,
      "step": 13345
    },
    {
      "entropy": 5.671395492553711,
      "epoch": 1.1215710985087166,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000487952681422508,
      "loss": 5.4195,
      "mean_token_accuracy": 0.16594525426626205,
      "num_tokens": 24626986.0,
      "step": 13350
    },
    {
      "entropy": 5.57235198020935,
      "epoch": 1.1219911783238816,
      "grad_norm": 1.40625,
      "learning_rate": 0.000487943011838728,
      "loss": 5.2896,
      "mean_token_accuracy": 0.17551097571849822,
      "num_tokens": 24635283.0,
      "step": 13355
    },
    {
      "entropy": 5.603815412521362,
      "epoch": 1.1224112581390464,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004879333384827722,
      "loss": 5.4099,
      "mean_token_accuracy": 0.16004293411970139,
      "num_tokens": 24644451.0,
      "step": 13360
    },
    {
      "entropy": 5.793188524246216,
      "epoch": 1.1228313379542114,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004879236613548119,
      "loss": 5.5573,
      "mean_token_accuracy": 0.15772061198949813,
      "num_tokens": 24654811.0,
      "step": 13365
    },
    {
      "entropy": 5.736780405044556,
      "epoch": 1.1232514177693762,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004879139804550187,
      "loss": 5.4654,
      "mean_token_accuracy": 0.16606503427028657,
      "num_tokens": 24663712.0,
      "step": 13370
    },
    {
      "entropy": 5.724746227264404,
      "epoch": 1.123671497584541,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048790429578356387,
      "loss": 5.5665,
      "mean_token_accuracy": 0.15672928094863892,
      "num_tokens": 24672518.0,
      "step": 13375
    },
    {
      "entropy": 5.733008527755738,
      "epoch": 1.124091577399706,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048789460734061915,
      "loss": 5.4963,
      "mean_token_accuracy": 0.15912166833877564,
      "num_tokens": 24681900.0,
      "step": 13380
    },
    {
      "entropy": 5.731820964813233,
      "epoch": 1.1245116572148708,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004878849151263561,
      "loss": 5.4631,
      "mean_token_accuracy": 0.16634195446968078,
      "num_tokens": 24691760.0,
      "step": 13385
    },
    {
      "entropy": 5.733888721466064,
      "epoch": 1.1249317370300358,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004878752191409463,
      "loss": 5.4154,
      "mean_token_accuracy": 0.16850126534700394,
      "num_tokens": 24700742.0,
      "step": 13390
    },
    {
      "entropy": 5.593951797485351,
      "epoch": 1.1253518168452006,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004878655193845616,
      "loss": 5.4872,
      "mean_token_accuracy": 0.1581694796681404,
      "num_tokens": 24709329.0,
      "step": 13395
    },
    {
      "entropy": 5.692504644393921,
      "epoch": 1.1257718966603654,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048785581585737394,
      "loss": 5.6223,
      "mean_token_accuracy": 0.15529729127883912,
      "num_tokens": 24718475.0,
      "step": 13400
    },
    {
      "entropy": 5.802961778640747,
      "epoch": 1.1261919764755304,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000487846108559555,
      "loss": 5.4954,
      "mean_token_accuracy": 0.17018510401248932,
      "num_tokens": 24727817.0,
      "step": 13405
    },
    {
      "entropy": 5.656300020217896,
      "epoch": 1.1266120562906952,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048783639749127694,
      "loss": 5.4561,
      "mean_token_accuracy": 0.1604897528886795,
      "num_tokens": 24737057.0,
      "step": 13410
    },
    {
      "entropy": 5.670790195465088,
      "epoch": 1.1270321361058602,
      "grad_norm": 1.5,
      "learning_rate": 0.0004878266826527116,
      "loss": 5.5137,
      "mean_token_accuracy": 0.15796383768320083,
      "num_tokens": 24746016.0,
      "step": 13415
    },
    {
      "entropy": 5.821008825302124,
      "epoch": 1.127452215921025,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048781696404403126,
      "loss": 5.5078,
      "mean_token_accuracy": 0.16149032413959502,
      "num_tokens": 24755978.0,
      "step": 13420
    },
    {
      "entropy": 5.697250509262085,
      "epoch": 1.12787229573619,
      "grad_norm": 4.46875,
      "learning_rate": 0.00048780724166540794,
      "loss": 5.3952,
      "mean_token_accuracy": 0.16163186430931092,
      "num_tokens": 24765255.0,
      "step": 13425
    },
    {
      "entropy": 5.634158039093018,
      "epoch": 1.1282923755513548,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004877975155170139,
      "loss": 5.4805,
      "mean_token_accuracy": 0.15824283584952353,
      "num_tokens": 24774339.0,
      "step": 13430
    },
    {
      "entropy": 5.692260265350342,
      "epoch": 1.1287124553665198,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004877877855990215,
      "loss": 5.4813,
      "mean_token_accuracy": 0.15508203208446503,
      "num_tokens": 24783236.0,
      "step": 13435
    },
    {
      "entropy": 5.687100839614868,
      "epoch": 1.1291325351816845,
      "grad_norm": 1.5625,
      "learning_rate": 0.000487778051911603,
      "loss": 5.3898,
      "mean_token_accuracy": 0.1657722622156143,
      "num_tokens": 24792168.0,
      "step": 13440
    },
    {
      "entropy": 5.77692666053772,
      "epoch": 1.1295526149968493,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004877683144549308,
      "loss": 5.5277,
      "mean_token_accuracy": 0.1641865387558937,
      "num_tokens": 24800843.0,
      "step": 13445
    },
    {
      "entropy": 5.650931930541992,
      "epoch": 1.1299726948120143,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00048775857322917753,
      "loss": 5.426,
      "mean_token_accuracy": 0.16078004539012908,
      "num_tokens": 24810475.0,
      "step": 13450
    },
    {
      "entropy": 5.622085809707642,
      "epoch": 1.1303927746271791,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004877488282345158,
      "loss": 5.4865,
      "mean_token_accuracy": 0.16201561838388442,
      "num_tokens": 24820486.0,
      "step": 13455
    },
    {
      "entropy": 5.775274133682251,
      "epoch": 1.1308128544423441,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000487739079471118,
      "loss": 5.551,
      "mean_token_accuracy": 0.16165002584457397,
      "num_tokens": 24830243.0,
      "step": 13460
    },
    {
      "entropy": 5.764797496795654,
      "epoch": 1.131232934257509,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000487729326939157,
      "loss": 5.4526,
      "mean_token_accuracy": 0.1641850858926773,
      "num_tokens": 24839090.0,
      "step": 13465
    },
    {
      "entropy": 5.6410528182983395,
      "epoch": 1.1316530140726737,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048771957063880553,
      "loss": 5.4519,
      "mean_token_accuracy": 0.16257893294095993,
      "num_tokens": 24847933.0,
      "step": 13470
    },
    {
      "entropy": 5.754781723022461,
      "epoch": 1.1320730938878387,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004877098105702363,
      "loss": 5.4659,
      "mean_token_accuracy": 0.1691000446677208,
      "num_tokens": 24857037.0,
      "step": 13475
    },
    {
      "entropy": 5.613031244277954,
      "epoch": 1.1324931737030035,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048770004673362243,
      "loss": 5.2929,
      "mean_token_accuracy": 0.17315517216920853,
      "num_tokens": 24866042.0,
      "step": 13480
    },
    {
      "entropy": 5.554357528686523,
      "epoch": 1.1329132535181685,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048769027912913673,
      "loss": 5.2553,
      "mean_token_accuracy": 0.17282647341489793,
      "num_tokens": 24873735.0,
      "step": 13485
    },
    {
      "entropy": 5.519758033752441,
      "epoch": 1.1333333333333333,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004876805077569522,
      "loss": 5.3555,
      "mean_token_accuracy": 0.16498854458332063,
      "num_tokens": 24882277.0,
      "step": 13490
    },
    {
      "entropy": 5.610137462615967,
      "epoch": 1.133753413148498,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048767073261724204,
      "loss": 5.4597,
      "mean_token_accuracy": 0.1604814663529396,
      "num_tokens": 24891354.0,
      "step": 13495
    },
    {
      "entropy": 5.69881238937378,
      "epoch": 1.134173492963663,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004876609537101793,
      "loss": 5.4431,
      "mean_token_accuracy": 0.16179838478565217,
      "num_tokens": 24899887.0,
      "step": 13500
    },
    {
      "entropy": 5.7566437244415285,
      "epoch": 1.1345935727788279,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004876511710359374,
      "loss": 5.4648,
      "mean_token_accuracy": 0.16314765214920043,
      "num_tokens": 24908616.0,
      "step": 13505
    },
    {
      "entropy": 5.7111047267913815,
      "epoch": 1.135013652593993,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048764138459468935,
      "loss": 5.5047,
      "mean_token_accuracy": 0.1574953481554985,
      "num_tokens": 24917864.0,
      "step": 13510
    },
    {
      "entropy": 5.731399297714233,
      "epoch": 1.1354337324091577,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048763159438660876,
      "loss": 5.527,
      "mean_token_accuracy": 0.1585410252213478,
      "num_tokens": 24927864.0,
      "step": 13515
    },
    {
      "entropy": 5.650017118453979,
      "epoch": 1.1358538122243227,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048762180041186893,
      "loss": 5.4189,
      "mean_token_accuracy": 0.1666693225502968,
      "num_tokens": 24937146.0,
      "step": 13520
    },
    {
      "entropy": 5.721314907073975,
      "epoch": 1.1362738920394875,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004876120026706434,
      "loss": 5.5001,
      "mean_token_accuracy": 0.16370597183704377,
      "num_tokens": 24945694.0,
      "step": 13525
    },
    {
      "entropy": 5.627752208709717,
      "epoch": 1.1366939718546525,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004876022011631057,
      "loss": 5.3983,
      "mean_token_accuracy": 0.16853295415639877,
      "num_tokens": 24955325.0,
      "step": 13530
    },
    {
      "entropy": 5.646707582473755,
      "epoch": 1.1371140516698173,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004875923958894295,
      "loss": 5.2837,
      "mean_token_accuracy": 0.17064532786607742,
      "num_tokens": 24964028.0,
      "step": 13535
    },
    {
      "entropy": 5.666370153427124,
      "epoch": 1.137534131484982,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048758258684978846,
      "loss": 5.4713,
      "mean_token_accuracy": 0.16217924058437347,
      "num_tokens": 24972923.0,
      "step": 13540
    },
    {
      "entropy": 5.651957702636719,
      "epoch": 1.137954211300147,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048757277404435636,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1657715320587158,
      "num_tokens": 24982156.0,
      "step": 13545
    },
    {
      "entropy": 5.664628601074218,
      "epoch": 1.1383742911153119,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000487562957473307,
      "loss": 5.418,
      "mean_token_accuracy": 0.16670742779970169,
      "num_tokens": 24991616.0,
      "step": 13550
    },
    {
      "entropy": 5.60991358757019,
      "epoch": 1.1387943709304769,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004875531371368144,
      "loss": 5.4697,
      "mean_token_accuracy": 0.16192520707845687,
      "num_tokens": 25001140.0,
      "step": 13555
    },
    {
      "entropy": 5.696591901779175,
      "epoch": 1.1392144507456416,
      "grad_norm": 1.75,
      "learning_rate": 0.00048754331303505236,
      "loss": 5.3951,
      "mean_token_accuracy": 0.16725025027990342,
      "num_tokens": 25010863.0,
      "step": 13560
    },
    {
      "entropy": 5.726091575622559,
      "epoch": 1.1396345305608064,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048753348516819496,
      "loss": 5.5024,
      "mean_token_accuracy": 0.15817077159881593,
      "num_tokens": 25019770.0,
      "step": 13565
    },
    {
      "entropy": 5.7528482437133786,
      "epoch": 1.1400546103759714,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004875236535364163,
      "loss": 5.5333,
      "mean_token_accuracy": 0.15655099377036094,
      "num_tokens": 25029900.0,
      "step": 13570
    },
    {
      "entropy": 5.781103563308716,
      "epoch": 1.1404746901911362,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004875138181398906,
      "loss": 5.4958,
      "mean_token_accuracy": 0.1637464240193367,
      "num_tokens": 25039428.0,
      "step": 13575
    },
    {
      "entropy": 5.688928604125977,
      "epoch": 1.1408947700063012,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000487503978978792,
      "loss": 5.4784,
      "mean_token_accuracy": 0.15997918397188188,
      "num_tokens": 25049145.0,
      "step": 13580
    },
    {
      "entropy": 5.699268198013305,
      "epoch": 1.141314849821466,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048749413605329487,
      "loss": 5.5024,
      "mean_token_accuracy": 0.1619054026901722,
      "num_tokens": 25058772.0,
      "step": 13585
    },
    {
      "entropy": 5.69202241897583,
      "epoch": 1.141734929636631,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048748428936357346,
      "loss": 5.4063,
      "mean_token_accuracy": 0.16621732711791992,
      "num_tokens": 25067249.0,
      "step": 13590
    },
    {
      "entropy": 5.689562606811523,
      "epoch": 1.1421550094517958,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004874744389098024,
      "loss": 5.3995,
      "mean_token_accuracy": 0.1586827367544174,
      "num_tokens": 25076893.0,
      "step": 13595
    },
    {
      "entropy": 5.665656042098999,
      "epoch": 1.1425750892669608,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004874645846921559,
      "loss": 5.3994,
      "mean_token_accuracy": 0.1694393739104271,
      "num_tokens": 25086238.0,
      "step": 13600
    },
    {
      "entropy": 5.644594764709472,
      "epoch": 1.1429951690821256,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048745472671080884,
      "loss": 5.4037,
      "mean_token_accuracy": 0.16155755817890166,
      "num_tokens": 25095334.0,
      "step": 13605
    },
    {
      "entropy": 5.604833793640137,
      "epoch": 1.1434152488972904,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048744486496593565,
      "loss": 5.404,
      "mean_token_accuracy": 0.1675198942422867,
      "num_tokens": 25104136.0,
      "step": 13610
    },
    {
      "entropy": 5.678550148010254,
      "epoch": 1.1438353287124554,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000487434999457711,
      "loss": 5.3981,
      "mean_token_accuracy": 0.17361067086458207,
      "num_tokens": 25112629.0,
      "step": 13615
    },
    {
      "entropy": 5.709067010879517,
      "epoch": 1.1442554085276202,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004874251301863098,
      "loss": 5.4307,
      "mean_token_accuracy": 0.15858973413705826,
      "num_tokens": 25121014.0,
      "step": 13620
    },
    {
      "entropy": 5.631903171539307,
      "epoch": 1.1446754883427852,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048741525715190675,
      "loss": 5.4706,
      "mean_token_accuracy": 0.16206472367048264,
      "num_tokens": 25130097.0,
      "step": 13625
    },
    {
      "entropy": 5.711316680908203,
      "epoch": 1.14509556815795,
      "grad_norm": 1.25,
      "learning_rate": 0.0004874053803546769,
      "loss": 5.4696,
      "mean_token_accuracy": 0.16349728405475616,
      "num_tokens": 25139065.0,
      "step": 13630
    },
    {
      "entropy": 5.652597141265869,
      "epoch": 1.1455156479731148,
      "grad_norm": 1.625,
      "learning_rate": 0.000487395499794795,
      "loss": 5.4533,
      "mean_token_accuracy": 0.16656357645988465,
      "num_tokens": 25148852.0,
      "step": 13635
    },
    {
      "entropy": 5.617219066619873,
      "epoch": 1.1459357277882798,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004873856154724362,
      "loss": 5.3506,
      "mean_token_accuracy": 0.1754961669445038,
      "num_tokens": 25157580.0,
      "step": 13640
    },
    {
      "entropy": 5.680072975158692,
      "epoch": 1.1463558076034446,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004873757273877756,
      "loss": 5.4561,
      "mean_token_accuracy": 0.163452672958374,
      "num_tokens": 25166243.0,
      "step": 13645
    },
    {
      "entropy": 5.710014724731446,
      "epoch": 1.1467758874186096,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048736583554098836,
      "loss": 5.467,
      "mean_token_accuracy": 0.16175027936697006,
      "num_tokens": 25174674.0,
      "step": 13650
    },
    {
      "entropy": 5.641525459289551,
      "epoch": 1.1471959672337744,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048735593993224973,
      "loss": 5.3815,
      "mean_token_accuracy": 0.1715010792016983,
      "num_tokens": 25183892.0,
      "step": 13655
    },
    {
      "entropy": 5.654358148574829,
      "epoch": 1.1476160470489394,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048734604056173495,
      "loss": 5.4056,
      "mean_token_accuracy": 0.16546755135059357,
      "num_tokens": 25192731.0,
      "step": 13660
    },
    {
      "entropy": 5.659243631362915,
      "epoch": 1.1480361268641042,
      "grad_norm": 1.625,
      "learning_rate": 0.00048733613742961933,
      "loss": 5.5014,
      "mean_token_accuracy": 0.1672609716653824,
      "num_tokens": 25201280.0,
      "step": 13665
    },
    {
      "entropy": 5.653694009780883,
      "epoch": 1.1484562066792692,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048732623053607846,
      "loss": 5.3927,
      "mean_token_accuracy": 0.16543423086404802,
      "num_tokens": 25209929.0,
      "step": 13670
    },
    {
      "entropy": 5.695533609390258,
      "epoch": 1.148876286494434,
      "grad_norm": 1.625,
      "learning_rate": 0.0004873163198812877,
      "loss": 5.3275,
      "mean_token_accuracy": 0.17144448906183243,
      "num_tokens": 25218583.0,
      "step": 13675
    },
    {
      "entropy": 5.738392496109009,
      "epoch": 1.1492963663095987,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004873064054654227,
      "loss": 5.5484,
      "mean_token_accuracy": 0.1579354114830494,
      "num_tokens": 25228949.0,
      "step": 13680
    },
    {
      "entropy": 5.683894395828247,
      "epoch": 1.1497164461247638,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048729648728865904,
      "loss": 5.3731,
      "mean_token_accuracy": 0.1767183229327202,
      "num_tokens": 25238603.0,
      "step": 13685
    },
    {
      "entropy": 5.71501407623291,
      "epoch": 1.1501365259399285,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048728656535117237,
      "loss": 5.4992,
      "mean_token_accuracy": 0.15357865989208222,
      "num_tokens": 25248265.0,
      "step": 13690
    },
    {
      "entropy": 5.643803834915161,
      "epoch": 1.1505566057550936,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004872766396531386,
      "loss": 5.484,
      "mean_token_accuracy": 0.16071568578481674,
      "num_tokens": 25258195.0,
      "step": 13695
    },
    {
      "entropy": 5.702970218658447,
      "epoch": 1.1509766855702583,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048726671019473335,
      "loss": 5.4422,
      "mean_token_accuracy": 0.16466852724552156,
      "num_tokens": 25267886.0,
      "step": 13700
    },
    {
      "entropy": 5.746435308456421,
      "epoch": 1.1513967653854231,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048725677697613267,
      "loss": 5.4682,
      "mean_token_accuracy": 0.16264414340257644,
      "num_tokens": 25277304.0,
      "step": 13705
    },
    {
      "entropy": 5.713738203048706,
      "epoch": 1.1518168452005881,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004872468399975125,
      "loss": 5.4881,
      "mean_token_accuracy": 0.15376320034265517,
      "num_tokens": 25286771.0,
      "step": 13710
    },
    {
      "entropy": 5.780117321014404,
      "epoch": 1.152236925015753,
      "grad_norm": 4.0625,
      "learning_rate": 0.00048723689925904884,
      "loss": 5.5313,
      "mean_token_accuracy": 0.1592633917927742,
      "num_tokens": 25296018.0,
      "step": 13715
    },
    {
      "entropy": 5.741491746902466,
      "epoch": 1.152657004830918,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004872269547609179,
      "loss": 5.4953,
      "mean_token_accuracy": 0.16684821844100953,
      "num_tokens": 25305737.0,
      "step": 13720
    },
    {
      "entropy": 5.593285465240479,
      "epoch": 1.1530770846460827,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004872170065032956,
      "loss": 5.3045,
      "mean_token_accuracy": 0.16534190028905868,
      "num_tokens": 25314625.0,
      "step": 13725
    },
    {
      "entropy": 5.624049663543701,
      "epoch": 1.1534971644612477,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004872070544863584,
      "loss": 5.4599,
      "mean_token_accuracy": 0.16450536549091338,
      "num_tokens": 25323453.0,
      "step": 13730
    },
    {
      "entropy": 5.670636749267578,
      "epoch": 1.1539172442764125,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004871970987102824,
      "loss": 5.4532,
      "mean_token_accuracy": 0.16838308423757553,
      "num_tokens": 25333236.0,
      "step": 13735
    },
    {
      "entropy": 5.692652988433838,
      "epoch": 1.1543373240915775,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004871871391752442,
      "loss": 5.3671,
      "mean_token_accuracy": 0.16096194535493852,
      "num_tokens": 25341993.0,
      "step": 13740
    },
    {
      "entropy": 5.680402231216431,
      "epoch": 1.1547574039067423,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048717717588141993,
      "loss": 5.4006,
      "mean_token_accuracy": 0.1673072651028633,
      "num_tokens": 25350695.0,
      "step": 13745
    },
    {
      "entropy": 5.732087278366089,
      "epoch": 1.155177483721907,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004871672088289863,
      "loss": 5.4491,
      "mean_token_accuracy": 0.16335247606039047,
      "num_tokens": 25359044.0,
      "step": 13750
    },
    {
      "entropy": 5.652139139175415,
      "epoch": 1.155597563537072,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048715723801811986,
      "loss": 5.4682,
      "mean_token_accuracy": 0.1620037004351616,
      "num_tokens": 25367959.0,
      "step": 13755
    },
    {
      "entropy": 5.666346216201783,
      "epoch": 1.156017643352237,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048714726344899716,
      "loss": 5.471,
      "mean_token_accuracy": 0.16407882869243623,
      "num_tokens": 25376968.0,
      "step": 13760
    },
    {
      "entropy": 5.6227333545684814,
      "epoch": 1.156437723167402,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004871372851217949,
      "loss": 5.3543,
      "mean_token_accuracy": 0.1680816724896431,
      "num_tokens": 25385381.0,
      "step": 13765
    },
    {
      "entropy": 5.7028580665588375,
      "epoch": 1.1568578029825667,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004871273030366899,
      "loss": 5.4631,
      "mean_token_accuracy": 0.1649225354194641,
      "num_tokens": 25394647.0,
      "step": 13770
    },
    {
      "entropy": 5.648839139938355,
      "epoch": 1.1572778827977315,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004871173171938589,
      "loss": 5.4319,
      "mean_token_accuracy": 0.17046377956867217,
      "num_tokens": 25403973.0,
      "step": 13775
    },
    {
      "entropy": 5.6703736782073975,
      "epoch": 1.1576979626128965,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004871073275934789,
      "loss": 5.3984,
      "mean_token_accuracy": 0.1691466361284256,
      "num_tokens": 25412319.0,
      "step": 13780
    },
    {
      "entropy": 5.643203210830689,
      "epoch": 1.1581180424280613,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048709733423572685,
      "loss": 5.4302,
      "mean_token_accuracy": 0.16623932421207427,
      "num_tokens": 25420558.0,
      "step": 13785
    },
    {
      "entropy": 5.591461992263794,
      "epoch": 1.1585381222432263,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048708733712077973,
      "loss": 5.3857,
      "mean_token_accuracy": 0.16710084825754165,
      "num_tokens": 25429258.0,
      "step": 13790
    },
    {
      "entropy": 5.708824634552002,
      "epoch": 1.158958202058391,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004870773362488146,
      "loss": 5.3492,
      "mean_token_accuracy": 0.17254966199398042,
      "num_tokens": 25438005.0,
      "step": 13795
    },
    {
      "entropy": 5.66907844543457,
      "epoch": 1.159378281873556,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004870673316200087,
      "loss": 5.3715,
      "mean_token_accuracy": 0.1682361513376236,
      "num_tokens": 25447120.0,
      "step": 13800
    },
    {
      "entropy": 5.595958423614502,
      "epoch": 1.1597983616887209,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004870573232345392,
      "loss": 5.3666,
      "mean_token_accuracy": 0.16778069287538527,
      "num_tokens": 25456216.0,
      "step": 13805
    },
    {
      "entropy": 5.823038816452026,
      "epoch": 1.1602184415038856,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004870473110925834,
      "loss": 5.6262,
      "mean_token_accuracy": 0.1571579709649086,
      "num_tokens": 25466456.0,
      "step": 13810
    },
    {
      "entropy": 5.6383460521697994,
      "epoch": 1.1606385213190507,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004870372951943187,
      "loss": 5.293,
      "mean_token_accuracy": 0.17360034435987473,
      "num_tokens": 25475217.0,
      "step": 13815
    },
    {
      "entropy": 5.673424625396729,
      "epoch": 1.1610586011342154,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048702727553992243,
      "loss": 5.5869,
      "mean_token_accuracy": 0.15147149562835693,
      "num_tokens": 25484617.0,
      "step": 13820
    },
    {
      "entropy": 5.6951432704925535,
      "epoch": 1.1614786809493804,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048701725212957223,
      "loss": 5.3764,
      "mean_token_accuracy": 0.17048204690217972,
      "num_tokens": 25493936.0,
      "step": 13825
    },
    {
      "entropy": 5.621255588531494,
      "epoch": 1.1618987607645452,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004870072249634455,
      "loss": 5.3668,
      "mean_token_accuracy": 0.16855009347200395,
      "num_tokens": 25502306.0,
      "step": 13830
    },
    {
      "entropy": 5.569633483886719,
      "epoch": 1.1623188405797102,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048699719404172006,
      "loss": 5.4267,
      "mean_token_accuracy": 0.16455204039812088,
      "num_tokens": 25511247.0,
      "step": 13835
    },
    {
      "entropy": 5.657006931304932,
      "epoch": 1.162738920394875,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048698715936457344,
      "loss": 5.4694,
      "mean_token_accuracy": 0.16329195201396943,
      "num_tokens": 25520482.0,
      "step": 13840
    },
    {
      "entropy": 5.640268325805664,
      "epoch": 1.1631590002100398,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048697712093218336,
      "loss": 5.3744,
      "mean_token_accuracy": 0.16461358666419984,
      "num_tokens": 25529854.0,
      "step": 13845
    },
    {
      "entropy": 5.623713779449463,
      "epoch": 1.1635790800252048,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004869670787447279,
      "loss": 5.3002,
      "mean_token_accuracy": 0.16914840638637543,
      "num_tokens": 25538251.0,
      "step": 13850
    },
    {
      "entropy": 5.638497924804687,
      "epoch": 1.1639991598403696,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004869570328023846,
      "loss": 5.3871,
      "mean_token_accuracy": 0.16646116971969604,
      "num_tokens": 25546889.0,
      "step": 13855
    },
    {
      "entropy": 5.594804191589356,
      "epoch": 1.1644192396555346,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048694698310533177,
      "loss": 5.4603,
      "mean_token_accuracy": 0.16453583240509034,
      "num_tokens": 25557040.0,
      "step": 13860
    },
    {
      "entropy": 5.753741073608398,
      "epoch": 1.1648393194706994,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004869369296537472,
      "loss": 5.609,
      "mean_token_accuracy": 0.15437615364789964,
      "num_tokens": 25565798.0,
      "step": 13865
    },
    {
      "entropy": 5.787939405441284,
      "epoch": 1.1652593992858642,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004869268724478091,
      "loss": 5.4344,
      "mean_token_accuracy": 0.16476142406463623,
      "num_tokens": 25575039.0,
      "step": 13870
    },
    {
      "entropy": 5.709807395935059,
      "epoch": 1.1656794791010292,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048691681148769545,
      "loss": 5.447,
      "mean_token_accuracy": 0.16519313603639602,
      "num_tokens": 25584635.0,
      "step": 13875
    },
    {
      "entropy": 5.570197010040284,
      "epoch": 1.166099558916194,
      "grad_norm": 2.0,
      "learning_rate": 0.0004869067467735847,
      "loss": 5.3732,
      "mean_token_accuracy": 0.16737187206745147,
      "num_tokens": 25593736.0,
      "step": 13880
    },
    {
      "entropy": 5.625331687927246,
      "epoch": 1.166519638731359,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004868966783056551,
      "loss": 5.3503,
      "mean_token_accuracy": 0.17617221772670746,
      "num_tokens": 25602685.0,
      "step": 13885
    },
    {
      "entropy": 5.6361452579498295,
      "epoch": 1.1669397185465238,
      "grad_norm": 1.890625,
      "learning_rate": 0.00048688660608408484,
      "loss": 5.4137,
      "mean_token_accuracy": 0.16371521204710007,
      "num_tokens": 25610690.0,
      "step": 13890
    },
    {
      "entropy": 5.572535657882691,
      "epoch": 1.1673597983616888,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048687653010905254,
      "loss": 5.3114,
      "mean_token_accuracy": 0.17053646743297576,
      "num_tokens": 25619805.0,
      "step": 13895
    },
    {
      "entropy": 5.698605394363403,
      "epoch": 1.1677798781768536,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048686645038073664,
      "loss": 5.5349,
      "mean_token_accuracy": 0.15510550141334534,
      "num_tokens": 25629447.0,
      "step": 13900
    },
    {
      "entropy": 5.669313621520996,
      "epoch": 1.1681999579920186,
      "grad_norm": 1.75,
      "learning_rate": 0.00048685636689931554,
      "loss": 5.3748,
      "mean_token_accuracy": 0.1702971413731575,
      "num_tokens": 25638619.0,
      "step": 13905
    },
    {
      "entropy": 5.711242389678955,
      "epoch": 1.1686200378071834,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048684627966496803,
      "loss": 5.4614,
      "mean_token_accuracy": 0.1624944031238556,
      "num_tokens": 25648255.0,
      "step": 13910
    },
    {
      "entropy": 5.66675066947937,
      "epoch": 1.1690401176223482,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048683618867787284,
      "loss": 5.5036,
      "mean_token_accuracy": 0.16320302784442903,
      "num_tokens": 25657881.0,
      "step": 13915
    },
    {
      "entropy": 5.758802509307861,
      "epoch": 1.1694601974375132,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004868260939382086,
      "loss": 5.5189,
      "mean_token_accuracy": 0.16214758455753325,
      "num_tokens": 25666773.0,
      "step": 13920
    },
    {
      "entropy": 5.709087944030761,
      "epoch": 1.169880277252678,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004868159954461542,
      "loss": 5.4196,
      "mean_token_accuracy": 0.1658247232437134,
      "num_tokens": 25675152.0,
      "step": 13925
    },
    {
      "entropy": 5.730292272567749,
      "epoch": 1.170300357067843,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048680589320188847,
      "loss": 5.5228,
      "mean_token_accuracy": 0.15880742371082307,
      "num_tokens": 25684962.0,
      "step": 13930
    },
    {
      "entropy": 5.679729652404785,
      "epoch": 1.1707204368830078,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004867957872055904,
      "loss": 5.4287,
      "mean_token_accuracy": 0.16511483937501908,
      "num_tokens": 25693782.0,
      "step": 13935
    },
    {
      "entropy": 5.641356992721557,
      "epoch": 1.1711405166981725,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048678567745743905,
      "loss": 5.3885,
      "mean_token_accuracy": 0.17264550924301147,
      "num_tokens": 25703081.0,
      "step": 13940
    },
    {
      "entropy": 5.658506870269775,
      "epoch": 1.1715605965133375,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004867755639576135,
      "loss": 5.4104,
      "mean_token_accuracy": 0.16793930828571318,
      "num_tokens": 25711628.0,
      "step": 13945
    },
    {
      "entropy": 5.640485239028931,
      "epoch": 1.1719806763285023,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004867654467062928,
      "loss": 5.4241,
      "mean_token_accuracy": 0.16928267627954482,
      "num_tokens": 25720676.0,
      "step": 13950
    },
    {
      "entropy": 5.649695205688476,
      "epoch": 1.1724007561436673,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048675532570365633,
      "loss": 5.392,
      "mean_token_accuracy": 0.16928528249263763,
      "num_tokens": 25729920.0,
      "step": 13955
    },
    {
      "entropy": 5.625086498260498,
      "epoch": 1.1728208359588321,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048674520094988327,
      "loss": 5.3654,
      "mean_token_accuracy": 0.17307248711585999,
      "num_tokens": 25739745.0,
      "step": 13960
    },
    {
      "entropy": 5.627538824081421,
      "epoch": 1.1732409157739971,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048673507244515303,
      "loss": 5.3779,
      "mean_token_accuracy": 0.16306281685829163,
      "num_tokens": 25748636.0,
      "step": 13965
    },
    {
      "entropy": 5.715080642700196,
      "epoch": 1.173660995589162,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000486724940189645,
      "loss": 5.5204,
      "mean_token_accuracy": 0.1618019700050354,
      "num_tokens": 25758393.0,
      "step": 13970
    },
    {
      "entropy": 5.68854718208313,
      "epoch": 1.174081075404327,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004867148041835386,
      "loss": 5.5155,
      "mean_token_accuracy": 0.1531493306159973,
      "num_tokens": 25768520.0,
      "step": 13975
    },
    {
      "entropy": 5.570581007003784,
      "epoch": 1.1745011552194917,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004867046644270136,
      "loss": 5.3033,
      "mean_token_accuracy": 0.16933976262807846,
      "num_tokens": 25777168.0,
      "step": 13980
    },
    {
      "entropy": 5.755339241027832,
      "epoch": 1.1749212350346565,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004866945209202494,
      "loss": 5.6177,
      "mean_token_accuracy": 0.14852596670389176,
      "num_tokens": 25787042.0,
      "step": 13985
    },
    {
      "entropy": 5.755776023864746,
      "epoch": 1.1753413148498215,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004866843736634258,
      "loss": 5.4891,
      "mean_token_accuracy": 0.16482731252908706,
      "num_tokens": 25796784.0,
      "step": 13990
    },
    {
      "entropy": 5.821707010269165,
      "epoch": 1.1757613946649863,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004866742226567225,
      "loss": 5.5454,
      "mean_token_accuracy": 0.1584312066435814,
      "num_tokens": 25806285.0,
      "step": 13995
    },
    {
      "entropy": 5.697404479980468,
      "epoch": 1.1761814744801513,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048666406790031936,
      "loss": 5.3646,
      "mean_token_accuracy": 0.16209536790847778,
      "num_tokens": 25814889.0,
      "step": 14000
    },
    {
      "entropy": 5.6022241592407225,
      "epoch": 1.176601554295316,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004866539093943962,
      "loss": 5.4378,
      "mean_token_accuracy": 0.1663196012377739,
      "num_tokens": 25824551.0,
      "step": 14005
    },
    {
      "entropy": 5.756335592269897,
      "epoch": 1.1770216341104809,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048664374713913304,
      "loss": 5.5054,
      "mean_token_accuracy": 0.15789050087332726,
      "num_tokens": 25834482.0,
      "step": 14010
    },
    {
      "entropy": 5.777653217315674,
      "epoch": 1.177441713925646,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004866335811347099,
      "loss": 5.5015,
      "mean_token_accuracy": 0.15598200112581254,
      "num_tokens": 25843274.0,
      "step": 14015
    },
    {
      "entropy": 5.761362743377686,
      "epoch": 1.1778617937408107,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00048662341138130683,
      "loss": 5.4877,
      "mean_token_accuracy": 0.158286252617836,
      "num_tokens": 25852482.0,
      "step": 14020
    },
    {
      "entropy": 5.748376035690308,
      "epoch": 1.1782818735559757,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048661323787910405,
      "loss": 5.4725,
      "mean_token_accuracy": 0.1600512832403183,
      "num_tokens": 25862657.0,
      "step": 14025
    },
    {
      "entropy": 5.64421010017395,
      "epoch": 1.1787019533711405,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004866030606282817,
      "loss": 5.4185,
      "mean_token_accuracy": 0.1709764003753662,
      "num_tokens": 25871492.0,
      "step": 14030
    },
    {
      "entropy": 5.638793468475342,
      "epoch": 1.1791220331863055,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048659287962902006,
      "loss": 5.4203,
      "mean_token_accuracy": 0.1675298035144806,
      "num_tokens": 25880979.0,
      "step": 14035
    },
    {
      "entropy": 5.696149921417236,
      "epoch": 1.1795421130014703,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048658269488149945,
      "loss": 5.4218,
      "mean_token_accuracy": 0.1625008463859558,
      "num_tokens": 25891060.0,
      "step": 14040
    },
    {
      "entropy": 5.761597537994385,
      "epoch": 1.1799621928166353,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004865725063859005,
      "loss": 5.544,
      "mean_token_accuracy": 0.16607259213924408,
      "num_tokens": 25900421.0,
      "step": 14045
    },
    {
      "entropy": 5.653324842453003,
      "epoch": 1.1803822726318,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048656231414240345,
      "loss": 5.4313,
      "mean_token_accuracy": 0.1618879944086075,
      "num_tokens": 25909614.0,
      "step": 14050
    },
    {
      "entropy": 5.616987228393555,
      "epoch": 1.1808023524469649,
      "grad_norm": 1.421875,
      "learning_rate": 0.000486552118151189,
      "loss": 5.4679,
      "mean_token_accuracy": 0.15664371848106384,
      "num_tokens": 25919324.0,
      "step": 14055
    },
    {
      "entropy": 5.717600774765015,
      "epoch": 1.1812224322621299,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048654191841243763,
      "loss": 5.4666,
      "mean_token_accuracy": 0.16498553603887559,
      "num_tokens": 25928818.0,
      "step": 14060
    },
    {
      "entropy": 5.759012937545776,
      "epoch": 1.1816425120772946,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004865317149263301,
      "loss": 5.5292,
      "mean_token_accuracy": 0.1587470844388008,
      "num_tokens": 25938148.0,
      "step": 14065
    },
    {
      "entropy": 5.616943883895874,
      "epoch": 1.1820625918924597,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004865215076930473,
      "loss": 5.4301,
      "mean_token_accuracy": 0.1658810332417488,
      "num_tokens": 25947210.0,
      "step": 14070
    },
    {
      "entropy": 5.620659303665161,
      "epoch": 1.1824826717076244,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004865112967127697,
      "loss": 5.4247,
      "mean_token_accuracy": 0.16346285492181778,
      "num_tokens": 25955949.0,
      "step": 14075
    },
    {
      "entropy": 5.575088262557983,
      "epoch": 1.1829027515227892,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004865010819856786,
      "loss": 5.3599,
      "mean_token_accuracy": 0.16429171711206436,
      "num_tokens": 25964193.0,
      "step": 14080
    },
    {
      "entropy": 5.644189167022705,
      "epoch": 1.1833228313379542,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004864908635119546,
      "loss": 5.4325,
      "mean_token_accuracy": 0.16631924957036973,
      "num_tokens": 25973141.0,
      "step": 14085
    },
    {
      "entropy": 5.723185777664185,
      "epoch": 1.183742911153119,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004864806412917788,
      "loss": 5.5128,
      "mean_token_accuracy": 0.16188513338565827,
      "num_tokens": 25982650.0,
      "step": 14090
    },
    {
      "entropy": 5.752957391738891,
      "epoch": 1.184162990968284,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004864704153253325,
      "loss": 5.5041,
      "mean_token_accuracy": 0.15529987812042237,
      "num_tokens": 25992096.0,
      "step": 14095
    },
    {
      "entropy": 5.73392686843872,
      "epoch": 1.1845830707834488,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048646018561279665,
      "loss": 5.4828,
      "mean_token_accuracy": 0.16362193524837493,
      "num_tokens": 26002063.0,
      "step": 14100
    },
    {
      "entropy": 5.537550973892212,
      "epoch": 1.1850031505986138,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048644995215435245,
      "loss": 5.3166,
      "mean_token_accuracy": 0.17091075628995894,
      "num_tokens": 26010716.0,
      "step": 14105
    },
    {
      "entropy": 5.664225101470947,
      "epoch": 1.1854232304137786,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004864397149501812,
      "loss": 5.3975,
      "mean_token_accuracy": 0.1652759626507759,
      "num_tokens": 26019136.0,
      "step": 14110
    },
    {
      "entropy": 5.679570627212525,
      "epoch": 1.1858433102289434,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048642947400046434,
      "loss": 5.4205,
      "mean_token_accuracy": 0.17468804866075516,
      "num_tokens": 26028029.0,
      "step": 14115
    },
    {
      "entropy": 5.725770425796509,
      "epoch": 1.1862633900441084,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048641922930538325,
      "loss": 5.578,
      "mean_token_accuracy": 0.15540485978126525,
      "num_tokens": 26038025.0,
      "step": 14120
    },
    {
      "entropy": 5.6925132274627686,
      "epoch": 1.1866834698592732,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004864089808651193,
      "loss": 5.5527,
      "mean_token_accuracy": 0.1457514375448227,
      "num_tokens": 26048427.0,
      "step": 14125
    },
    {
      "entropy": 5.681820917129516,
      "epoch": 1.1871035496744382,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004863987286798541,
      "loss": 5.3589,
      "mean_token_accuracy": 0.1656435549259186,
      "num_tokens": 26057682.0,
      "step": 14130
    },
    {
      "entropy": 5.619316673278808,
      "epoch": 1.187523629489603,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004863884727497693,
      "loss": 5.421,
      "mean_token_accuracy": 0.16528261303901673,
      "num_tokens": 26066562.0,
      "step": 14135
    },
    {
      "entropy": 5.603818464279175,
      "epoch": 1.187943709304768,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004863782130750466,
      "loss": 5.3489,
      "mean_token_accuracy": 0.16920543015003203,
      "num_tokens": 26075633.0,
      "step": 14140
    },
    {
      "entropy": 5.7123847007751465,
      "epoch": 1.1883637891199328,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048636794965586764,
      "loss": 5.5142,
      "mean_token_accuracy": 0.1616382658481598,
      "num_tokens": 26085160.0,
      "step": 14145
    },
    {
      "entropy": 5.671110486984253,
      "epoch": 1.1887838689350976,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048635768249241434,
      "loss": 5.3997,
      "mean_token_accuracy": 0.1647752344608307,
      "num_tokens": 26094157.0,
      "step": 14150
    },
    {
      "entropy": 5.748544454574585,
      "epoch": 1.1892039487502626,
      "grad_norm": 1.75,
      "learning_rate": 0.0004863474115848685,
      "loss": 5.5073,
      "mean_token_accuracy": 0.16841865181922913,
      "num_tokens": 26104459.0,
      "step": 14155
    },
    {
      "entropy": 5.6415163516998295,
      "epoch": 1.1896240285654274,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048633713693341214,
      "loss": 5.4513,
      "mean_token_accuracy": 0.16421539783477784,
      "num_tokens": 26114468.0,
      "step": 14160
    },
    {
      "entropy": 5.602667903900146,
      "epoch": 1.1900441083805924,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048632685853822714,
      "loss": 5.4233,
      "mean_token_accuracy": 0.16077221781015397,
      "num_tokens": 26123408.0,
      "step": 14165
    },
    {
      "entropy": 5.65820803642273,
      "epoch": 1.1904641881957572,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004863165763994957,
      "loss": 5.4277,
      "mean_token_accuracy": 0.16007840856909752,
      "num_tokens": 26132692.0,
      "step": 14170
    },
    {
      "entropy": 5.831601476669311,
      "epoch": 1.190884268010922,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004863062905173999,
      "loss": 5.6045,
      "mean_token_accuracy": 0.15579294115304948,
      "num_tokens": 26142259.0,
      "step": 14175
    },
    {
      "entropy": 5.72370457649231,
      "epoch": 1.191304347826087,
      "grad_norm": 1.515625,
      "learning_rate": 0.000486296000892122,
      "loss": 5.457,
      "mean_token_accuracy": 0.15974125415086746,
      "num_tokens": 26151782.0,
      "step": 14180
    },
    {
      "entropy": 5.591598129272461,
      "epoch": 1.1917244276412517,
      "grad_norm": 4.0625,
      "learning_rate": 0.00048628570752384424,
      "loss": 5.2925,
      "mean_token_accuracy": 0.17002766877412795,
      "num_tokens": 26160449.0,
      "step": 14185
    },
    {
      "entropy": 5.6798159122467045,
      "epoch": 1.1921445074564168,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048627541041274897,
      "loss": 5.5294,
      "mean_token_accuracy": 0.15847962498664855,
      "num_tokens": 26169764.0,
      "step": 14190
    },
    {
      "entropy": 5.6722887516021725,
      "epoch": 1.1925645872715815,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048626510955901854,
      "loss": 5.388,
      "mean_token_accuracy": 0.16532097309827803,
      "num_tokens": 26178759.0,
      "step": 14195
    },
    {
      "entropy": 5.752083778381348,
      "epoch": 1.1929846670867466,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004862548049628356,
      "loss": 5.5175,
      "mean_token_accuracy": 0.1655479848384857,
      "num_tokens": 26187904.0,
      "step": 14200
    },
    {
      "entropy": 5.709446907043457,
      "epoch": 1.1934047469019113,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004862444966243824,
      "loss": 5.4362,
      "mean_token_accuracy": 0.16655133068561553,
      "num_tokens": 26196563.0,
      "step": 14205
    },
    {
      "entropy": 5.716876649856568,
      "epoch": 1.1938248267170763,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004862341845438419,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16468843668699265,
      "num_tokens": 26206573.0,
      "step": 14210
    },
    {
      "entropy": 5.640651273727417,
      "epoch": 1.1942449065322411,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048622386872139645,
      "loss": 5.3703,
      "mean_token_accuracy": 0.16637713611125945,
      "num_tokens": 26215308.0,
      "step": 14215
    },
    {
      "entropy": 5.623022794723511,
      "epoch": 1.194664986347406,
      "grad_norm": 1.5,
      "learning_rate": 0.000486213549157229,
      "loss": 5.4308,
      "mean_token_accuracy": 0.1651561066508293,
      "num_tokens": 26224379.0,
      "step": 14220
    },
    {
      "entropy": 5.606683444976807,
      "epoch": 1.195085066162571,
      "grad_norm": 2.125,
      "learning_rate": 0.0004862032258515222,
      "loss": 5.4237,
      "mean_token_accuracy": 0.1671927824616432,
      "num_tokens": 26233620.0,
      "step": 14225
    },
    {
      "entropy": 5.685428619384766,
      "epoch": 1.1955051459777357,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004861928988044592,
      "loss": 5.4885,
      "mean_token_accuracy": 0.16064939498901368,
      "num_tokens": 26242556.0,
      "step": 14230
    },
    {
      "entropy": 5.704376649856568,
      "epoch": 1.1959252257929007,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004861825680162226,
      "loss": 5.4799,
      "mean_token_accuracy": 0.16530386954545975,
      "num_tokens": 26251561.0,
      "step": 14235
    },
    {
      "entropy": 5.671692752838135,
      "epoch": 1.1963453056080655,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048617223348699546,
      "loss": 5.395,
      "mean_token_accuracy": 0.1631077140569687,
      "num_tokens": 26261115.0,
      "step": 14240
    },
    {
      "entropy": 5.7645580768585205,
      "epoch": 1.1967653854232303,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004861618952169611,
      "loss": 5.5554,
      "mean_token_accuracy": 0.1650903783738613,
      "num_tokens": 26271165.0,
      "step": 14245
    },
    {
      "entropy": 5.668324375152588,
      "epoch": 1.1971854652383953,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004861515532063025,
      "loss": 5.5081,
      "mean_token_accuracy": 0.16101304292678834,
      "num_tokens": 26280822.0,
      "step": 14250
    },
    {
      "entropy": 5.678185224533081,
      "epoch": 1.19760554505356,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048614120745520275,
      "loss": 5.3902,
      "mean_token_accuracy": 0.16933465003967285,
      "num_tokens": 26288747.0,
      "step": 14255
    },
    {
      "entropy": 5.661726713180542,
      "epoch": 1.198025624868725,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048613085796384524,
      "loss": 5.4684,
      "mean_token_accuracy": 0.1568397268652916,
      "num_tokens": 26298387.0,
      "step": 14260
    },
    {
      "entropy": 5.607639598846435,
      "epoch": 1.19844570468389,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048612050473241335,
      "loss": 5.3588,
      "mean_token_accuracy": 0.16898838877677919,
      "num_tokens": 26307016.0,
      "step": 14265
    },
    {
      "entropy": 5.663268566131592,
      "epoch": 1.198865784499055,
      "grad_norm": 1.375,
      "learning_rate": 0.0004861101477610905,
      "loss": 5.4788,
      "mean_token_accuracy": 0.1621081128716469,
      "num_tokens": 26316296.0,
      "step": 14270
    },
    {
      "entropy": 5.6753627300262455,
      "epoch": 1.1992858643142197,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048609978705006,
      "loss": 5.464,
      "mean_token_accuracy": 0.16156177669763566,
      "num_tokens": 26325525.0,
      "step": 14275
    },
    {
      "entropy": 5.665592432022095,
      "epoch": 1.1997059441293847,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004860894225995055,
      "loss": 5.3402,
      "mean_token_accuracy": 0.16874586492776872,
      "num_tokens": 26334195.0,
      "step": 14280
    },
    {
      "entropy": 5.6383813381195065,
      "epoch": 1.2001260239445495,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048607905440961054,
      "loss": 5.4703,
      "mean_token_accuracy": 0.1640348732471466,
      "num_tokens": 26343933.0,
      "step": 14285
    },
    {
      "entropy": 5.629398059844971,
      "epoch": 1.2005461037597143,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048606868248055887,
      "loss": 5.422,
      "mean_token_accuracy": 0.16718345433473586,
      "num_tokens": 26353455.0,
      "step": 14290
    },
    {
      "entropy": 5.7174492359161375,
      "epoch": 1.2009661835748793,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004860583068125341,
      "loss": 5.4115,
      "mean_token_accuracy": 0.17318314611911773,
      "num_tokens": 26362662.0,
      "step": 14295
    },
    {
      "entropy": 5.647911453247071,
      "epoch": 1.201386263390044,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004860479274057202,
      "loss": 5.4264,
      "mean_token_accuracy": 0.16339855939149855,
      "num_tokens": 26371536.0,
      "step": 14300
    },
    {
      "entropy": 5.729653596878052,
      "epoch": 1.201806343205209,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048603754426030087,
      "loss": 5.52,
      "mean_token_accuracy": 0.16162212640047074,
      "num_tokens": 26381925.0,
      "step": 14305
    },
    {
      "entropy": 5.602517032623291,
      "epoch": 1.2022264230203739,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048602715737646016,
      "loss": 5.3863,
      "mean_token_accuracy": 0.16718171089887618,
      "num_tokens": 26391111.0,
      "step": 14310
    },
    {
      "entropy": 5.763671970367431,
      "epoch": 1.2026465028355386,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048601676675438197,
      "loss": 5.5468,
      "mean_token_accuracy": 0.1514241561293602,
      "num_tokens": 26401667.0,
      "step": 14315
    },
    {
      "entropy": 5.683077478408814,
      "epoch": 1.2030665826507037,
      "grad_norm": 1.5,
      "learning_rate": 0.00048600637239425045,
      "loss": 5.3705,
      "mean_token_accuracy": 0.1719566836953163,
      "num_tokens": 26411261.0,
      "step": 14320
    },
    {
      "entropy": 5.641879558563232,
      "epoch": 1.2034866624658684,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048599597429624966,
      "loss": 5.5082,
      "mean_token_accuracy": 0.16119082495570183,
      "num_tokens": 26419808.0,
      "step": 14325
    },
    {
      "entropy": 5.675163888931275,
      "epoch": 1.2039067422810334,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048598557246056385,
      "loss": 5.4305,
      "mean_token_accuracy": 0.16555157899856568,
      "num_tokens": 26429160.0,
      "step": 14330
    },
    {
      "entropy": 5.699215984344482,
      "epoch": 1.2043268220961982,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048597516688737727,
      "loss": 5.3938,
      "mean_token_accuracy": 0.16645348966121673,
      "num_tokens": 26437675.0,
      "step": 14335
    },
    {
      "entropy": 5.680204963684082,
      "epoch": 1.2047469019113632,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048596475757687425,
      "loss": 5.449,
      "mean_token_accuracy": 0.16452098041772842,
      "num_tokens": 26446317.0,
      "step": 14340
    },
    {
      "entropy": 5.699273586273193,
      "epoch": 1.205166981726528,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048595434452923915,
      "loss": 5.4949,
      "mean_token_accuracy": 0.16454333513975145,
      "num_tokens": 26456183.0,
      "step": 14345
    },
    {
      "entropy": 5.6539466857910154,
      "epoch": 1.205587061541693,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048594392774465656,
      "loss": 5.4311,
      "mean_token_accuracy": 0.15812044590711594,
      "num_tokens": 26466324.0,
      "step": 14350
    },
    {
      "entropy": 5.667297124862671,
      "epoch": 1.2060071413568578,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048593350722331074,
      "loss": 5.4471,
      "mean_token_accuracy": 0.16772420853376388,
      "num_tokens": 26475560.0,
      "step": 14355
    },
    {
      "entropy": 5.610403156280517,
      "epoch": 1.2064272211720226,
      "grad_norm": 1.875,
      "learning_rate": 0.00048592308296538654,
      "loss": 5.4173,
      "mean_token_accuracy": 0.16000428274273873,
      "num_tokens": 26484955.0,
      "step": 14360
    },
    {
      "entropy": 5.65930495262146,
      "epoch": 1.2068473009871876,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004859126549710686,
      "loss": 5.3811,
      "mean_token_accuracy": 0.1743646025657654,
      "num_tokens": 26494306.0,
      "step": 14365
    },
    {
      "entropy": 5.64382495880127,
      "epoch": 1.2072673808023524,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048590222324054153,
      "loss": 5.3942,
      "mean_token_accuracy": 0.1713266119360924,
      "num_tokens": 26503871.0,
      "step": 14370
    },
    {
      "entropy": 5.720830965042114,
      "epoch": 1.2076874606175174,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004858917877739901,
      "loss": 5.4942,
      "mean_token_accuracy": 0.16498788744211196,
      "num_tokens": 26511929.0,
      "step": 14375
    },
    {
      "entropy": 5.695744895935059,
      "epoch": 1.2081075404326822,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004858813485715994,
      "loss": 5.4881,
      "mean_token_accuracy": 0.15547695755958557,
      "num_tokens": 26520469.0,
      "step": 14380
    },
    {
      "entropy": 5.6886962890625,
      "epoch": 1.208527620247847,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004858709056335541,
      "loss": 5.458,
      "mean_token_accuracy": 0.16531693637371064,
      "num_tokens": 26530102.0,
      "step": 14385
    },
    {
      "entropy": 5.668837690353394,
      "epoch": 1.208947700063012,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048586045896003926,
      "loss": 5.4623,
      "mean_token_accuracy": 0.16233531534671783,
      "num_tokens": 26538705.0,
      "step": 14390
    },
    {
      "entropy": 5.7534499168396,
      "epoch": 1.2093677798781768,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004858500085512401,
      "loss": 5.5504,
      "mean_token_accuracy": 0.16140262633562089,
      "num_tokens": 26548315.0,
      "step": 14395
    },
    {
      "entropy": 5.734239721298218,
      "epoch": 1.2097878596933418,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048583955440734144,
      "loss": 5.3839,
      "mean_token_accuracy": 0.16452737748622895,
      "num_tokens": 26556412.0,
      "step": 14400
    },
    {
      "entropy": 5.757485628128052,
      "epoch": 1.2102079395085066,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048582909652852873,
      "loss": 5.5363,
      "mean_token_accuracy": 0.15494186282157899,
      "num_tokens": 26566146.0,
      "step": 14405
    },
    {
      "entropy": 5.666713428497315,
      "epoch": 1.2106280193236716,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004858186349149871,
      "loss": 5.4291,
      "mean_token_accuracy": 0.16879764646291734,
      "num_tokens": 26576019.0,
      "step": 14410
    },
    {
      "entropy": 5.547569131851196,
      "epoch": 1.2110480991388364,
      "grad_norm": 1.640625,
      "learning_rate": 0.000485808169566902,
      "loss": 5.2952,
      "mean_token_accuracy": 0.16971922367811204,
      "num_tokens": 26585461.0,
      "step": 14415
    },
    {
      "entropy": 5.549987030029297,
      "epoch": 1.2114681789540014,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048579770048445863,
      "loss": 5.3289,
      "mean_token_accuracy": 0.18390253633260728,
      "num_tokens": 26594021.0,
      "step": 14420
    },
    {
      "entropy": 5.748401546478272,
      "epoch": 1.2118882587691662,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048578722766784253,
      "loss": 5.481,
      "mean_token_accuracy": 0.16007982641458512,
      "num_tokens": 26602712.0,
      "step": 14425
    },
    {
      "entropy": 5.627417659759521,
      "epoch": 1.212308338584331,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048577675111723925,
      "loss": 5.1783,
      "mean_token_accuracy": 0.1806653767824173,
      "num_tokens": 26610970.0,
      "step": 14430
    },
    {
      "entropy": 5.627402400970459,
      "epoch": 1.212728418399496,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00048576627083283435,
      "loss": 5.4717,
      "mean_token_accuracy": 0.1654764398932457,
      "num_tokens": 26619840.0,
      "step": 14435
    },
    {
      "entropy": 5.632164764404297,
      "epoch": 1.2131484982146608,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004857557868148136,
      "loss": 5.3626,
      "mean_token_accuracy": 0.17133797258138656,
      "num_tokens": 26629271.0,
      "step": 14440
    },
    {
      "entropy": 5.661153793334961,
      "epoch": 1.2135685780298258,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004857452990633625,
      "loss": 5.4142,
      "mean_token_accuracy": 0.1605449289083481,
      "num_tokens": 26638610.0,
      "step": 14445
    },
    {
      "entropy": 5.73473424911499,
      "epoch": 1.2139886578449905,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048573480757866695,
      "loss": 5.5548,
      "mean_token_accuracy": 0.1624301940202713,
      "num_tokens": 26648504.0,
      "step": 14450
    },
    {
      "entropy": 5.697289085388183,
      "epoch": 1.2144087376601553,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048572431236091284,
      "loss": 5.4489,
      "mean_token_accuracy": 0.1593151181936264,
      "num_tokens": 26658084.0,
      "step": 14455
    },
    {
      "entropy": 5.722338914871216,
      "epoch": 1.2148288174753203,
      "grad_norm": 1.921875,
      "learning_rate": 0.00048571381341028604,
      "loss": 5.5185,
      "mean_token_accuracy": 0.16407164931297302,
      "num_tokens": 26666933.0,
      "step": 14460
    },
    {
      "entropy": 5.739486169815064,
      "epoch": 1.2152488972904851,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004857033107269725,
      "loss": 5.4105,
      "mean_token_accuracy": 0.16758025884628297,
      "num_tokens": 26675049.0,
      "step": 14465
    },
    {
      "entropy": 5.621294927597046,
      "epoch": 1.2156689771056501,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048569280431115823,
      "loss": 5.4754,
      "mean_token_accuracy": 0.16362982392311096,
      "num_tokens": 26684223.0,
      "step": 14470
    },
    {
      "entropy": 5.661048316955567,
      "epoch": 1.216089056920815,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004856822941630296,
      "loss": 5.4118,
      "mean_token_accuracy": 0.160876327753067,
      "num_tokens": 26693605.0,
      "step": 14475
    },
    {
      "entropy": 5.75145697593689,
      "epoch": 1.2165091367359797,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048567178028277255,
      "loss": 5.473,
      "mean_token_accuracy": 0.1669105038046837,
      "num_tokens": 26702829.0,
      "step": 14480
    },
    {
      "entropy": 5.730476236343383,
      "epoch": 1.2169292165511447,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004856612626705733,
      "loss": 5.5238,
      "mean_token_accuracy": 0.15920734852552415,
      "num_tokens": 26712466.0,
      "step": 14485
    },
    {
      "entropy": 5.672338390350342,
      "epoch": 1.2173492963663095,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004856507413266183,
      "loss": 5.4036,
      "mean_token_accuracy": 0.1744743689894676,
      "num_tokens": 26721730.0,
      "step": 14490
    },
    {
      "entropy": 5.648630571365357,
      "epoch": 1.2177693761814745,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000485640216251094,
      "loss": 5.476,
      "mean_token_accuracy": 0.16147005409002305,
      "num_tokens": 26731017.0,
      "step": 14495
    },
    {
      "entropy": 5.728749513626099,
      "epoch": 1.2181894559966393,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048562968744418665,
      "loss": 5.4411,
      "mean_token_accuracy": 0.16135525703430176,
      "num_tokens": 26739588.0,
      "step": 14500
    },
    {
      "entropy": 5.750088310241699,
      "epoch": 1.2186095358118043,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004856191549060828,
      "loss": 5.5545,
      "mean_token_accuracy": 0.15382490158081055,
      "num_tokens": 26748889.0,
      "step": 14505
    },
    {
      "entropy": 5.731989622116089,
      "epoch": 1.219029615626969,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048560861863696913,
      "loss": 5.503,
      "mean_token_accuracy": 0.16170719712972642,
      "num_tokens": 26757979.0,
      "step": 14510
    },
    {
      "entropy": 5.755252075195313,
      "epoch": 1.219449695442134,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004855980786370322,
      "loss": 5.4249,
      "mean_token_accuracy": 0.16454763114452362,
      "num_tokens": 26767225.0,
      "step": 14515
    },
    {
      "entropy": 5.638862180709839,
      "epoch": 1.219869775257299,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004855875349064588,
      "loss": 5.3742,
      "mean_token_accuracy": 0.16612455993890762,
      "num_tokens": 26776289.0,
      "step": 14520
    },
    {
      "entropy": 5.709742593765259,
      "epoch": 1.2202898550724637,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004855769874454356,
      "loss": 5.4903,
      "mean_token_accuracy": 0.15808172821998595,
      "num_tokens": 26785631.0,
      "step": 14525
    },
    {
      "entropy": 5.647239732742309,
      "epoch": 1.2207099348876287,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004855664362541495,
      "loss": 5.5121,
      "mean_token_accuracy": 0.15897773057222367,
      "num_tokens": 26795285.0,
      "step": 14530
    },
    {
      "entropy": 5.628441572189331,
      "epoch": 1.2211300147027935,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048555588133278744,
      "loss": 5.4128,
      "mean_token_accuracy": 0.16046550869941711,
      "num_tokens": 26804584.0,
      "step": 14535
    },
    {
      "entropy": 5.607101106643677,
      "epoch": 1.2215500945179585,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004855453226815363,
      "loss": 5.3055,
      "mean_token_accuracy": 0.16931912302970886,
      "num_tokens": 26814354.0,
      "step": 14540
    },
    {
      "entropy": 5.595360374450683,
      "epoch": 1.2219701743331233,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048553476030058326,
      "loss": 5.3156,
      "mean_token_accuracy": 0.1787246897816658,
      "num_tokens": 26824274.0,
      "step": 14545
    },
    {
      "entropy": 5.593567895889282,
      "epoch": 1.222390254148288,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048552419419011536,
      "loss": 5.4488,
      "mean_token_accuracy": 0.16586222499608994,
      "num_tokens": 26833155.0,
      "step": 14550
    },
    {
      "entropy": 5.637729072570801,
      "epoch": 1.222810333963453,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004855136243503196,
      "loss": 5.3859,
      "mean_token_accuracy": 0.16326625347137452,
      "num_tokens": 26842545.0,
      "step": 14555
    },
    {
      "entropy": 5.714113759994507,
      "epoch": 1.2232304137786179,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048550305078138363,
      "loss": 5.4666,
      "mean_token_accuracy": 0.16542484909296035,
      "num_tokens": 26851772.0,
      "step": 14560
    },
    {
      "entropy": 5.639671373367309,
      "epoch": 1.2236504935937829,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048549247348349435,
      "loss": 5.3469,
      "mean_token_accuracy": 0.16841816902160645,
      "num_tokens": 26860884.0,
      "step": 14565
    },
    {
      "entropy": 5.659556913375854,
      "epoch": 1.2240705734089476,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048548189245683934,
      "loss": 5.4783,
      "mean_token_accuracy": 0.16680554449558258,
      "num_tokens": 26869435.0,
      "step": 14570
    },
    {
      "entropy": 5.661346864700318,
      "epoch": 1.2244906532241127,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048547130770160596,
      "loss": 5.38,
      "mean_token_accuracy": 0.16684748679399491,
      "num_tokens": 26878852.0,
      "step": 14575
    },
    {
      "entropy": 5.670219469070434,
      "epoch": 1.2249107330392774,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004854607192179817,
      "loss": 5.3417,
      "mean_token_accuracy": 0.16878951340913773,
      "num_tokens": 26887532.0,
      "step": 14580
    },
    {
      "entropy": 5.798138046264649,
      "epoch": 1.2253308128544425,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004854501270061543,
      "loss": 5.5841,
      "mean_token_accuracy": 0.16193009316921234,
      "num_tokens": 26897459.0,
      "step": 14585
    },
    {
      "entropy": 5.631389856338501,
      "epoch": 1.2257508926696072,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048543953106631115,
      "loss": 5.3437,
      "mean_token_accuracy": 0.17152508199214936,
      "num_tokens": 26907156.0,
      "step": 14590
    },
    {
      "entropy": 5.7155900478363035,
      "epoch": 1.226170972484772,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004854289313986401,
      "loss": 5.4399,
      "mean_token_accuracy": 0.1658498004078865,
      "num_tokens": 26915764.0,
      "step": 14595
    },
    {
      "entropy": 5.578080129623413,
      "epoch": 1.226591052299937,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004854183280033289,
      "loss": 5.293,
      "mean_token_accuracy": 0.17005468606948854,
      "num_tokens": 26924166.0,
      "step": 14600
    },
    {
      "entropy": 5.705555152893067,
      "epoch": 1.2270111321151018,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004854077208805654,
      "loss": 5.5469,
      "mean_token_accuracy": 0.1581970199942589,
      "num_tokens": 26933546.0,
      "step": 14605
    },
    {
      "entropy": 5.711961030960083,
      "epoch": 1.2274312119302668,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004853971100305374,
      "loss": 5.4648,
      "mean_token_accuracy": 0.16808405220508577,
      "num_tokens": 26943213.0,
      "step": 14610
    },
    {
      "entropy": 5.702235364913941,
      "epoch": 1.2278512917454316,
      "grad_norm": 2.203125,
      "learning_rate": 0.000485386495453433,
      "loss": 5.4126,
      "mean_token_accuracy": 0.1717974692583084,
      "num_tokens": 26952968.0,
      "step": 14615
    },
    {
      "entropy": 5.671736001968384,
      "epoch": 1.2282713715605964,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048537587714944007,
      "loss": 5.41,
      "mean_token_accuracy": 0.16740047186613083,
      "num_tokens": 26962230.0,
      "step": 14620
    },
    {
      "entropy": 5.6375446796417235,
      "epoch": 1.2286914513757614,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004853652551187469,
      "loss": 5.4774,
      "mean_token_accuracy": 0.17203135490417482,
      "num_tokens": 26970985.0,
      "step": 14625
    },
    {
      "entropy": 5.717348384857178,
      "epoch": 1.2291115311909262,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048535462936154147,
      "loss": 5.5114,
      "mean_token_accuracy": 0.160255528986454,
      "num_tokens": 26981138.0,
      "step": 14630
    },
    {
      "entropy": 5.623097467422485,
      "epoch": 1.2295316110060912,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004853439998780122,
      "loss": 5.3563,
      "mean_token_accuracy": 0.17389864474534988,
      "num_tokens": 26990158.0,
      "step": 14635
    },
    {
      "entropy": 5.647943782806396,
      "epoch": 1.229951690821256,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004853333666683472,
      "loss": 5.507,
      "mean_token_accuracy": 0.15743157267570496,
      "num_tokens": 26998889.0,
      "step": 14640
    },
    {
      "entropy": 5.691283798217773,
      "epoch": 1.230371770636421,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048532272973273496,
      "loss": 5.4327,
      "mean_token_accuracy": 0.16522250920534134,
      "num_tokens": 27008912.0,
      "step": 14645
    },
    {
      "entropy": 5.639300966262818,
      "epoch": 1.2307918504515858,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048531208907136384,
      "loss": 5.3326,
      "mean_token_accuracy": 0.17845280468463898,
      "num_tokens": 27017573.0,
      "step": 14650
    },
    {
      "entropy": 5.629958868026733,
      "epoch": 1.2312119302667508,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048530144468442236,
      "loss": 5.4223,
      "mean_token_accuracy": 0.15466590747237205,
      "num_tokens": 27027205.0,
      "step": 14655
    },
    {
      "entropy": 5.702744960784912,
      "epoch": 1.2316320100819156,
      "grad_norm": 1.6875,
      "learning_rate": 0.00048529079657209906,
      "loss": 5.3471,
      "mean_token_accuracy": 0.1684443920850754,
      "num_tokens": 27035882.0,
      "step": 14660
    },
    {
      "entropy": 5.582692718505859,
      "epoch": 1.2320520898970804,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004852801447345826,
      "loss": 5.4266,
      "mean_token_accuracy": 0.17018674015998841,
      "num_tokens": 27044761.0,
      "step": 14665
    },
    {
      "entropy": 5.673105096817016,
      "epoch": 1.2324721697122454,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004852694891720617,
      "loss": 5.4613,
      "mean_token_accuracy": 0.16327601224184035,
      "num_tokens": 27054149.0,
      "step": 14670
    },
    {
      "entropy": 5.7073362350463865,
      "epoch": 1.2328922495274102,
      "grad_norm": 1.5,
      "learning_rate": 0.000485258829884725,
      "loss": 5.49,
      "mean_token_accuracy": 0.16526736468076705,
      "num_tokens": 27063145.0,
      "step": 14675
    },
    {
      "entropy": 5.735353565216064,
      "epoch": 1.2333123293425752,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004852481668727614,
      "loss": 5.4537,
      "mean_token_accuracy": 0.16542838364839554,
      "num_tokens": 27072378.0,
      "step": 14680
    },
    {
      "entropy": 5.6080183506011965,
      "epoch": 1.23373240915774,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048523750013635986,
      "loss": 5.3308,
      "mean_token_accuracy": 0.16522957533597946,
      "num_tokens": 27082241.0,
      "step": 14685
    },
    {
      "entropy": 5.59213662147522,
      "epoch": 1.2341524889729047,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004852268296757092,
      "loss": 5.3492,
      "mean_token_accuracy": 0.16783910393714904,
      "num_tokens": 27091488.0,
      "step": 14690
    },
    {
      "entropy": 5.69573221206665,
      "epoch": 1.2345725687880698,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004852161554909985,
      "loss": 5.4111,
      "mean_token_accuracy": 0.16741783171892166,
      "num_tokens": 27100378.0,
      "step": 14695
    },
    {
      "entropy": 5.6991900444030765,
      "epoch": 1.2349926486032345,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048520547758241686,
      "loss": 5.4165,
      "mean_token_accuracy": 0.1690449208021164,
      "num_tokens": 27110341.0,
      "step": 14700
    },
    {
      "entropy": 5.683678340911865,
      "epoch": 1.2354127284183996,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048519479595015343,
      "loss": 5.3629,
      "mean_token_accuracy": 0.16223382204771042,
      "num_tokens": 27119381.0,
      "step": 14705
    },
    {
      "entropy": 5.618437147140503,
      "epoch": 1.2358328082335643,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048518411059439746,
      "loss": 5.4655,
      "mean_token_accuracy": 0.15772671699523927,
      "num_tokens": 27129167.0,
      "step": 14710
    },
    {
      "entropy": 5.634173774719239,
      "epoch": 1.2362528880487293,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048517342151533813,
      "loss": 5.4712,
      "mean_token_accuracy": 0.1590983271598816,
      "num_tokens": 27138479.0,
      "step": 14715
    },
    {
      "entropy": 5.642277622222901,
      "epoch": 1.2366729678638941,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004851627287131649,
      "loss": 5.3531,
      "mean_token_accuracy": 0.1717111274600029,
      "num_tokens": 27147197.0,
      "step": 14720
    },
    {
      "entropy": 5.628361082077026,
      "epoch": 1.2370930476790591,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004851520321880672,
      "loss": 5.3891,
      "mean_token_accuracy": 0.17068208754062653,
      "num_tokens": 27155854.0,
      "step": 14725
    },
    {
      "entropy": 5.6246418952941895,
      "epoch": 1.237513127494224,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004851413319402344,
      "loss": 5.3439,
      "mean_token_accuracy": 0.16404491513967515,
      "num_tokens": 27165069.0,
      "step": 14730
    },
    {
      "entropy": 5.666826486587524,
      "epoch": 1.2379332073093887,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004851306279698561,
      "loss": 5.4071,
      "mean_token_accuracy": 0.16024885028600694,
      "num_tokens": 27174070.0,
      "step": 14735
    },
    {
      "entropy": 5.772764015197754,
      "epoch": 1.2383532871245537,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004851199202771219,
      "loss": 5.483,
      "mean_token_accuracy": 0.16502515822649003,
      "num_tokens": 27182903.0,
      "step": 14740
    },
    {
      "entropy": 5.702397966384888,
      "epoch": 1.2387733669397185,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004851092088622216,
      "loss": 5.401,
      "mean_token_accuracy": 0.16937888264656067,
      "num_tokens": 27192747.0,
      "step": 14745
    },
    {
      "entropy": 5.664963483810425,
      "epoch": 1.2391934467548835,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004850984937253448,
      "loss": 5.3985,
      "mean_token_accuracy": 0.16960911005735396,
      "num_tokens": 27201657.0,
      "step": 14750
    },
    {
      "entropy": 5.632458209991455,
      "epoch": 1.2396135265700483,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004850877748666814,
      "loss": 5.4448,
      "mean_token_accuracy": 0.16701988875865936,
      "num_tokens": 27211794.0,
      "step": 14755
    },
    {
      "entropy": 5.656436538696289,
      "epoch": 1.240033606385213,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048507705228642117,
      "loss": 5.4039,
      "mean_token_accuracy": 0.16116943657398225,
      "num_tokens": 27221852.0,
      "step": 14760
    },
    {
      "entropy": 5.649289464950561,
      "epoch": 1.240453686200378,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004850663259847542,
      "loss": 5.4512,
      "mean_token_accuracy": 0.16168087124824523,
      "num_tokens": 27231558.0,
      "step": 14765
    },
    {
      "entropy": 5.6387444019317625,
      "epoch": 1.240873766015543,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048505559596187037,
      "loss": 5.4236,
      "mean_token_accuracy": 0.16527110785245896,
      "num_tokens": 27241053.0,
      "step": 14770
    },
    {
      "entropy": 5.636253309249878,
      "epoch": 1.241293845830708,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004850448622179599,
      "loss": 5.3175,
      "mean_token_accuracy": 0.16779372990131378,
      "num_tokens": 27249770.0,
      "step": 14775
    },
    {
      "entropy": 5.738742828369141,
      "epoch": 1.2417139256458727,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004850341247532128,
      "loss": 5.562,
      "mean_token_accuracy": 0.1625165030360222,
      "num_tokens": 27258883.0,
      "step": 14780
    },
    {
      "entropy": 5.749982023239136,
      "epoch": 1.2421340054610377,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004850233835678194,
      "loss": 5.4633,
      "mean_token_accuracy": 0.1648506224155426,
      "num_tokens": 27268056.0,
      "step": 14785
    },
    {
      "entropy": 5.663808441162109,
      "epoch": 1.2425540852762025,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004850126386619699,
      "loss": 5.3236,
      "mean_token_accuracy": 0.17531862407922744,
      "num_tokens": 27276965.0,
      "step": 14790
    },
    {
      "entropy": 5.587924385070801,
      "epoch": 1.2429741650913673,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004850018900358545,
      "loss": 5.4055,
      "mean_token_accuracy": 0.17150697410106658,
      "num_tokens": 27286173.0,
      "step": 14795
    },
    {
      "entropy": 5.602390480041504,
      "epoch": 1.2433942449065323,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048499113768966386,
      "loss": 5.4064,
      "mean_token_accuracy": 0.16569943949580193,
      "num_tokens": 27294863.0,
      "step": 14800
    },
    {
      "entropy": 5.70964732170105,
      "epoch": 1.243814324721697,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004849803816235884,
      "loss": 5.4411,
      "mean_token_accuracy": 0.16401146501302719,
      "num_tokens": 27304427.0,
      "step": 14805
    },
    {
      "entropy": 5.745283937454223,
      "epoch": 1.244234404536862,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004849696218378185,
      "loss": 5.4928,
      "mean_token_accuracy": 0.16396332681179046,
      "num_tokens": 27313716.0,
      "step": 14810
    },
    {
      "entropy": 5.715461683273316,
      "epoch": 1.2446544843520269,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004849588583325449,
      "loss": 5.3965,
      "mean_token_accuracy": 0.17563249170780182,
      "num_tokens": 27322342.0,
      "step": 14815
    },
    {
      "entropy": 5.7087640285491945,
      "epoch": 1.2450745641671919,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004849480911079583,
      "loss": 5.4743,
      "mean_token_accuracy": 0.15280741229653358,
      "num_tokens": 27331892.0,
      "step": 14820
    },
    {
      "entropy": 5.690652322769165,
      "epoch": 1.2454946439823567,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004849373201642493,
      "loss": 5.4503,
      "mean_token_accuracy": 0.15473639070987702,
      "num_tokens": 27340428.0,
      "step": 14825
    },
    {
      "entropy": 5.666205930709839,
      "epoch": 1.2459147237975214,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004849265455016088,
      "loss": 5.4488,
      "mean_token_accuracy": 0.16578658074140548,
      "num_tokens": 27349224.0,
      "step": 14830
    },
    {
      "entropy": 5.656325817108154,
      "epoch": 1.2463348036126864,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004849157671202277,
      "loss": 5.4257,
      "mean_token_accuracy": 0.16552276760339737,
      "num_tokens": 27357480.0,
      "step": 14835
    },
    {
      "entropy": 5.63496470451355,
      "epoch": 1.2467548834278512,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004849049850202968,
      "loss": 5.3417,
      "mean_token_accuracy": 0.17225963473320008,
      "num_tokens": 27366732.0,
      "step": 14840
    },
    {
      "entropy": 5.706405782699585,
      "epoch": 1.2471749632430162,
      "grad_norm": 2.375,
      "learning_rate": 0.0004848941992020072,
      "loss": 5.4366,
      "mean_token_accuracy": 0.16214968562126159,
      "num_tokens": 27375834.0,
      "step": 14845
    },
    {
      "entropy": 5.714145278930664,
      "epoch": 1.247595043058181,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004848834096655499,
      "loss": 5.4407,
      "mean_token_accuracy": 0.16688166558742523,
      "num_tokens": 27385311.0,
      "step": 14850
    },
    {
      "entropy": 5.674659585952758,
      "epoch": 1.2480151228733458,
      "grad_norm": 2.125,
      "learning_rate": 0.00048487261641111607,
      "loss": 5.4804,
      "mean_token_accuracy": 0.1633618250489235,
      "num_tokens": 27394587.0,
      "step": 14855
    },
    {
      "entropy": 5.5959290027618405,
      "epoch": 1.2484352026885108,
      "grad_norm": 1.671875,
      "learning_rate": 0.000484861819438897,
      "loss": 5.3548,
      "mean_token_accuracy": 0.1634133994579315,
      "num_tokens": 27403316.0,
      "step": 14860
    },
    {
      "entropy": 5.690115690231323,
      "epoch": 1.2488552825036756,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004848510187490838,
      "loss": 5.4139,
      "mean_token_accuracy": 0.16767150908708572,
      "num_tokens": 27412709.0,
      "step": 14865
    },
    {
      "entropy": 5.697337675094604,
      "epoch": 1.2492753623188406,
      "grad_norm": 2.875,
      "learning_rate": 0.0004848402143418679,
      "loss": 5.4513,
      "mean_token_accuracy": 0.16266510039567947,
      "num_tokens": 27422004.0,
      "step": 14870
    },
    {
      "entropy": 5.71270170211792,
      "epoch": 1.2496954421340054,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048482940621744053,
      "loss": 5.492,
      "mean_token_accuracy": 0.16200196892023086,
      "num_tokens": 27431931.0,
      "step": 14875
    },
    {
      "entropy": 5.625197458267212,
      "epoch": 1.2501155219491704,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004848185943759934,
      "loss": 5.3174,
      "mean_token_accuracy": 0.17668476104736328,
      "num_tokens": 27441527.0,
      "step": 14880
    },
    {
      "entropy": 5.715439796447754,
      "epoch": 1.2505356017643352,
      "grad_norm": 1.75,
      "learning_rate": 0.00048480777881771786,
      "loss": 5.4617,
      "mean_token_accuracy": 0.16427915468811988,
      "num_tokens": 27449964.0,
      "step": 14885
    },
    {
      "entropy": 5.682178783416748,
      "epoch": 1.2509556815795002,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004847969595428056,
      "loss": 5.4647,
      "mean_token_accuracy": 0.16355433017015458,
      "num_tokens": 27459044.0,
      "step": 14890
    },
    {
      "entropy": 5.620614910125733,
      "epoch": 1.251375761394665,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00048478613655144817,
      "loss": 5.4447,
      "mean_token_accuracy": 0.16300222128629685,
      "num_tokens": 27467644.0,
      "step": 14895
    },
    {
      "entropy": 5.696579885482788,
      "epoch": 1.2517958412098298,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004847753098438374,
      "loss": 5.4725,
      "mean_token_accuracy": 0.1569360613822937,
      "num_tokens": 27476899.0,
      "step": 14900
    },
    {
      "entropy": 5.6774050235748295,
      "epoch": 1.2522159210249948,
      "grad_norm": 1.515625,
      "learning_rate": 0.000484764479420165,
      "loss": 5.365,
      "mean_token_accuracy": 0.17198057770729064,
      "num_tokens": 27485167.0,
      "step": 14905
    },
    {
      "entropy": 5.703144598007202,
      "epoch": 1.2526360008401596,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048475364528062287,
      "loss": 5.4154,
      "mean_token_accuracy": 0.16219851225614548,
      "num_tokens": 27493986.0,
      "step": 14910
    },
    {
      "entropy": 5.72310881614685,
      "epoch": 1.2530560806553246,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004847428074254029,
      "loss": 5.4616,
      "mean_token_accuracy": 0.16988417655229568,
      "num_tokens": 27503896.0,
      "step": 14915
    },
    {
      "entropy": 5.662931871414185,
      "epoch": 1.2534761604704894,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00048473196585469713,
      "loss": 5.4167,
      "mean_token_accuracy": 0.16819938123226166,
      "num_tokens": 27513485.0,
      "step": 14920
    },
    {
      "entropy": 5.721698808670044,
      "epoch": 1.2538962402856542,
      "grad_norm": 1.8125,
      "learning_rate": 0.00048472112056869763,
      "loss": 5.4664,
      "mean_token_accuracy": 0.1657877966761589,
      "num_tokens": 27523164.0,
      "step": 14925
    },
    {
      "entropy": 5.728442049026489,
      "epoch": 1.2543163201008192,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004847102715675964,
      "loss": 5.4042,
      "mean_token_accuracy": 0.16405912339687348,
      "num_tokens": 27531387.0,
      "step": 14930
    },
    {
      "entropy": 5.639872455596924,
      "epoch": 1.254736399915984,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004846994188515857,
      "loss": 5.4354,
      "mean_token_accuracy": 0.168145589530468,
      "num_tokens": 27541754.0,
      "step": 14935
    },
    {
      "entropy": 5.789422988891602,
      "epoch": 1.255156479731149,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004846885624208578,
      "loss": 5.5008,
      "mean_token_accuracy": 0.1617642045021057,
      "num_tokens": 27551458.0,
      "step": 14940
    },
    {
      "entropy": 5.697876119613648,
      "epoch": 1.2555765595463138,
      "grad_norm": 1.5,
      "learning_rate": 0.000484677702275605,
      "loss": 5.4113,
      "mean_token_accuracy": 0.170192214846611,
      "num_tokens": 27560797.0,
      "step": 14945
    },
    {
      "entropy": 5.651928424835205,
      "epoch": 1.2559966393614788,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00048466683841601963,
      "loss": 5.3958,
      "mean_token_accuracy": 0.16396012008190156,
      "num_tokens": 27570166.0,
      "step": 14950
    },
    {
      "entropy": 5.603499841690064,
      "epoch": 1.2564167191766435,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048465597084229416,
      "loss": 5.308,
      "mean_token_accuracy": 0.17143625617027283,
      "num_tokens": 27579411.0,
      "step": 14955
    },
    {
      "entropy": 5.695507955551148,
      "epoch": 1.2568367989918086,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004846450995546212,
      "loss": 5.5591,
      "mean_token_accuracy": 0.15752519071102142,
      "num_tokens": 27589124.0,
      "step": 14960
    },
    {
      "entropy": 5.721073627471924,
      "epoch": 1.2572568788069733,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004846342245531932,
      "loss": 5.5081,
      "mean_token_accuracy": 0.15465038716793061,
      "num_tokens": 27598664.0,
      "step": 14965
    },
    {
      "entropy": 5.7139750003814695,
      "epoch": 1.2576769586221381,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004846233458382029,
      "loss": 5.4358,
      "mean_token_accuracy": 0.16635446548461913,
      "num_tokens": 27607189.0,
      "step": 14970
    },
    {
      "entropy": 5.7070770263671875,
      "epoch": 1.2580970384373031,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048461246340984293,
      "loss": 5.4657,
      "mean_token_accuracy": 0.16415384262800217,
      "num_tokens": 27616415.0,
      "step": 14975
    },
    {
      "entropy": 5.698441743850708,
      "epoch": 1.258517118252468,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004846015772683061,
      "loss": 5.4595,
      "mean_token_accuracy": 0.16718726754188537,
      "num_tokens": 27624492.0,
      "step": 14980
    },
    {
      "entropy": 5.598542785644531,
      "epoch": 1.258937198067633,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048459068741378526,
      "loss": 5.3518,
      "mean_token_accuracy": 0.1675553500652313,
      "num_tokens": 27634243.0,
      "step": 14985
    },
    {
      "entropy": 5.647716331481933,
      "epoch": 1.2593572778827977,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004845797938464734,
      "loss": 5.4522,
      "mean_token_accuracy": 0.16648911386728288,
      "num_tokens": 27642887.0,
      "step": 14990
    },
    {
      "entropy": 5.7321514129638675,
      "epoch": 1.2597773576979625,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004845688965665633,
      "loss": 5.4703,
      "mean_token_accuracy": 0.15985920876264573,
      "num_tokens": 27652524.0,
      "step": 14995
    },
    {
      "entropy": 5.682502317428589,
      "epoch": 1.2601974375131275,
      "grad_norm": 1.625,
      "learning_rate": 0.00048455799557424814,
      "loss": 5.3173,
      "mean_token_accuracy": 0.17918399572372437,
      "num_tokens": 27661306.0,
      "step": 15000
    },
    {
      "epoch": 1.2601974375131275,
      "eval_entropy": 5.539990809356291,
      "eval_loss": 5.50443172454834,
      "eval_mean_token_accuracy": 0.17097199404162772,
      "eval_num_tokens": 27661306.0,
      "eval_runtime": 27.3249,
      "eval_samples_per_second": 1367.473,
      "eval_steps_per_second": 170.943,
      "step": 15000
    },
    {
      "entropy": 5.740118932723999,
      "epoch": 1.2606175173282923,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004845470908697209,
      "loss": 5.528,
      "mean_token_accuracy": 0.16602981388568877,
      "num_tokens": 27671728.0,
      "step": 15005
    },
    {
      "entropy": 5.626253604888916,
      "epoch": 1.2610375971434573,
      "grad_norm": 1.375,
      "learning_rate": 0.000484536182453175,
      "loss": 5.3218,
      "mean_token_accuracy": 0.17333738803863524,
      "num_tokens": 27680740.0,
      "step": 15010
    },
    {
      "entropy": 5.6302649021148685,
      "epoch": 1.261457676958622,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004845252703248035,
      "loss": 5.3716,
      "mean_token_accuracy": 0.16665417850017547,
      "num_tokens": 27689865.0,
      "step": 15015
    },
    {
      "entropy": 5.65459942817688,
      "epoch": 1.2618777567737869,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004845143544847997,
      "loss": 5.4306,
      "mean_token_accuracy": 0.16595624536275863,
      "num_tokens": 27700366.0,
      "step": 15020
    },
    {
      "entropy": 5.705493068695068,
      "epoch": 1.262297836588952,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048450343493335697,
      "loss": 5.3119,
      "mean_token_accuracy": 0.1745210662484169,
      "num_tokens": 27708893.0,
      "step": 15025
    },
    {
      "entropy": 5.5806032657623295,
      "epoch": 1.262717916404117,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004844925116706688,
      "loss": 5.36,
      "mean_token_accuracy": 0.16658202111721038,
      "num_tokens": 27717494.0,
      "step": 15030
    },
    {
      "entropy": 5.524540996551513,
      "epoch": 1.2631379962192817,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048448158469692866,
      "loss": 5.2822,
      "mean_token_accuracy": 0.18211074471473693,
      "num_tokens": 27726487.0,
      "step": 15035
    },
    {
      "entropy": 5.747773456573486,
      "epoch": 1.2635580760344465,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004844706540123301,
      "loss": 5.5093,
      "mean_token_accuracy": 0.15818075388669967,
      "num_tokens": 27736602.0,
      "step": 15040
    },
    {
      "entropy": 5.821777057647705,
      "epoch": 1.2639781558496115,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048445971961706675,
      "loss": 5.5215,
      "mean_token_accuracy": 0.1600981444120407,
      "num_tokens": 27746322.0,
      "step": 15045
    },
    {
      "entropy": 5.59310622215271,
      "epoch": 1.2643982356647763,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004844487815113323,
      "loss": 5.3842,
      "mean_token_accuracy": 0.17132694125175477,
      "num_tokens": 27754941.0,
      "step": 15050
    },
    {
      "entropy": 5.586697721481324,
      "epoch": 1.2648183154799413,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004844378396953206,
      "loss": 5.4566,
      "mean_token_accuracy": 0.16294800937175752,
      "num_tokens": 27763941.0,
      "step": 15055
    },
    {
      "entropy": 5.711082839965821,
      "epoch": 1.265238395295106,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048442689416922536,
      "loss": 5.4534,
      "mean_token_accuracy": 0.1667443260550499,
      "num_tokens": 27773087.0,
      "step": 15060
    },
    {
      "entropy": 5.6144452571868895,
      "epoch": 1.2656584751102709,
      "grad_norm": 1.9375,
      "learning_rate": 0.00048441594493324057,
      "loss": 5.2765,
      "mean_token_accuracy": 0.17230459451675414,
      "num_tokens": 27782648.0,
      "step": 15065
    },
    {
      "entropy": 5.612412023544311,
      "epoch": 1.2660785549254359,
      "grad_norm": 3.453125,
      "learning_rate": 0.00048440499198756015,
      "loss": 5.4682,
      "mean_token_accuracy": 0.16322605609893798,
      "num_tokens": 27791567.0,
      "step": 15070
    },
    {
      "entropy": 5.6634868621826175,
      "epoch": 1.2664986347406006,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048439403533237816,
      "loss": 5.481,
      "mean_token_accuracy": 0.15995543897151948,
      "num_tokens": 27801397.0,
      "step": 15075
    },
    {
      "entropy": 5.7584997653961185,
      "epoch": 1.2669187145557657,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004843830749678886,
      "loss": 5.4874,
      "mean_token_accuracy": 0.16391463428735734,
      "num_tokens": 27810831.0,
      "step": 15080
    },
    {
      "entropy": 5.650566339492798,
      "epoch": 1.2673387943709304,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004843721108942856,
      "loss": 5.3917,
      "mean_token_accuracy": 0.16781074553728104,
      "num_tokens": 27819591.0,
      "step": 15085
    },
    {
      "entropy": 5.629691410064697,
      "epoch": 1.2677588741860952,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004843611431117636,
      "loss": 5.3844,
      "mean_token_accuracy": 0.1685127153992653,
      "num_tokens": 27828614.0,
      "step": 15090
    },
    {
      "entropy": 5.648036146163941,
      "epoch": 1.2681789540012602,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004843501716205167,
      "loss": 5.4323,
      "mean_token_accuracy": 0.1689649283885956,
      "num_tokens": 27837549.0,
      "step": 15095
    },
    {
      "entropy": 5.694434404373169,
      "epoch": 1.2685990338164252,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004843391964207393,
      "loss": 5.4409,
      "mean_token_accuracy": 0.16338130384683608,
      "num_tokens": 27846678.0,
      "step": 15100
    },
    {
      "entropy": 5.701109600067139,
      "epoch": 1.26901911363159,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004843282175126258,
      "loss": 5.4579,
      "mean_token_accuracy": 0.1680167555809021,
      "num_tokens": 27855734.0,
      "step": 15105
    },
    {
      "entropy": 5.698972129821778,
      "epoch": 1.2694391934467548,
      "grad_norm": 1.625,
      "learning_rate": 0.00048431723489637086,
      "loss": 5.3932,
      "mean_token_accuracy": 0.16749608069658278,
      "num_tokens": 27865111.0,
      "step": 15110
    },
    {
      "entropy": 5.712488508224487,
      "epoch": 1.2698592732619198,
      "grad_norm": 1.875,
      "learning_rate": 0.00048430624857216876,
      "loss": 5.4281,
      "mean_token_accuracy": 0.16414488554000856,
      "num_tokens": 27874495.0,
      "step": 15115
    },
    {
      "entropy": 5.650420093536377,
      "epoch": 1.2702793530770846,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004842952585402143,
      "loss": 5.4437,
      "mean_token_accuracy": 0.16803406924009323,
      "num_tokens": 27884531.0,
      "step": 15120
    },
    {
      "entropy": 5.598265886306763,
      "epoch": 1.2706994328922496,
      "grad_norm": 1.6875,
      "learning_rate": 0.000484284264800702,
      "loss": 5.3274,
      "mean_token_accuracy": 0.17601890116930008,
      "num_tokens": 27893463.0,
      "step": 15125
    },
    {
      "entropy": 5.671769857406616,
      "epoch": 1.2711195127074144,
      "grad_norm": 2.0,
      "learning_rate": 0.00048427326735382687,
      "loss": 5.4507,
      "mean_token_accuracy": 0.16206626147031783,
      "num_tokens": 27903015.0,
      "step": 15130
    },
    {
      "entropy": 5.66033525466919,
      "epoch": 1.2715395925225792,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004842622661997834,
      "loss": 5.4295,
      "mean_token_accuracy": 0.17019102573394776,
      "num_tokens": 27912207.0,
      "step": 15135
    },
    {
      "entropy": 5.675687265396118,
      "epoch": 1.2719596723377442,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004842512613387668,
      "loss": 5.4242,
      "mean_token_accuracy": 0.15651100426912307,
      "num_tokens": 27921566.0,
      "step": 15140
    },
    {
      "entropy": 5.686455583572387,
      "epoch": 1.272379752152909,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004842402527709718,
      "loss": 5.3861,
      "mean_token_accuracy": 0.16979311257600785,
      "num_tokens": 27930633.0,
      "step": 15145
    },
    {
      "entropy": 5.69820990562439,
      "epoch": 1.272799831968074,
      "grad_norm": 1.375,
      "learning_rate": 0.0004842292404965934,
      "loss": 5.4906,
      "mean_token_accuracy": 0.15941346883773805,
      "num_tokens": 27939887.0,
      "step": 15150
    },
    {
      "entropy": 5.731826877593994,
      "epoch": 1.2732199117832388,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004842182245158268,
      "loss": 5.4904,
      "mean_token_accuracy": 0.1680808886885643,
      "num_tokens": 27949090.0,
      "step": 15155
    },
    {
      "entropy": 5.622611045837402,
      "epoch": 1.2736399915984036,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048420720482886715,
      "loss": 5.3229,
      "mean_token_accuracy": 0.16923893690109254,
      "num_tokens": 27958141.0,
      "step": 15160
    },
    {
      "entropy": 5.558681201934815,
      "epoch": 1.2740600714135686,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004841961814359095,
      "loss": 5.3694,
      "mean_token_accuracy": 0.17056505382061005,
      "num_tokens": 27967780.0,
      "step": 15165
    },
    {
      "entropy": 5.695374393463135,
      "epoch": 1.2744801512287336,
      "grad_norm": 1.375,
      "learning_rate": 0.00048418515433714917,
      "loss": 5.4675,
      "mean_token_accuracy": 0.16432348042726516,
      "num_tokens": 27976243.0,
      "step": 15170
    },
    {
      "entropy": 5.652808141708374,
      "epoch": 1.2749002310438984,
      "grad_norm": 1.875,
      "learning_rate": 0.0004841741235327817,
      "loss": 5.3246,
      "mean_token_accuracy": 0.17720907926559448,
      "num_tokens": 27985874.0,
      "step": 15175
    },
    {
      "entropy": 5.783726453781128,
      "epoch": 1.2753203108590632,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048416308902300215,
      "loss": 5.5611,
      "mean_token_accuracy": 0.15746224224567412,
      "num_tokens": 27995111.0,
      "step": 15180
    },
    {
      "entropy": 5.611493015289307,
      "epoch": 1.2757403906742282,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004841520508080063,
      "loss": 5.3865,
      "mean_token_accuracy": 0.1658123567700386,
      "num_tokens": 28003948.0,
      "step": 15185
    },
    {
      "entropy": 5.584348535537719,
      "epoch": 1.276160470489393,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048414100888798957,
      "loss": 5.4025,
      "mean_token_accuracy": 0.1662847191095352,
      "num_tokens": 28012941.0,
      "step": 15190
    },
    {
      "entropy": 5.594786882400513,
      "epoch": 1.276580550304558,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004841299632631475,
      "loss": 5.3754,
      "mean_token_accuracy": 0.16421881020069123,
      "num_tokens": 28022195.0,
      "step": 15195
    },
    {
      "entropy": 5.690154695510865,
      "epoch": 1.2770006301197228,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004841189139336759,
      "loss": 5.3385,
      "mean_token_accuracy": 0.167516852915287,
      "num_tokens": 28031446.0,
      "step": 15200
    },
    {
      "entropy": 5.657998752593994,
      "epoch": 1.2774207099348875,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004841078608997703,
      "loss": 5.3479,
      "mean_token_accuracy": 0.17438205927610398,
      "num_tokens": 28040906.0,
      "step": 15205
    },
    {
      "entropy": 5.6006917476654055,
      "epoch": 1.2778407897500526,
      "grad_norm": 1.5,
      "learning_rate": 0.0004840968041616267,
      "loss": 5.3613,
      "mean_token_accuracy": 0.16892288625240326,
      "num_tokens": 28049848.0,
      "step": 15210
    },
    {
      "entropy": 5.636824655532837,
      "epoch": 1.2782608695652173,
      "grad_norm": 1.625,
      "learning_rate": 0.00048408574371944094,
      "loss": 5.3648,
      "mean_token_accuracy": 0.16786900609731675,
      "num_tokens": 28058276.0,
      "step": 15215
    },
    {
      "entropy": 5.644614505767822,
      "epoch": 1.2786809493803823,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004840746795734088,
      "loss": 5.4813,
      "mean_token_accuracy": 0.16076158583164216,
      "num_tokens": 28068185.0,
      "step": 15220
    },
    {
      "entropy": 5.695021247863769,
      "epoch": 1.2791010291955471,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004840636117237264,
      "loss": 5.5161,
      "mean_token_accuracy": 0.16435035318136215,
      "num_tokens": 28077532.0,
      "step": 15225
    },
    {
      "entropy": 5.677987861633301,
      "epoch": 1.279521109010712,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004840525401705897,
      "loss": 5.3842,
      "mean_token_accuracy": 0.16521210521459578,
      "num_tokens": 28087593.0,
      "step": 15230
    },
    {
      "entropy": 5.6469056606292725,
      "epoch": 1.279941188825877,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048404146491419503,
      "loss": 5.3477,
      "mean_token_accuracy": 0.1719723403453827,
      "num_tokens": 28096256.0,
      "step": 15235
    },
    {
      "entropy": 5.675913572311401,
      "epoch": 1.2803612686410417,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048403038595473837,
      "loss": 5.3823,
      "mean_token_accuracy": 0.16878153681755065,
      "num_tokens": 28105048.0,
      "step": 15240
    },
    {
      "entropy": 5.681619501113891,
      "epoch": 1.2807813484562067,
      "grad_norm": 2.25,
      "learning_rate": 0.000484019303292416,
      "loss": 5.4377,
      "mean_token_accuracy": 0.15800079107284545,
      "num_tokens": 28114330.0,
      "step": 15245
    },
    {
      "entropy": 5.592029476165772,
      "epoch": 1.2812014282713715,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048400821692742434,
      "loss": 5.3553,
      "mean_token_accuracy": 0.17437646985054017,
      "num_tokens": 28123147.0,
      "step": 15250
    },
    {
      "entropy": 5.681383657455444,
      "epoch": 1.2816215080865365,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048399712685995983,
      "loss": 5.4772,
      "mean_token_accuracy": 0.16735728532075883,
      "num_tokens": 28132477.0,
      "step": 15255
    },
    {
      "entropy": 5.651606941223145,
      "epoch": 1.2820415879017013,
      "grad_norm": 1.625,
      "learning_rate": 0.00048398603309021877,
      "loss": 5.4705,
      "mean_token_accuracy": 0.16156791150569916,
      "num_tokens": 28141350.0,
      "step": 15260
    },
    {
      "entropy": 5.672251844406128,
      "epoch": 1.2824616677168663,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004839749356183978,
      "loss": 5.4191,
      "mean_token_accuracy": 0.16051417142152785,
      "num_tokens": 28149522.0,
      "step": 15265
    },
    {
      "entropy": 5.71749963760376,
      "epoch": 1.282881747532031,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004839638344446933,
      "loss": 5.5426,
      "mean_token_accuracy": 0.16109504848718642,
      "num_tokens": 28159646.0,
      "step": 15270
    },
    {
      "entropy": 5.791233873367309,
      "epoch": 1.283301827347196,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004839527295693023,
      "loss": 5.4266,
      "mean_token_accuracy": 0.16974038779735565,
      "num_tokens": 28168408.0,
      "step": 15275
    },
    {
      "entropy": 5.688838005065918,
      "epoch": 1.283721907162361,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004839416209924211,
      "loss": 5.4485,
      "mean_token_accuracy": 0.16379015445709227,
      "num_tokens": 28177744.0,
      "step": 15280
    },
    {
      "entropy": 5.6691258430480955,
      "epoch": 1.2841419869775257,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048393050871424676,
      "loss": 5.4993,
      "mean_token_accuracy": 0.16459273397922516,
      "num_tokens": 28186811.0,
      "step": 15285
    },
    {
      "entropy": 5.659898042678833,
      "epoch": 1.2845620667926907,
      "grad_norm": 2.0,
      "learning_rate": 0.000483919392734976,
      "loss": 5.4727,
      "mean_token_accuracy": 0.16189465820789337,
      "num_tokens": 28197052.0,
      "step": 15290
    },
    {
      "entropy": 5.635334396362305,
      "epoch": 1.2849821466078555,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004839082730548058,
      "loss": 5.3189,
      "mean_token_accuracy": 0.17702167481184006,
      "num_tokens": 28206000.0,
      "step": 15295
    },
    {
      "entropy": 5.633714246749878,
      "epoch": 1.2854022264230203,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004838971496739331,
      "loss": 5.3145,
      "mean_token_accuracy": 0.1692175582051277,
      "num_tokens": 28214679.0,
      "step": 15300
    },
    {
      "entropy": 5.572221660614014,
      "epoch": 1.2858223062381853,
      "grad_norm": 1.53125,
      "learning_rate": 0.000483886022592555,
      "loss": 5.4257,
      "mean_token_accuracy": 0.16499675512313844,
      "num_tokens": 28223890.0,
      "step": 15305
    },
    {
      "entropy": 5.599172353744507,
      "epoch": 1.28624238605335,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004838748918108685,
      "loss": 5.3496,
      "mean_token_accuracy": 0.17307732552289962,
      "num_tokens": 28232422.0,
      "step": 15310
    },
    {
      "entropy": 5.668398380279541,
      "epoch": 1.286662465868515,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048386375732907083,
      "loss": 5.3892,
      "mean_token_accuracy": 0.16759901046752929,
      "num_tokens": 28242079.0,
      "step": 15315
    },
    {
      "entropy": 5.778624057769775,
      "epoch": 1.2870825456836799,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00048385261914735936,
      "loss": 5.5994,
      "mean_token_accuracy": 0.16109821647405625,
      "num_tokens": 28252510.0,
      "step": 15320
    },
    {
      "entropy": 5.7435403823852536,
      "epoch": 1.2875026254988446,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048384147726593125,
      "loss": 5.4997,
      "mean_token_accuracy": 0.16292525976896285,
      "num_tokens": 28261348.0,
      "step": 15325
    },
    {
      "entropy": 5.6811802864074705,
      "epoch": 1.2879227053140097,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004838303316849839,
      "loss": 5.4134,
      "mean_token_accuracy": 0.1626274198293686,
      "num_tokens": 28270739.0,
      "step": 15330
    },
    {
      "entropy": 5.687950229644775,
      "epoch": 1.2883427851291747,
      "grad_norm": 1.5,
      "learning_rate": 0.00048381918240471473,
      "loss": 5.4736,
      "mean_token_accuracy": 0.15857867076992987,
      "num_tokens": 28279370.0,
      "step": 15335
    },
    {
      "entropy": 5.691525173187256,
      "epoch": 1.2887628649443394,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048380802942532124,
      "loss": 5.3952,
      "mean_token_accuracy": 0.16189926713705063,
      "num_tokens": 28287955.0,
      "step": 15340
    },
    {
      "entropy": 5.6139708995819095,
      "epoch": 1.2891829447595042,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048379687274700107,
      "loss": 5.3408,
      "mean_token_accuracy": 0.17437266409397126,
      "num_tokens": 28296832.0,
      "step": 15345
    },
    {
      "entropy": 5.614264583587646,
      "epoch": 1.2896030245746692,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048378571236995185,
      "loss": 5.3748,
      "mean_token_accuracy": 0.16980076879262923,
      "num_tokens": 28305778.0,
      "step": 15350
    },
    {
      "entropy": 5.657407474517822,
      "epoch": 1.290023104389834,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048377454829437124,
      "loss": 5.4187,
      "mean_token_accuracy": 0.16050552129745482,
      "num_tokens": 28314615.0,
      "step": 15355
    },
    {
      "entropy": 5.7148003578186035,
      "epoch": 1.290443184204999,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004837633805204569,
      "loss": 5.4905,
      "mean_token_accuracy": 0.16279885321855544,
      "num_tokens": 28324478.0,
      "step": 15360
    },
    {
      "entropy": 5.712179803848267,
      "epoch": 1.2908632640201638,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004837522090484069,
      "loss": 5.4509,
      "mean_token_accuracy": 0.16506802439689636,
      "num_tokens": 28333532.0,
      "step": 15365
    },
    {
      "entropy": 5.721324396133423,
      "epoch": 1.2912833438353286,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048374103387841894,
      "loss": 5.4081,
      "mean_token_accuracy": 0.15882999151945115,
      "num_tokens": 28343723.0,
      "step": 15370
    },
    {
      "entropy": 5.696778392791748,
      "epoch": 1.2917034236504936,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048372985501069106,
      "loss": 5.4023,
      "mean_token_accuracy": 0.16522992253303528,
      "num_tokens": 28351992.0,
      "step": 15375
    },
    {
      "entropy": 5.643481302261352,
      "epoch": 1.2921235034656584,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004837186724454213,
      "loss": 5.3962,
      "mean_token_accuracy": 0.1686009034514427,
      "num_tokens": 28361141.0,
      "step": 15380
    },
    {
      "entropy": 5.668659353256226,
      "epoch": 1.2925435832808234,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004837074861828077,
      "loss": 5.381,
      "mean_token_accuracy": 0.16764274686574937,
      "num_tokens": 28370339.0,
      "step": 15385
    },
    {
      "entropy": 5.67275333404541,
      "epoch": 1.2929636630959882,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004836962962230485,
      "loss": 5.4889,
      "mean_token_accuracy": 0.16424557715654373,
      "num_tokens": 28379242.0,
      "step": 15390
    },
    {
      "entropy": 5.640101718902588,
      "epoch": 1.293383742911153,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004836851025663418,
      "loss": 5.3747,
      "mean_token_accuracy": 0.17321812808513642,
      "num_tokens": 28388864.0,
      "step": 15395
    },
    {
      "entropy": 5.718721628189087,
      "epoch": 1.293803822726318,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000483673905212886,
      "loss": 5.4796,
      "mean_token_accuracy": 0.16458262205123902,
      "num_tokens": 28398000.0,
      "step": 15400
    },
    {
      "entropy": 5.579117107391357,
      "epoch": 1.294223902541483,
      "grad_norm": 1.625,
      "learning_rate": 0.0004836627041628794,
      "loss": 5.4191,
      "mean_token_accuracy": 0.16662654727697374,
      "num_tokens": 28407652.0,
      "step": 15405
    },
    {
      "entropy": 5.7414247512817385,
      "epoch": 1.2946439823566478,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004836514994165205,
      "loss": 5.4666,
      "mean_token_accuracy": 0.16150236427783965,
      "num_tokens": 28417694.0,
      "step": 15410
    },
    {
      "entropy": 5.731302785873413,
      "epoch": 1.2950640621718126,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048364029097400777,
      "loss": 5.4012,
      "mean_token_accuracy": 0.17097476571798326,
      "num_tokens": 28426928.0,
      "step": 15415
    },
    {
      "entropy": 5.666651439666748,
      "epoch": 1.2954841419869776,
      "grad_norm": 1.875,
      "learning_rate": 0.00048362907883553956,
      "loss": 5.4414,
      "mean_token_accuracy": 0.1606253504753113,
      "num_tokens": 28436176.0,
      "step": 15420
    },
    {
      "entropy": 5.678581762313843,
      "epoch": 1.2959042218021424,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048361786300131477,
      "loss": 5.5218,
      "mean_token_accuracy": 0.15690904185175897,
      "num_tokens": 28445277.0,
      "step": 15425
    },
    {
      "entropy": 5.7580749034881595,
      "epoch": 1.2963243016173074,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004836066434715319,
      "loss": 5.3988,
      "mean_token_accuracy": 0.16545010805130006,
      "num_tokens": 28453959.0,
      "step": 15430
    },
    {
      "entropy": 5.706592798233032,
      "epoch": 1.2967443814324722,
      "grad_norm": 1.75,
      "learning_rate": 0.0004835954202463898,
      "loss": 5.4844,
      "mean_token_accuracy": 0.16176826804876326,
      "num_tokens": 28463780.0,
      "step": 15435
    },
    {
      "entropy": 5.591341590881347,
      "epoch": 1.297164461247637,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004835841933260872,
      "loss": 5.3451,
      "mean_token_accuracy": 0.16188596040010453,
      "num_tokens": 28473299.0,
      "step": 15440
    },
    {
      "entropy": 5.68267674446106,
      "epoch": 1.297584541062802,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048357296271082305,
      "loss": 5.3804,
      "mean_token_accuracy": 0.16492928117513656,
      "num_tokens": 28481859.0,
      "step": 15445
    },
    {
      "entropy": 5.71458158493042,
      "epoch": 1.2980046208779668,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048356172840079625,
      "loss": 5.4446,
      "mean_token_accuracy": 0.1674533635377884,
      "num_tokens": 28491034.0,
      "step": 15450
    },
    {
      "entropy": 5.6562601089477536,
      "epoch": 1.2984247006931318,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004835504903962058,
      "loss": 5.3785,
      "mean_token_accuracy": 0.16320030838251115,
      "num_tokens": 28499829.0,
      "step": 15455
    },
    {
      "entropy": 5.62461109161377,
      "epoch": 1.2988447805082965,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048353924869725084,
      "loss": 5.359,
      "mean_token_accuracy": 0.17031383961439134,
      "num_tokens": 28508188.0,
      "step": 15460
    },
    {
      "entropy": 5.584475469589234,
      "epoch": 1.2992648603234613,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004835280033041305,
      "loss": 5.2664,
      "mean_token_accuracy": 0.17198913246393205,
      "num_tokens": 28516509.0,
      "step": 15465
    },
    {
      "entropy": 5.596610498428345,
      "epoch": 1.2996849401386263,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004835167542170439,
      "loss": 5.5167,
      "mean_token_accuracy": 0.15922940373420716,
      "num_tokens": 28526457.0,
      "step": 15470
    },
    {
      "entropy": 5.652245616912841,
      "epoch": 1.3001050199537914,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004835055014361904,
      "loss": 5.4266,
      "mean_token_accuracy": 0.16186550334095956,
      "num_tokens": 28536149.0,
      "step": 15475
    },
    {
      "entropy": 5.7870481491088865,
      "epoch": 1.3005250997689561,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048349424496176924,
      "loss": 5.4822,
      "mean_token_accuracy": 0.16614470705389978,
      "num_tokens": 28545486.0,
      "step": 15480
    },
    {
      "entropy": 5.6642012119293215,
      "epoch": 1.300945179584121,
      "grad_norm": 1.375,
      "learning_rate": 0.00048348298479397996,
      "loss": 5.3852,
      "mean_token_accuracy": 0.16934724897146225,
      "num_tokens": 28554555.0,
      "step": 15485
    },
    {
      "entropy": 5.580156564712524,
      "epoch": 1.301365259399286,
      "grad_norm": 2.0,
      "learning_rate": 0.00048347172093302196,
      "loss": 5.3941,
      "mean_token_accuracy": 0.17442938536405564,
      "num_tokens": 28563387.0,
      "step": 15490
    },
    {
      "entropy": 5.627208614349366,
      "epoch": 1.3017853392144507,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048346045337909475,
      "loss": 5.396,
      "mean_token_accuracy": 0.1660616174340248,
      "num_tokens": 28573437.0,
      "step": 15495
    },
    {
      "entropy": 5.598279333114624,
      "epoch": 1.3022054190296157,
      "grad_norm": 1.421875,
      "learning_rate": 0.000483449182132398,
      "loss": 5.3416,
      "mean_token_accuracy": 0.18041471093893052,
      "num_tokens": 28583362.0,
      "step": 15500
    },
    {
      "entropy": 5.792739629745483,
      "epoch": 1.3026254988447805,
      "grad_norm": 1.921875,
      "learning_rate": 0.00048343790719313124,
      "loss": 5.5214,
      "mean_token_accuracy": 0.16025079190731048,
      "num_tokens": 28593201.0,
      "step": 15505
    },
    {
      "entropy": 5.6832780838012695,
      "epoch": 1.3030455786599453,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048342662856149427,
      "loss": 5.4209,
      "mean_token_accuracy": 0.1612033486366272,
      "num_tokens": 28602486.0,
      "step": 15510
    },
    {
      "entropy": 5.60623607635498,
      "epoch": 1.3034656584751103,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000483415346237687,
      "loss": 5.4308,
      "mean_token_accuracy": 0.1651290088891983,
      "num_tokens": 28611643.0,
      "step": 15515
    },
    {
      "entropy": 5.68533353805542,
      "epoch": 1.303885738290275,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004834040602219091,
      "loss": 5.4872,
      "mean_token_accuracy": 0.1661544680595398,
      "num_tokens": 28620545.0,
      "step": 15520
    },
    {
      "entropy": 5.621874284744263,
      "epoch": 1.30430581810544,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048339277051436067,
      "loss": 5.4148,
      "mean_token_accuracy": 0.17089460641145707,
      "num_tokens": 28630024.0,
      "step": 15525
    },
    {
      "entropy": 5.761952877044678,
      "epoch": 1.304725897920605,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004833814771152415,
      "loss": 5.4714,
      "mean_token_accuracy": 0.1678002029657364,
      "num_tokens": 28638995.0,
      "step": 15530
    },
    {
      "entropy": 5.670332241058349,
      "epoch": 1.3051459777357697,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048337018002475184,
      "loss": 5.4258,
      "mean_token_accuracy": 0.1655592367053032,
      "num_tokens": 28647833.0,
      "step": 15535
    },
    {
      "entropy": 5.599709606170654,
      "epoch": 1.3055660575509347,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004833588792430917,
      "loss": 5.3235,
      "mean_token_accuracy": 0.17410400807857512,
      "num_tokens": 28657441.0,
      "step": 15540
    },
    {
      "entropy": 5.640621328353882,
      "epoch": 1.3059861373660997,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004833475747704614,
      "loss": 5.4465,
      "mean_token_accuracy": 0.1664034828543663,
      "num_tokens": 28666666.0,
      "step": 15545
    },
    {
      "entropy": 5.636472892761231,
      "epoch": 1.3064062171812645,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000483336266607061,
      "loss": 5.4568,
      "mean_token_accuracy": 0.16376900374889375,
      "num_tokens": 28676770.0,
      "step": 15550
    },
    {
      "entropy": 5.644525098800659,
      "epoch": 1.3068262969964293,
      "grad_norm": 1.5,
      "learning_rate": 0.00048332495475309097,
      "loss": 5.3537,
      "mean_token_accuracy": 0.16782074570655822,
      "num_tokens": 28685610.0,
      "step": 15555
    },
    {
      "entropy": 5.69416651725769,
      "epoch": 1.3072463768115943,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048331363920875155,
      "loss": 5.4508,
      "mean_token_accuracy": 0.1647176504135132,
      "num_tokens": 28695082.0,
      "step": 15560
    },
    {
      "entropy": 5.631888675689697,
      "epoch": 1.307666456626759,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048330231997424335,
      "loss": 5.3897,
      "mean_token_accuracy": 0.16708039939403535,
      "num_tokens": 28704006.0,
      "step": 15565
    },
    {
      "entropy": 5.651385116577148,
      "epoch": 1.308086536441924,
      "grad_norm": 1.875,
      "learning_rate": 0.0004832909970497668,
      "loss": 5.4031,
      "mean_token_accuracy": 0.16770894974470138,
      "num_tokens": 28713665.0,
      "step": 15570
    },
    {
      "entropy": 5.675976800918579,
      "epoch": 1.3085066162570889,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048327967043552245,
      "loss": 5.3657,
      "mean_token_accuracy": 0.17080232203006745,
      "num_tokens": 28722920.0,
      "step": 15575
    },
    {
      "entropy": 5.640656042098999,
      "epoch": 1.3089266960722536,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048326834013171107,
      "loss": 5.3121,
      "mean_token_accuracy": 0.1753455638885498,
      "num_tokens": 28731689.0,
      "step": 15580
    },
    {
      "entropy": 5.6382190704345705,
      "epoch": 1.3093467758874187,
      "grad_norm": 2.0,
      "learning_rate": 0.0004832570061385332,
      "loss": 5.4336,
      "mean_token_accuracy": 0.17522858083248138,
      "num_tokens": 28741308.0,
      "step": 15585
    },
    {
      "entropy": 5.6088639259338375,
      "epoch": 1.3097668557025834,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004832456684561898,
      "loss": 5.4034,
      "mean_token_accuracy": 0.16869428902864456,
      "num_tokens": 28750190.0,
      "step": 15590
    },
    {
      "entropy": 5.671949577331543,
      "epoch": 1.3101869355177485,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004832343270848815,
      "loss": 5.4826,
      "mean_token_accuracy": 0.15759723484516144,
      "num_tokens": 28759588.0,
      "step": 15595
    },
    {
      "entropy": 5.726993370056152,
      "epoch": 1.3106070153329132,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048322298202480935,
      "loss": 5.472,
      "mean_token_accuracy": 0.1627465769648552,
      "num_tokens": 28768800.0,
      "step": 15600
    },
    {
      "entropy": 5.7425398349761965,
      "epoch": 1.311027095148078,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048321163327617433,
      "loss": 5.4138,
      "mean_token_accuracy": 0.16536730974912645,
      "num_tokens": 28778108.0,
      "step": 15605
    },
    {
      "entropy": 5.728103113174439,
      "epoch": 1.311447174963243,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004832002808391775,
      "loss": 5.4102,
      "mean_token_accuracy": 0.16822538524866104,
      "num_tokens": 28787202.0,
      "step": 15610
    },
    {
      "entropy": 5.677258062362671,
      "epoch": 1.3118672547784078,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004831889247140198,
      "loss": 5.4366,
      "mean_token_accuracy": 0.1647050052881241,
      "num_tokens": 28797482.0,
      "step": 15615
    },
    {
      "entropy": 5.5999133586883545,
      "epoch": 1.3122873345935728,
      "grad_norm": 1.75,
      "learning_rate": 0.00048317756490090253,
      "loss": 5.3542,
      "mean_token_accuracy": 0.17025308758020402,
      "num_tokens": 28805872.0,
      "step": 15620
    },
    {
      "entropy": 5.667055320739746,
      "epoch": 1.3127074144087376,
      "grad_norm": 1.9375,
      "learning_rate": 0.00048316620140002685,
      "loss": 5.4762,
      "mean_token_accuracy": 0.16081090718507768,
      "num_tokens": 28814836.0,
      "step": 15625
    },
    {
      "entropy": 5.7981573104858395,
      "epoch": 1.3131274942239024,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004831548342115942,
      "loss": 5.4579,
      "mean_token_accuracy": 0.15672310143709184,
      "num_tokens": 28824727.0,
      "step": 15630
    },
    {
      "entropy": 5.807303428649902,
      "epoch": 1.3135475740390674,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048314346333580576,
      "loss": 5.5685,
      "mean_token_accuracy": 0.16139665246009827,
      "num_tokens": 28833848.0,
      "step": 15635
    },
    {
      "entropy": 5.657500743865967,
      "epoch": 1.3139676538542324,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004831320887728631,
      "loss": 5.3201,
      "mean_token_accuracy": 0.1758513405919075,
      "num_tokens": 28842198.0,
      "step": 15640
    },
    {
      "entropy": 5.653538751602173,
      "epoch": 1.3143877336693972,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004831207105229676,
      "loss": 5.4077,
      "mean_token_accuracy": 0.16642491072416304,
      "num_tokens": 28851804.0,
      "step": 15645
    },
    {
      "entropy": 5.571588611602783,
      "epoch": 1.314807813484562,
      "grad_norm": 1.71875,
      "learning_rate": 0.00048310932858632087,
      "loss": 5.3243,
      "mean_token_accuracy": 0.17372344583272933,
      "num_tokens": 28860181.0,
      "step": 15650
    },
    {
      "entropy": 5.602705335617065,
      "epoch": 1.315227893299727,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00048309794296312467,
      "loss": 5.3856,
      "mean_token_accuracy": 0.1769021064043045,
      "num_tokens": 28869945.0,
      "step": 15655
    },
    {
      "entropy": 5.6899755001068115,
      "epoch": 1.3156479731148918,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048308655365358053,
      "loss": 5.4468,
      "mean_token_accuracy": 0.17094084173440932,
      "num_tokens": 28880343.0,
      "step": 15660
    },
    {
      "entropy": 5.8332069396972654,
      "epoch": 1.3160680529300568,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048307516065789017,
      "loss": 5.5348,
      "mean_token_accuracy": 0.15504953935742377,
      "num_tokens": 28889441.0,
      "step": 15665
    },
    {
      "entropy": 5.709073162078857,
      "epoch": 1.3164881327452216,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048306376397625546,
      "loss": 5.4487,
      "mean_token_accuracy": 0.15954580307006835,
      "num_tokens": 28898154.0,
      "step": 15670
    },
    {
      "entropy": 5.675214862823486,
      "epoch": 1.3169082125603864,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048305236360887834,
      "loss": 5.442,
      "mean_token_accuracy": 0.16104377508163453,
      "num_tokens": 28908359.0,
      "step": 15675
    },
    {
      "entropy": 5.631436157226562,
      "epoch": 1.3173282923755514,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048304095955596074,
      "loss": 5.4521,
      "mean_token_accuracy": 0.1631111294031143,
      "num_tokens": 28918416.0,
      "step": 15680
    },
    {
      "entropy": 5.722534418106079,
      "epoch": 1.3177483721907162,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004830295518177047,
      "loss": 5.3755,
      "mean_token_accuracy": 0.17246500551700591,
      "num_tokens": 28927412.0,
      "step": 15685
    },
    {
      "entropy": 5.660516977310181,
      "epoch": 1.3181684520058812,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048301814039431227,
      "loss": 5.4203,
      "mean_token_accuracy": 0.16670238822698594,
      "num_tokens": 28936106.0,
      "step": 15690
    },
    {
      "entropy": 5.63556318283081,
      "epoch": 1.318588531821046,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048300672528598553,
      "loss": 5.4517,
      "mean_token_accuracy": 0.17143354564905167,
      "num_tokens": 28945197.0,
      "step": 15695
    },
    {
      "entropy": 5.727447652816773,
      "epoch": 1.3190086116362107,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004829953064929268,
      "loss": 5.469,
      "mean_token_accuracy": 0.16340519189834596,
      "num_tokens": 28954278.0,
      "step": 15700
    },
    {
      "entropy": 5.761994218826294,
      "epoch": 1.3194286914513758,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004829838840153383,
      "loss": 5.5355,
      "mean_token_accuracy": 0.16390469670295715,
      "num_tokens": 28963101.0,
      "step": 15705
    },
    {
      "entropy": 5.6193199634552,
      "epoch": 1.3198487712665408,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004829724578534224,
      "loss": 5.4311,
      "mean_token_accuracy": 0.16200152784585953,
      "num_tokens": 28972063.0,
      "step": 15710
    },
    {
      "entropy": 5.679441738128662,
      "epoch": 1.3202688510817056,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048296102800738153,
      "loss": 5.3728,
      "mean_token_accuracy": 0.171538782119751,
      "num_tokens": 28981617.0,
      "step": 15715
    },
    {
      "entropy": 5.693508291244507,
      "epoch": 1.3206889308968703,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048294959447741807,
      "loss": 5.3672,
      "mean_token_accuracy": 0.16616626530885698,
      "num_tokens": 28989442.0,
      "step": 15720
    },
    {
      "entropy": 5.627448081970215,
      "epoch": 1.3211090107120353,
      "grad_norm": 3.28125,
      "learning_rate": 0.00048293815726373467,
      "loss": 5.367,
      "mean_token_accuracy": 0.16815404444932938,
      "num_tokens": 28999104.0,
      "step": 15725
    },
    {
      "entropy": 5.652245903015137,
      "epoch": 1.3215290905272001,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048292671636653386,
      "loss": 5.4348,
      "mean_token_accuracy": 0.16832049638032914,
      "num_tokens": 29008645.0,
      "step": 15730
    },
    {
      "entropy": 5.697985315322876,
      "epoch": 1.3219491703423651,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004829152717860184,
      "loss": 5.4197,
      "mean_token_accuracy": 0.16434701383113862,
      "num_tokens": 29018655.0,
      "step": 15735
    },
    {
      "entropy": 5.697400093078613,
      "epoch": 1.32236925015753,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048290382352239087,
      "loss": 5.4205,
      "mean_token_accuracy": 0.16917917579412461,
      "num_tokens": 29027109.0,
      "step": 15740
    },
    {
      "entropy": 5.611954164505005,
      "epoch": 1.3227893299726947,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048289237157585424,
      "loss": 5.2315,
      "mean_token_accuracy": 0.18141870945692062,
      "num_tokens": 29035535.0,
      "step": 15745
    },
    {
      "entropy": 5.611086511611939,
      "epoch": 1.3232094097878597,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004828809159466112,
      "loss": 5.4058,
      "mean_token_accuracy": 0.16195948272943497,
      "num_tokens": 29044723.0,
      "step": 15750
    },
    {
      "entropy": 5.659829998016358,
      "epoch": 1.3236294896030245,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004828694566348648,
      "loss": 5.5628,
      "mean_token_accuracy": 0.15698053389787675,
      "num_tokens": 29053636.0,
      "step": 15755
    },
    {
      "entropy": 5.805431747436524,
      "epoch": 1.3240495694181895,
      "grad_norm": 2.375,
      "learning_rate": 0.00048285799364081806,
      "loss": 5.4536,
      "mean_token_accuracy": 0.15968141704797745,
      "num_tokens": 29062940.0,
      "step": 15760
    },
    {
      "entropy": 5.73027400970459,
      "epoch": 1.3244696492333543,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00048284652696467404,
      "loss": 5.3709,
      "mean_token_accuracy": 0.16666713058948518,
      "num_tokens": 29072159.0,
      "step": 15765
    },
    {
      "entropy": 5.6915867805480955,
      "epoch": 1.324889729048519,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048283505660663575,
      "loss": 5.4564,
      "mean_token_accuracy": 0.1679859459400177,
      "num_tokens": 29081544.0,
      "step": 15770
    },
    {
      "entropy": 5.63977689743042,
      "epoch": 1.325309808863684,
      "grad_norm": 1.75,
      "learning_rate": 0.0004828235825669064,
      "loss": 5.4084,
      "mean_token_accuracy": 0.1676102951169014,
      "num_tokens": 29090710.0,
      "step": 15775
    },
    {
      "entropy": 5.670703411102295,
      "epoch": 1.325729888678849,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048281210484568937,
      "loss": 5.4166,
      "mean_token_accuracy": 0.16657788977026938,
      "num_tokens": 29098988.0,
      "step": 15780
    },
    {
      "entropy": 5.573025465011597,
      "epoch": 1.326149968494014,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048280062344318794,
      "loss": 5.4649,
      "mean_token_accuracy": 0.1569308951497078,
      "num_tokens": 29108926.0,
      "step": 15785
    },
    {
      "entropy": 5.641130638122559,
      "epoch": 1.3265700483091787,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004827891383596054,
      "loss": 5.3712,
      "mean_token_accuracy": 0.16793366968631745,
      "num_tokens": 29118065.0,
      "step": 15790
    },
    {
      "entropy": 5.649385261535644,
      "epoch": 1.3269901281243437,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048277764959514524,
      "loss": 5.3288,
      "mean_token_accuracy": 0.16925195455551148,
      "num_tokens": 29127030.0,
      "step": 15795
    },
    {
      "entropy": 5.725132608413697,
      "epoch": 1.3274102079395085,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004827661571500111,
      "loss": 5.4628,
      "mean_token_accuracy": 0.16445249766111375,
      "num_tokens": 29137200.0,
      "step": 15800
    },
    {
      "entropy": 5.692111158370972,
      "epoch": 1.3278302877546735,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048275466102440644,
      "loss": 5.4538,
      "mean_token_accuracy": 0.16376561373472215,
      "num_tokens": 29147029.0,
      "step": 15805
    },
    {
      "entropy": 5.591803789138794,
      "epoch": 1.3282503675698383,
      "grad_norm": 1.6875,
      "learning_rate": 0.00048274316121853494,
      "loss": 5.3465,
      "mean_token_accuracy": 0.17478771060705184,
      "num_tokens": 29155675.0,
      "step": 15810
    },
    {
      "entropy": 5.679480457305909,
      "epoch": 1.328670447385003,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048273165773260023,
      "loss": 5.4006,
      "mean_token_accuracy": 0.16667551398277283,
      "num_tokens": 29164730.0,
      "step": 15815
    },
    {
      "entropy": 5.62858099937439,
      "epoch": 1.329090527200168,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004827201505668063,
      "loss": 5.4067,
      "mean_token_accuracy": 0.16634781211614608,
      "num_tokens": 29173074.0,
      "step": 15820
    },
    {
      "entropy": 5.733026361465454,
      "epoch": 1.3295106070153329,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004827086397213568,
      "loss": 5.5312,
      "mean_token_accuracy": 0.16071760803461074,
      "num_tokens": 29182175.0,
      "step": 15825
    },
    {
      "entropy": 5.862815237045288,
      "epoch": 1.3299306868304979,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004826971251964557,
      "loss": 5.723,
      "mean_token_accuracy": 0.15812294334173202,
      "num_tokens": 29192910.0,
      "step": 15830
    },
    {
      "entropy": 5.6796181201934814,
      "epoch": 1.3303507666456627,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000482685606992307,
      "loss": 5.3691,
      "mean_token_accuracy": 0.16864464730024337,
      "num_tokens": 29201969.0,
      "step": 15835
    },
    {
      "entropy": 5.702399587631225,
      "epoch": 1.3307708464608274,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048267408510911463,
      "loss": 5.5241,
      "mean_token_accuracy": 0.16161998659372329,
      "num_tokens": 29210475.0,
      "step": 15840
    },
    {
      "entropy": 5.650325727462769,
      "epoch": 1.3311909262759924,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004826625595470829,
      "loss": 5.3909,
      "mean_token_accuracy": 0.16496474742889405,
      "num_tokens": 29222586.0,
      "step": 15845
    },
    {
      "entropy": 5.668047189712524,
      "epoch": 1.3316110060911575,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048265103030641575,
      "loss": 5.4235,
      "mean_token_accuracy": 0.16500414907932281,
      "num_tokens": 29231503.0,
      "step": 15850
    },
    {
      "entropy": 5.621324348449707,
      "epoch": 1.3320310859063222,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004826394973873176,
      "loss": 5.4152,
      "mean_token_accuracy": 0.16088854372501374,
      "num_tokens": 29241534.0,
      "step": 15855
    },
    {
      "entropy": 5.654465675354004,
      "epoch": 1.332451165721487,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048262796078999266,
      "loss": 5.4137,
      "mean_token_accuracy": 0.16738684028387069,
      "num_tokens": 29250381.0,
      "step": 15860
    },
    {
      "entropy": 5.705436277389526,
      "epoch": 1.332871245536652,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004826164205146453,
      "loss": 5.5007,
      "mean_token_accuracy": 0.15687062889337539,
      "num_tokens": 29259205.0,
      "step": 15865
    },
    {
      "entropy": 5.607155609130859,
      "epoch": 1.3332913253518168,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048260487656147995,
      "loss": 5.382,
      "mean_token_accuracy": 0.1687689259648323,
      "num_tokens": 29267723.0,
      "step": 15870
    },
    {
      "entropy": 5.626588535308838,
      "epoch": 1.3337114051669818,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048259332893070106,
      "loss": 5.3938,
      "mean_token_accuracy": 0.17040273696184158,
      "num_tokens": 29277102.0,
      "step": 15875
    },
    {
      "entropy": 5.686921262741089,
      "epoch": 1.3341314849821466,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004825817776225133,
      "loss": 5.3839,
      "mean_token_accuracy": 0.16327580213546752,
      "num_tokens": 29286484.0,
      "step": 15880
    },
    {
      "entropy": 5.651821851730347,
      "epoch": 1.3345515647973114,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048257022263712123,
      "loss": 5.4759,
      "mean_token_accuracy": 0.16907175481319428,
      "num_tokens": 29296528.0,
      "step": 15885
    },
    {
      "entropy": 5.558361577987671,
      "epoch": 1.3349716446124764,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048255866397472954,
      "loss": 5.3099,
      "mean_token_accuracy": 0.1763760641217232,
      "num_tokens": 29305283.0,
      "step": 15890
    },
    {
      "entropy": 5.700358200073242,
      "epoch": 1.3353917244276412,
      "grad_norm": 1.484375,
      "learning_rate": 0.000482547101635543,
      "loss": 5.3272,
      "mean_token_accuracy": 0.1707787737250328,
      "num_tokens": 29315088.0,
      "step": 15895
    },
    {
      "entropy": 5.658198595046997,
      "epoch": 1.3358118042428062,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048253553561976645,
      "loss": 5.3482,
      "mean_token_accuracy": 0.17018151730298997,
      "num_tokens": 29323793.0,
      "step": 15900
    },
    {
      "entropy": 5.568337488174438,
      "epoch": 1.336231884057971,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004825239659276047,
      "loss": 5.4282,
      "mean_token_accuracy": 0.16275230050086975,
      "num_tokens": 29334015.0,
      "step": 15905
    },
    {
      "entropy": 5.725451993942261,
      "epoch": 1.3366519638731358,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004825123925592628,
      "loss": 5.5427,
      "mean_token_accuracy": 0.16163584142923354,
      "num_tokens": 29343221.0,
      "step": 15910
    },
    {
      "entropy": 5.648152446746826,
      "epoch": 1.3370720436883008,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048250081551494574,
      "loss": 5.3699,
      "mean_token_accuracy": 0.16689832657575607,
      "num_tokens": 29352261.0,
      "step": 15915
    },
    {
      "entropy": 5.632461404800415,
      "epoch": 1.3374921235034656,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004824892347948586,
      "loss": 5.4762,
      "mean_token_accuracy": 0.1636684164404869,
      "num_tokens": 29362138.0,
      "step": 15920
    },
    {
      "entropy": 5.657688474655151,
      "epoch": 1.3379122033186306,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004824776503992064,
      "loss": 5.37,
      "mean_token_accuracy": 0.16863038688898085,
      "num_tokens": 29371234.0,
      "step": 15925
    },
    {
      "entropy": 5.639244222640992,
      "epoch": 1.3383322831337954,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004824660623281945,
      "loss": 5.4042,
      "mean_token_accuracy": 0.17468772679567338,
      "num_tokens": 29380371.0,
      "step": 15930
    },
    {
      "entropy": 5.7573333263397215,
      "epoch": 1.3387523629489604,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048245447058202815,
      "loss": 5.5393,
      "mean_token_accuracy": 0.16147254556417465,
      "num_tokens": 29389230.0,
      "step": 15935
    },
    {
      "entropy": 5.7053827285766605,
      "epoch": 1.3391724427641252,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004824428751609126,
      "loss": 5.4136,
      "mean_token_accuracy": 0.17178767770528794,
      "num_tokens": 29398753.0,
      "step": 15940
    },
    {
      "entropy": 5.669283819198609,
      "epoch": 1.3395925225792902,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048243127606505343,
      "loss": 5.38,
      "mean_token_accuracy": 0.17132503688335418,
      "num_tokens": 29407487.0,
      "step": 15945
    },
    {
      "entropy": 5.549468564987182,
      "epoch": 1.340012602394455,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000482419673294656,
      "loss": 5.3813,
      "mean_token_accuracy": 0.16664464697241782,
      "num_tokens": 29416140.0,
      "step": 15950
    },
    {
      "entropy": 5.57112455368042,
      "epoch": 1.3404326822096198,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004824080668499259,
      "loss": 5.4062,
      "mean_token_accuracy": 0.16573573499917985,
      "num_tokens": 29424763.0,
      "step": 15955
    },
    {
      "entropy": 5.731543159484863,
      "epoch": 1.3408527620247848,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048239645673106855,
      "loss": 5.4124,
      "mean_token_accuracy": 0.16358748376369475,
      "num_tokens": 29434589.0,
      "step": 15960
    },
    {
      "entropy": 5.723189926147461,
      "epoch": 1.3412728418399495,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048238484293828995,
      "loss": 5.4248,
      "mean_token_accuracy": 0.16633280068635942,
      "num_tokens": 29443549.0,
      "step": 15965
    },
    {
      "entropy": 5.6820426940917965,
      "epoch": 1.3416929216551146,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004823732254717955,
      "loss": 5.4188,
      "mean_token_accuracy": 0.16279302090406417,
      "num_tokens": 29452457.0,
      "step": 15970
    },
    {
      "entropy": 5.589213705062866,
      "epoch": 1.3421130014702793,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004823616043317912,
      "loss": 5.3997,
      "mean_token_accuracy": 0.16387947052717208,
      "num_tokens": 29461238.0,
      "step": 15975
    },
    {
      "entropy": 5.69581823348999,
      "epoch": 1.3425330812854441,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048234997951848284,
      "loss": 5.4603,
      "mean_token_accuracy": 0.1613929435610771,
      "num_tokens": 29471170.0,
      "step": 15980
    },
    {
      "entropy": 5.76471643447876,
      "epoch": 1.3429531611006091,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004823383510320764,
      "loss": 5.4759,
      "mean_token_accuracy": 0.15751562193036078,
      "num_tokens": 29481017.0,
      "step": 15985
    },
    {
      "entropy": 5.739363956451416,
      "epoch": 1.343373240915774,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048232671887277786,
      "loss": 5.419,
      "mean_token_accuracy": 0.1685940816998482,
      "num_tokens": 29489809.0,
      "step": 15990
    },
    {
      "entropy": 5.631502962112426,
      "epoch": 1.343793320730939,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048231508304079313,
      "loss": 5.4566,
      "mean_token_accuracy": 0.16690857708454132,
      "num_tokens": 29499499.0,
      "step": 15995
    },
    {
      "entropy": 5.655130910873413,
      "epoch": 1.3442134005461037,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048230344353632855,
      "loss": 5.4091,
      "mean_token_accuracy": 0.1686744846403599,
      "num_tokens": 29508526.0,
      "step": 16000
    },
    {
      "entropy": 5.685823059082031,
      "epoch": 1.3446334803612685,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004822918003595902,
      "loss": 5.3444,
      "mean_token_accuracy": 0.1660786375403404,
      "num_tokens": 29517516.0,
      "step": 16005
    },
    {
      "entropy": 5.658749771118164,
      "epoch": 1.3450535601764335,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004822801535107843,
      "loss": 5.4503,
      "mean_token_accuracy": 0.16219758689403535,
      "num_tokens": 29526949.0,
      "step": 16010
    },
    {
      "entropy": 5.59309139251709,
      "epoch": 1.3454736399915985,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004822685029901173,
      "loss": 5.3416,
      "mean_token_accuracy": 0.16975321769714355,
      "num_tokens": 29536696.0,
      "step": 16015
    },
    {
      "entropy": 5.675222110748291,
      "epoch": 1.3458937198067633,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004822568487977954,
      "loss": 5.4269,
      "mean_token_accuracy": 0.17550380378961564,
      "num_tokens": 29545672.0,
      "step": 16020
    },
    {
      "entropy": 5.687830686569214,
      "epoch": 1.346313799621928,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048224519093402517,
      "loss": 5.4681,
      "mean_token_accuracy": 0.165561743080616,
      "num_tokens": 29554888.0,
      "step": 16025
    },
    {
      "entropy": 5.640204286575317,
      "epoch": 1.346733879437093,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048223352939901317,
      "loss": 5.392,
      "mean_token_accuracy": 0.17039609402418138,
      "num_tokens": 29564798.0,
      "step": 16030
    },
    {
      "entropy": 5.640942716598511,
      "epoch": 1.347153959252258,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004822218641929658,
      "loss": 5.4129,
      "mean_token_accuracy": 0.17124655097723007,
      "num_tokens": 29574802.0,
      "step": 16035
    },
    {
      "entropy": 5.74286994934082,
      "epoch": 1.347574039067423,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004822101953160899,
      "loss": 5.4086,
      "mean_token_accuracy": 0.16020314395427704,
      "num_tokens": 29583056.0,
      "step": 16040
    },
    {
      "entropy": 5.669561100006104,
      "epoch": 1.3479941188825877,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000482198522768592,
      "loss": 5.3908,
      "mean_token_accuracy": 0.1646451786160469,
      "num_tokens": 29591935.0,
      "step": 16045
    },
    {
      "entropy": 5.581321716308594,
      "epoch": 1.3484141986977525,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048218684655067907,
      "loss": 5.3277,
      "mean_token_accuracy": 0.17401027381420137,
      "num_tokens": 29600812.0,
      "step": 16050
    },
    {
      "entropy": 5.693939971923828,
      "epoch": 1.3488342785129175,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004821751666625577,
      "loss": 5.4566,
      "mean_token_accuracy": 0.16635051518678665,
      "num_tokens": 29610735.0,
      "step": 16055
    },
    {
      "entropy": 5.685873317718506,
      "epoch": 1.3492543583280823,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048216348310443506,
      "loss": 5.3747,
      "mean_token_accuracy": 0.16337648183107376,
      "num_tokens": 29620295.0,
      "step": 16060
    },
    {
      "entropy": 5.5523065567016605,
      "epoch": 1.3496744381432473,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048215179587651795,
      "loss": 5.2543,
      "mean_token_accuracy": 0.17642796337604522,
      "num_tokens": 29628214.0,
      "step": 16065
    },
    {
      "entropy": 5.5786426067352295,
      "epoch": 1.350094517958412,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004821401049790134,
      "loss": 5.3666,
      "mean_token_accuracy": 0.17696074694395064,
      "num_tokens": 29636598.0,
      "step": 16070
    },
    {
      "entropy": 5.657120943069458,
      "epoch": 1.3505145977735769,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004821284104121286,
      "loss": 5.3638,
      "mean_token_accuracy": 0.17189116179943084,
      "num_tokens": 29646052.0,
      "step": 16075
    },
    {
      "entropy": 5.596641874313354,
      "epoch": 1.3509346775887419,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048211671217607066,
      "loss": 5.3991,
      "mean_token_accuracy": 0.15655021667480468,
      "num_tokens": 29655310.0,
      "step": 16080
    },
    {
      "entropy": 5.6666512966156,
      "epoch": 1.3513547574039069,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004821050102710468,
      "loss": 5.3816,
      "mean_token_accuracy": 0.16549568772315978,
      "num_tokens": 29664020.0,
      "step": 16085
    },
    {
      "entropy": 5.623274230957032,
      "epoch": 1.3517748372190717,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048209330469726433,
      "loss": 5.467,
      "mean_token_accuracy": 0.16062849909067153,
      "num_tokens": 29672416.0,
      "step": 16090
    },
    {
      "entropy": 5.669020700454712,
      "epoch": 1.3521949170342364,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048208159545493057,
      "loss": 5.3496,
      "mean_token_accuracy": 0.1768997758626938,
      "num_tokens": 29681148.0,
      "step": 16095
    },
    {
      "entropy": 5.650693941116333,
      "epoch": 1.3526149968494015,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004820698825442531,
      "loss": 5.3282,
      "mean_token_accuracy": 0.17453669309616088,
      "num_tokens": 29689089.0,
      "step": 16100
    },
    {
      "entropy": 5.621245336532593,
      "epoch": 1.3530350766645662,
      "grad_norm": 1.5,
      "learning_rate": 0.00048205816596543914,
      "loss": 5.4492,
      "mean_token_accuracy": 0.16295104026794432,
      "num_tokens": 29697704.0,
      "step": 16105
    },
    {
      "entropy": 5.728252172470093,
      "epoch": 1.3534551564797312,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048204644571869646,
      "loss": 5.4602,
      "mean_token_accuracy": 0.16218211352825165,
      "num_tokens": 29706966.0,
      "step": 16110
    },
    {
      "entropy": 5.6020056247711185,
      "epoch": 1.353875236294896,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004820347218042326,
      "loss": 5.3677,
      "mean_token_accuracy": 0.1657108783721924,
      "num_tokens": 29715817.0,
      "step": 16115
    },
    {
      "entropy": 5.6382012367248535,
      "epoch": 1.3542953161100608,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004820229942222553,
      "loss": 5.471,
      "mean_token_accuracy": 0.16378311514854432,
      "num_tokens": 29725500.0,
      "step": 16120
    },
    {
      "entropy": 5.654023885726929,
      "epoch": 1.3547153959252258,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048201126297297214,
      "loss": 5.3703,
      "mean_token_accuracy": 0.16789211332798004,
      "num_tokens": 29734774.0,
      "step": 16125
    },
    {
      "entropy": 5.649758577346802,
      "epoch": 1.3551354757403906,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004819995280565911,
      "loss": 5.344,
      "mean_token_accuracy": 0.1713784232735634,
      "num_tokens": 29744667.0,
      "step": 16130
    },
    {
      "entropy": 5.726012516021728,
      "epoch": 1.3555555555555556,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048198778947332,
      "loss": 5.4539,
      "mean_token_accuracy": 0.16631684005260466,
      "num_tokens": 29753644.0,
      "step": 16135
    },
    {
      "entropy": 5.724636268615723,
      "epoch": 1.3559756353707204,
      "grad_norm": 1.5,
      "learning_rate": 0.0004819760472233668,
      "loss": 5.4064,
      "mean_token_accuracy": 0.1708889827132225,
      "num_tokens": 29762977.0,
      "step": 16140
    },
    {
      "entropy": 5.610619306564331,
      "epoch": 1.3563957151858852,
      "grad_norm": 1.6875,
      "learning_rate": 0.00048196430130693956,
      "loss": 5.3913,
      "mean_token_accuracy": 0.16885729730129242,
      "num_tokens": 29772221.0,
      "step": 16145
    },
    {
      "entropy": 5.648658275604248,
      "epoch": 1.3568157950010502,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048195255172424627,
      "loss": 5.3756,
      "mean_token_accuracy": 0.1758609801530838,
      "num_tokens": 29781240.0,
      "step": 16150
    },
    {
      "entropy": 5.700412368774414,
      "epoch": 1.3572358748162152,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048194079847549507,
      "loss": 5.3655,
      "mean_token_accuracy": 0.16792706549167633,
      "num_tokens": 29790330.0,
      "step": 16155
    },
    {
      "entropy": 5.715218544006348,
      "epoch": 1.35765595463138,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004819290415608942,
      "loss": 5.4906,
      "mean_token_accuracy": 0.1646016627550125,
      "num_tokens": 29800945.0,
      "step": 16160
    },
    {
      "entropy": 5.721516180038452,
      "epoch": 1.3580760344465448,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004819172809806519,
      "loss": 5.5395,
      "mean_token_accuracy": 0.16160600483417512,
      "num_tokens": 29810391.0,
      "step": 16165
    },
    {
      "entropy": 5.70505657196045,
      "epoch": 1.3584961142617098,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048190551673497645,
      "loss": 5.3917,
      "mean_token_accuracy": 0.168082594871521,
      "num_tokens": 29819511.0,
      "step": 16170
    },
    {
      "entropy": 5.667353963851928,
      "epoch": 1.3589161940768746,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004818937488240764,
      "loss": 5.4171,
      "mean_token_accuracy": 0.1622082620859146,
      "num_tokens": 29828313.0,
      "step": 16175
    },
    {
      "entropy": 5.583069562911987,
      "epoch": 1.3593362738920396,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048188197724816014,
      "loss": 5.337,
      "mean_token_accuracy": 0.17444069236516951,
      "num_tokens": 29837940.0,
      "step": 16180
    },
    {
      "entropy": 5.619865131378174,
      "epoch": 1.3597563537072044,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048187020200743613,
      "loss": 5.3184,
      "mean_token_accuracy": 0.1743628665804863,
      "num_tokens": 29846799.0,
      "step": 16185
    },
    {
      "entropy": 5.64148211479187,
      "epoch": 1.3601764335223692,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000481858423102113,
      "loss": 5.4569,
      "mean_token_accuracy": 0.16412997096776963,
      "num_tokens": 29856263.0,
      "step": 16190
    },
    {
      "entropy": 5.636111736297607,
      "epoch": 1.3605965133375342,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004818466405323994,
      "loss": 5.3716,
      "mean_token_accuracy": 0.16748650670051574,
      "num_tokens": 29864335.0,
      "step": 16195
    },
    {
      "entropy": 5.714676570892334,
      "epoch": 1.361016593152699,
      "grad_norm": 1.6875,
      "learning_rate": 0.00048183485429850417,
      "loss": 5.4178,
      "mean_token_accuracy": 0.16076132357120515,
      "num_tokens": 29873466.0,
      "step": 16200
    },
    {
      "entropy": 5.582801818847656,
      "epoch": 1.361436672967864,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004818230644006359,
      "loss": 5.4041,
      "mean_token_accuracy": 0.17851744890213012,
      "num_tokens": 29883051.0,
      "step": 16205
    },
    {
      "entropy": 5.632126808166504,
      "epoch": 1.3618567527830288,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004818112708390036,
      "loss": 5.351,
      "mean_token_accuracy": 0.16902291923761367,
      "num_tokens": 29891823.0,
      "step": 16210
    },
    {
      "entropy": 5.635466575622559,
      "epoch": 1.3622768325981935,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004817994736138162,
      "loss": 5.3548,
      "mean_token_accuracy": 0.16993682384490966,
      "num_tokens": 29900735.0,
      "step": 16215
    },
    {
      "entropy": 5.644832944869995,
      "epoch": 1.3626969124133586,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004817876727252824,
      "loss": 5.4477,
      "mean_token_accuracy": 0.16936271041631698,
      "num_tokens": 29910345.0,
      "step": 16220
    },
    {
      "entropy": 5.6721724510192875,
      "epoch": 1.3631169922285233,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00048177586817361166,
      "loss": 5.3877,
      "mean_token_accuracy": 0.16757078915834428,
      "num_tokens": 29919650.0,
      "step": 16225
    },
    {
      "entropy": 5.722408390045166,
      "epoch": 1.3635370720436883,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004817640599590128,
      "loss": 5.4311,
      "mean_token_accuracy": 0.16306850612163543,
      "num_tokens": 29928851.0,
      "step": 16230
    },
    {
      "entropy": 5.7201415538787845,
      "epoch": 1.3639571518588531,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048175224808169506,
      "loss": 5.5249,
      "mean_token_accuracy": 0.15673743486404418,
      "num_tokens": 29939146.0,
      "step": 16235
    },
    {
      "entropy": 5.6970385074615475,
      "epoch": 1.3643772316740181,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048174043254186775,
      "loss": 5.3631,
      "mean_token_accuracy": 0.1684204787015915,
      "num_tokens": 29947556.0,
      "step": 16240
    },
    {
      "entropy": 5.721104145050049,
      "epoch": 1.364797311489183,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004817286133397401,
      "loss": 5.4662,
      "mean_token_accuracy": 0.16225254982709886,
      "num_tokens": 29957319.0,
      "step": 16245
    },
    {
      "entropy": 5.686917638778686,
      "epoch": 1.365217391304348,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004817167904755216,
      "loss": 5.4307,
      "mean_token_accuracy": 0.1654740959405899,
      "num_tokens": 29966697.0,
      "step": 16250
    },
    {
      "entropy": 5.667904806137085,
      "epoch": 1.3656374711195127,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048170496394942154,
      "loss": 5.4284,
      "mean_token_accuracy": 0.16945357620716095,
      "num_tokens": 29975103.0,
      "step": 16255
    },
    {
      "entropy": 5.602601337432861,
      "epoch": 1.3660575509346775,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048169313376164943,
      "loss": 5.332,
      "mean_token_accuracy": 0.16808460801839828,
      "num_tokens": 29984865.0,
      "step": 16260
    },
    {
      "entropy": 5.671368885040283,
      "epoch": 1.3664776307498425,
      "grad_norm": 1.625,
      "learning_rate": 0.00048168129991241497,
      "loss": 5.3684,
      "mean_token_accuracy": 0.16912826001644135,
      "num_tokens": 29994376.0,
      "step": 16265
    },
    {
      "entropy": 5.768115043640137,
      "epoch": 1.3668977105650073,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004816694624019277,
      "loss": 5.5964,
      "mean_token_accuracy": 0.1639996662735939,
      "num_tokens": 30004846.0,
      "step": 16270
    },
    {
      "entropy": 5.667897796630859,
      "epoch": 1.3673177903801723,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048165762123039723,
      "loss": 5.373,
      "mean_token_accuracy": 0.16785126626491548,
      "num_tokens": 30014083.0,
      "step": 16275
    },
    {
      "entropy": 5.634545803070068,
      "epoch": 1.367737870195337,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048164577639803354,
      "loss": 5.3769,
      "mean_token_accuracy": 0.17043836861848832,
      "num_tokens": 30023606.0,
      "step": 16280
    },
    {
      "entropy": 5.582794809341431,
      "epoch": 1.3681579500105019,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004816339279050463,
      "loss": 5.3653,
      "mean_token_accuracy": 0.15966161638498305,
      "num_tokens": 30033657.0,
      "step": 16285
    },
    {
      "entropy": 5.64022970199585,
      "epoch": 1.368578029825667,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048162207575164537,
      "loss": 5.4065,
      "mean_token_accuracy": 0.1677308440208435,
      "num_tokens": 30043230.0,
      "step": 16290
    },
    {
      "entropy": 5.681842994689942,
      "epoch": 1.3689981096408317,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048161021993804075,
      "loss": 5.4394,
      "mean_token_accuracy": 0.16544794291257858,
      "num_tokens": 30054457.0,
      "step": 16295
    },
    {
      "entropy": 5.607257509231568,
      "epoch": 1.3694181894559967,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048159836046444255,
      "loss": 5.2931,
      "mean_token_accuracy": 0.17434093803167344,
      "num_tokens": 30062912.0,
      "step": 16300
    },
    {
      "entropy": 5.6496968269348145,
      "epoch": 1.3698382692711615,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004815864973310607,
      "loss": 5.449,
      "mean_token_accuracy": 0.16527730822563172,
      "num_tokens": 30071340.0,
      "step": 16305
    },
    {
      "entropy": 5.761241769790649,
      "epoch": 1.3702583490863263,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048157463053810553,
      "loss": 5.5307,
      "mean_token_accuracy": 0.15745776742696763,
      "num_tokens": 30080334.0,
      "step": 16310
    },
    {
      "entropy": 5.688508176803589,
      "epoch": 1.3706784289014913,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048156276008578706,
      "loss": 5.3762,
      "mean_token_accuracy": 0.16883545219898224,
      "num_tokens": 30089391.0,
      "step": 16315
    },
    {
      "entropy": 5.621118450164795,
      "epoch": 1.3710985087166563,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004815508859743157,
      "loss": 5.3457,
      "mean_token_accuracy": 0.1709073156118393,
      "num_tokens": 30099027.0,
      "step": 16320
    },
    {
      "entropy": 5.624277973175049,
      "epoch": 1.371518588531821,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004815390082039017,
      "loss": 5.3511,
      "mean_token_accuracy": 0.16971062421798705,
      "num_tokens": 30108088.0,
      "step": 16325
    },
    {
      "entropy": 5.581058311462402,
      "epoch": 1.3719386683469859,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048152712677475556,
      "loss": 5.3499,
      "mean_token_accuracy": 0.1683119624853134,
      "num_tokens": 30117768.0,
      "step": 16330
    },
    {
      "entropy": 5.678498029708862,
      "epoch": 1.3723587481621509,
      "grad_norm": 1.625,
      "learning_rate": 0.00048151524168708773,
      "loss": 5.4536,
      "mean_token_accuracy": 0.16896748319268226,
      "num_tokens": 30126364.0,
      "step": 16335
    },
    {
      "entropy": 5.64181604385376,
      "epoch": 1.3727788279773157,
      "grad_norm": 1.859375,
      "learning_rate": 0.00048150335294110867,
      "loss": 5.4052,
      "mean_token_accuracy": 0.1731199026107788,
      "num_tokens": 30135365.0,
      "step": 16340
    },
    {
      "entropy": 5.717572402954102,
      "epoch": 1.3731989077924807,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048149146053702915,
      "loss": 5.4736,
      "mean_token_accuracy": 0.1794845297932625,
      "num_tokens": 30145542.0,
      "step": 16345
    },
    {
      "entropy": 5.690555429458618,
      "epoch": 1.3736189876076454,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004814795644750597,
      "loss": 5.4662,
      "mean_token_accuracy": 0.16346006095409393,
      "num_tokens": 30154100.0,
      "step": 16350
    },
    {
      "entropy": 5.612636661529541,
      "epoch": 1.3740390674228102,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048146766475541105,
      "loss": 5.365,
      "mean_token_accuracy": 0.16891611516475677,
      "num_tokens": 30162647.0,
      "step": 16355
    },
    {
      "entropy": 5.798755931854248,
      "epoch": 1.3744591472379752,
      "grad_norm": 2.125,
      "learning_rate": 0.00048145576137829406,
      "loss": 5.5193,
      "mean_token_accuracy": 0.15578468888998032,
      "num_tokens": 30172518.0,
      "step": 16360
    },
    {
      "entropy": 5.685378837585449,
      "epoch": 1.37487922705314,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004814438543439195,
      "loss": 5.4457,
      "mean_token_accuracy": 0.17112136483192444,
      "num_tokens": 30183124.0,
      "step": 16365
    },
    {
      "entropy": 5.700458574295044,
      "epoch": 1.375299306868305,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004814319436524984,
      "loss": 5.3898,
      "mean_token_accuracy": 0.1719352975487709,
      "num_tokens": 30191861.0,
      "step": 16370
    },
    {
      "entropy": 5.552261209487915,
      "epoch": 1.3757193866834698,
      "grad_norm": 1.625,
      "learning_rate": 0.00048142002930424174,
      "loss": 5.2934,
      "mean_token_accuracy": 0.16797118484973908,
      "num_tokens": 30200308.0,
      "step": 16375
    },
    {
      "entropy": 5.676715898513794,
      "epoch": 1.3761394664986346,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004814081112993605,
      "loss": 5.4149,
      "mean_token_accuracy": 0.17060285806655884,
      "num_tokens": 30209380.0,
      "step": 16380
    },
    {
      "entropy": 5.703636837005615,
      "epoch": 1.3765595463137996,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004813961896380659,
      "loss": 5.5023,
      "mean_token_accuracy": 0.1652629256248474,
      "num_tokens": 30218549.0,
      "step": 16385
    },
    {
      "entropy": 5.668811988830567,
      "epoch": 1.3769796261289646,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004813842643205691,
      "loss": 5.4297,
      "mean_token_accuracy": 0.16347114741802216,
      "num_tokens": 30228119.0,
      "step": 16390
    },
    {
      "entropy": 5.704423522949218,
      "epoch": 1.3773997059441294,
      "grad_norm": 2.125,
      "learning_rate": 0.0004813723353470813,
      "loss": 5.4086,
      "mean_token_accuracy": 0.16577320992946626,
      "num_tokens": 30236765.0,
      "step": 16395
    },
    {
      "entropy": 5.680452489852906,
      "epoch": 1.3778197857592942,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004813604027178139,
      "loss": 5.3459,
      "mean_token_accuracy": 0.16742168515920638,
      "num_tokens": 30246089.0,
      "step": 16400
    },
    {
      "entropy": 5.632819128036499,
      "epoch": 1.3782398655744592,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048134846643297817,
      "loss": 5.464,
      "mean_token_accuracy": 0.1612554356455803,
      "num_tokens": 30255806.0,
      "step": 16405
    },
    {
      "entropy": 5.71699275970459,
      "epoch": 1.378659945389624,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004813365264927856,
      "loss": 5.5247,
      "mean_token_accuracy": 0.1526763767004013,
      "num_tokens": 30267112.0,
      "step": 16410
    },
    {
      "entropy": 5.667221641540527,
      "epoch": 1.379080025204789,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004813245828974477,
      "loss": 5.3842,
      "mean_token_accuracy": 0.1672530382871628,
      "num_tokens": 30276168.0,
      "step": 16415
    },
    {
      "entropy": 5.661026334762573,
      "epoch": 1.3795001050199538,
      "grad_norm": 1.625,
      "learning_rate": 0.0004813126356471761,
      "loss": 5.4154,
      "mean_token_accuracy": 0.16718480885028839,
      "num_tokens": 30285723.0,
      "step": 16420
    },
    {
      "entropy": 5.776949548721314,
      "epoch": 1.3799201848351186,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004813006847421824,
      "loss": 5.4541,
      "mean_token_accuracy": 0.1688556745648384,
      "num_tokens": 30294790.0,
      "step": 16425
    },
    {
      "entropy": 5.642568206787109,
      "epoch": 1.3803402646502836,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004812887301826783,
      "loss": 5.3799,
      "mean_token_accuracy": 0.16735500395298003,
      "num_tokens": 30303439.0,
      "step": 16430
    },
    {
      "entropy": 5.613499641418457,
      "epoch": 1.3807603444654484,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004812767719688755,
      "loss": 5.3822,
      "mean_token_accuracy": 0.15868924260139466,
      "num_tokens": 30312493.0,
      "step": 16435
    },
    {
      "entropy": 5.6956870555877686,
      "epoch": 1.3811804242806134,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004812648101009859,
      "loss": 5.4251,
      "mean_token_accuracy": 0.17203418612480165,
      "num_tokens": 30321637.0,
      "step": 16440
    },
    {
      "entropy": 5.757980585098267,
      "epoch": 1.3816005040957782,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004812528445792215,
      "loss": 5.5473,
      "mean_token_accuracy": 0.1549142688512802,
      "num_tokens": 30330730.0,
      "step": 16445
    },
    {
      "entropy": 5.631748342514038,
      "epoch": 1.382020583910943,
      "grad_norm": 1.5,
      "learning_rate": 0.00048124087540379407,
      "loss": 5.3721,
      "mean_token_accuracy": 0.16817360520362853,
      "num_tokens": 30339568.0,
      "step": 16450
    },
    {
      "entropy": 5.633518838882447,
      "epoch": 1.382440663726108,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00048122890257491573,
      "loss": 5.4227,
      "mean_token_accuracy": 0.16262441575527192,
      "num_tokens": 30349225.0,
      "step": 16455
    },
    {
      "entropy": 5.706428813934326,
      "epoch": 1.382860743541273,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048121692609279866,
      "loss": 5.4231,
      "mean_token_accuracy": 0.1687069818377495,
      "num_tokens": 30358804.0,
      "step": 16460
    },
    {
      "entropy": 5.686781787872315,
      "epoch": 1.3832808233564378,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004812049459576549,
      "loss": 5.4869,
      "mean_token_accuracy": 0.16923241466283798,
      "num_tokens": 30368490.0,
      "step": 16465
    },
    {
      "entropy": 5.763137435913086,
      "epoch": 1.3837009031716025,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004811929621696966,
      "loss": 5.3709,
      "mean_token_accuracy": 0.16324265003204347,
      "num_tokens": 30377117.0,
      "step": 16470
    },
    {
      "entropy": 5.576843166351319,
      "epoch": 1.3841209829867676,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048118097472913627,
      "loss": 5.2725,
      "mean_token_accuracy": 0.17726120352745056,
      "num_tokens": 30385151.0,
      "step": 16475
    },
    {
      "entropy": 5.522637939453125,
      "epoch": 1.3845410628019323,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004811689836361861,
      "loss": 5.2973,
      "mean_token_accuracy": 0.1718326285481453,
      "num_tokens": 30394837.0,
      "step": 16480
    },
    {
      "entropy": 5.665565729141235,
      "epoch": 1.3849611426170974,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004811569888910585,
      "loss": 5.3868,
      "mean_token_accuracy": 0.17299704551696776,
      "num_tokens": 30403507.0,
      "step": 16485
    },
    {
      "entropy": 5.618129730224609,
      "epoch": 1.3853812224322621,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004811449904939661,
      "loss": 5.3707,
      "mean_token_accuracy": 0.172273900359869,
      "num_tokens": 30412941.0,
      "step": 16490
    },
    {
      "entropy": 5.627029800415039,
      "epoch": 1.385801302247427,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048113298844512127,
      "loss": 5.3529,
      "mean_token_accuracy": 0.1789866179227829,
      "num_tokens": 30421823.0,
      "step": 16495
    },
    {
      "entropy": 5.669685316085816,
      "epoch": 1.386221382062592,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004811209827447367,
      "loss": 5.4705,
      "mean_token_accuracy": 0.1607930526137352,
      "num_tokens": 30431901.0,
      "step": 16500
    },
    {
      "entropy": 5.660027360916137,
      "epoch": 1.3866414618777567,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048110897339302504,
      "loss": 5.3853,
      "mean_token_accuracy": 0.1660548597574234,
      "num_tokens": 30442037.0,
      "step": 16505
    },
    {
      "entropy": 5.662465572357178,
      "epoch": 1.3870615416929217,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048109696039019915,
      "loss": 5.3702,
      "mean_token_accuracy": 0.17271846532821655,
      "num_tokens": 30451189.0,
      "step": 16510
    },
    {
      "entropy": 5.68119535446167,
      "epoch": 1.3874816215080865,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004810849437364716,
      "loss": 5.4395,
      "mean_token_accuracy": 0.1687230348587036,
      "num_tokens": 30460214.0,
      "step": 16515
    },
    {
      "entropy": 5.703924751281738,
      "epoch": 1.3879017013232513,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048107292343205546,
      "loss": 5.4496,
      "mean_token_accuracy": 0.1623838424682617,
      "num_tokens": 30469936.0,
      "step": 16520
    },
    {
      "entropy": 5.6473335266113285,
      "epoch": 1.3883217811384163,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004810608994771636,
      "loss": 5.3955,
      "mean_token_accuracy": 0.16575901359319686,
      "num_tokens": 30479282.0,
      "step": 16525
    },
    {
      "entropy": 5.694073867797852,
      "epoch": 1.388741860953581,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000481048871872009,
      "loss": 5.4413,
      "mean_token_accuracy": 0.16611346304416658,
      "num_tokens": 30487839.0,
      "step": 16530
    },
    {
      "entropy": 5.725026845932007,
      "epoch": 1.389161940768746,
      "grad_norm": 1.703125,
      "learning_rate": 0.00048103684061680463,
      "loss": 5.4804,
      "mean_token_accuracy": 0.1627011075615883,
      "num_tokens": 30497327.0,
      "step": 16535
    },
    {
      "entropy": 5.669433879852295,
      "epoch": 1.389582020583911,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048102480571176384,
      "loss": 5.3926,
      "mean_token_accuracy": 0.16877335906028748,
      "num_tokens": 30506996.0,
      "step": 16540
    },
    {
      "entropy": 5.659185171127319,
      "epoch": 1.390002100399076,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004810127671570997,
      "loss": 5.3141,
      "mean_token_accuracy": 0.17210612148046495,
      "num_tokens": 30515627.0,
      "step": 16545
    },
    {
      "entropy": 5.720813179016114,
      "epoch": 1.3904221802142407,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048100072495302544,
      "loss": 5.4572,
      "mean_token_accuracy": 0.16103183925151826,
      "num_tokens": 30525858.0,
      "step": 16550
    },
    {
      "entropy": 5.610604095458984,
      "epoch": 1.3908422600294057,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004809886790997544,
      "loss": 5.3545,
      "mean_token_accuracy": 0.17273420989513397,
      "num_tokens": 30536331.0,
      "step": 16555
    },
    {
      "entropy": 5.624381971359253,
      "epoch": 1.3912623398445705,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004809766295975,
      "loss": 5.3914,
      "mean_token_accuracy": 0.16571607887744905,
      "num_tokens": 30545329.0,
      "step": 16560
    },
    {
      "entropy": 5.64850263595581,
      "epoch": 1.3916824196597353,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004809645764464757,
      "loss": 5.3438,
      "mean_token_accuracy": 0.16861116588115693,
      "num_tokens": 30554357.0,
      "step": 16565
    },
    {
      "entropy": 5.687424087524414,
      "epoch": 1.3921024994749003,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048095251964689494,
      "loss": 5.5077,
      "mean_token_accuracy": 0.1639639601111412,
      "num_tokens": 30563548.0,
      "step": 16570
    },
    {
      "entropy": 5.633093214035034,
      "epoch": 1.392522579290065,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048094045919897134,
      "loss": 5.3943,
      "mean_token_accuracy": 0.16770315915346146,
      "num_tokens": 30572844.0,
      "step": 16575
    },
    {
      "entropy": 5.584927940368653,
      "epoch": 1.39294265910523,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004809283951029185,
      "loss": 5.3214,
      "mean_token_accuracy": 0.17551473677158355,
      "num_tokens": 30580930.0,
      "step": 16580
    },
    {
      "entropy": 5.675092315673828,
      "epoch": 1.3933627389203949,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004809163273589503,
      "loss": 5.3292,
      "mean_token_accuracy": 0.17460571080446244,
      "num_tokens": 30589917.0,
      "step": 16585
    },
    {
      "entropy": 5.625373315811157,
      "epoch": 1.3937828187355596,
      "grad_norm": 1.5,
      "learning_rate": 0.00048090425596728035,
      "loss": 5.4199,
      "mean_token_accuracy": 0.16480701714754104,
      "num_tokens": 30599282.0,
      "step": 16590
    },
    {
      "entropy": 5.65163745880127,
      "epoch": 1.3942028985507247,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048089218092812254,
      "loss": 5.4304,
      "mean_token_accuracy": 0.16113353818655013,
      "num_tokens": 30608244.0,
      "step": 16595
    },
    {
      "entropy": 5.724131345748901,
      "epoch": 1.3946229783658894,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048088010224169064,
      "loss": 5.5432,
      "mean_token_accuracy": 0.1666365347802639,
      "num_tokens": 30617340.0,
      "step": 16600
    },
    {
      "entropy": 5.750015497207642,
      "epoch": 1.3950430581810545,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048086801990819886,
      "loss": 5.4522,
      "mean_token_accuracy": 0.16224676370620728,
      "num_tokens": 30626244.0,
      "step": 16605
    },
    {
      "entropy": 5.699167776107788,
      "epoch": 1.3954631379962192,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048085593392786113,
      "loss": 5.433,
      "mean_token_accuracy": 0.17035846263170243,
      "num_tokens": 30635279.0,
      "step": 16610
    },
    {
      "entropy": 5.704600381851196,
      "epoch": 1.395883217811384,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004808438443008915,
      "loss": 5.5861,
      "mean_token_accuracy": 0.1579027071595192,
      "num_tokens": 30645790.0,
      "step": 16615
    },
    {
      "entropy": 5.651020908355713,
      "epoch": 1.396303297626549,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004808317510275041,
      "loss": 5.3936,
      "mean_token_accuracy": 0.16329846829175948,
      "num_tokens": 30654497.0,
      "step": 16620
    },
    {
      "entropy": 5.6952495098114015,
      "epoch": 1.396723377441714,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004808196541079133,
      "loss": 5.4784,
      "mean_token_accuracy": 0.15707603394985198,
      "num_tokens": 30663760.0,
      "step": 16625
    },
    {
      "entropy": 5.660111951828003,
      "epoch": 1.3971434572568788,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048080755354233326,
      "loss": 5.4888,
      "mean_token_accuracy": 0.16803241819143294,
      "num_tokens": 30674263.0,
      "step": 16630
    },
    {
      "entropy": 5.663553285598755,
      "epoch": 1.3975635370720436,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004807954493309784,
      "loss": 5.3481,
      "mean_token_accuracy": 0.17060723453760146,
      "num_tokens": 30683501.0,
      "step": 16635
    },
    {
      "entropy": 5.662658262252807,
      "epoch": 1.3979836168872086,
      "grad_norm": 1.890625,
      "learning_rate": 0.00048078334147406314,
      "loss": 5.3451,
      "mean_token_accuracy": 0.17941270768642426,
      "num_tokens": 30691917.0,
      "step": 16640
    },
    {
      "entropy": 5.650649309158325,
      "epoch": 1.3984036967023734,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048077122997180197,
      "loss": 5.4363,
      "mean_token_accuracy": 0.16697393506765365,
      "num_tokens": 30701753.0,
      "step": 16645
    },
    {
      "entropy": 5.5801185131072994,
      "epoch": 1.3988237765175384,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004807591148244093,
      "loss": 5.4032,
      "mean_token_accuracy": 0.16500199735164642,
      "num_tokens": 30710878.0,
      "step": 16650
    },
    {
      "entropy": 5.643789529800415,
      "epoch": 1.3992438563327032,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004807469960321,
      "loss": 5.2945,
      "mean_token_accuracy": 0.1748718097805977,
      "num_tokens": 30719372.0,
      "step": 16655
    },
    {
      "entropy": 5.688783597946167,
      "epoch": 1.399663936147868,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048073487359508854,
      "loss": 5.4628,
      "mean_token_accuracy": 0.16029228419065475,
      "num_tokens": 30728529.0,
      "step": 16660
    },
    {
      "entropy": 5.726004886627197,
      "epoch": 1.400084015963033,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048072274751358976,
      "loss": 5.4148,
      "mean_token_accuracy": 0.1690065696835518,
      "num_tokens": 30737704.0,
      "step": 16665
    },
    {
      "entropy": 5.6314037322998045,
      "epoch": 1.4005040957781978,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048071061778781843,
      "loss": 5.385,
      "mean_token_accuracy": 0.16517933458089828,
      "num_tokens": 30747836.0,
      "step": 16670
    },
    {
      "entropy": 5.638839244842529,
      "epoch": 1.4009241755933628,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004806984844179894,
      "loss": 5.4406,
      "mean_token_accuracy": 0.1636255532503128,
      "num_tokens": 30757881.0,
      "step": 16675
    },
    {
      "entropy": 5.710208129882813,
      "epoch": 1.4013442554085276,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048068634740431774,
      "loss": 5.4362,
      "mean_token_accuracy": 0.16064029261469842,
      "num_tokens": 30767592.0,
      "step": 16680
    },
    {
      "entropy": 5.66591444015503,
      "epoch": 1.4017643352236924,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004806742067470182,
      "loss": 5.4298,
      "mean_token_accuracy": 0.17142573595046998,
      "num_tokens": 30776633.0,
      "step": 16685
    },
    {
      "entropy": 5.671479415893555,
      "epoch": 1.4021844150388574,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048066206244630613,
      "loss": 5.3659,
      "mean_token_accuracy": 0.1664132371544838,
      "num_tokens": 30785195.0,
      "step": 16690
    },
    {
      "entropy": 5.544547271728516,
      "epoch": 1.4026044948540224,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048064991450239643,
      "loss": 5.3734,
      "mean_token_accuracy": 0.16558644473552703,
      "num_tokens": 30794397.0,
      "step": 16695
    },
    {
      "entropy": 5.755392265319824,
      "epoch": 1.4030245746691872,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00048063776291550444,
      "loss": 5.5283,
      "mean_token_accuracy": 0.16150481402873992,
      "num_tokens": 30803312.0,
      "step": 16700
    },
    {
      "entropy": 5.715979433059692,
      "epoch": 1.403444654484352,
      "grad_norm": 1.609375,
      "learning_rate": 0.00048062560768584537,
      "loss": 5.4309,
      "mean_token_accuracy": 0.1731870412826538,
      "num_tokens": 30812519.0,
      "step": 16705
    },
    {
      "entropy": 5.607736825942993,
      "epoch": 1.403864734299517,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048061344881363444,
      "loss": 5.3721,
      "mean_token_accuracy": 0.17038303017616271,
      "num_tokens": 30821558.0,
      "step": 16710
    },
    {
      "entropy": 5.6497814655303955,
      "epoch": 1.4042848141146818,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004806012862990873,
      "loss": 5.4012,
      "mean_token_accuracy": 0.16469183415174485,
      "num_tokens": 30831521.0,
      "step": 16715
    },
    {
      "entropy": 5.682461547851562,
      "epoch": 1.4047048939298468,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048058912014241914,
      "loss": 5.3868,
      "mean_token_accuracy": 0.1683809146285057,
      "num_tokens": 30841191.0,
      "step": 16720
    },
    {
      "entropy": 5.689334106445313,
      "epoch": 1.4051249737450116,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004805769503438456,
      "loss": 5.4883,
      "mean_token_accuracy": 0.1661745771765709,
      "num_tokens": 30850556.0,
      "step": 16725
    },
    {
      "entropy": 5.671855068206787,
      "epoch": 1.4055450535601763,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048056477690358227,
      "loss": 5.3843,
      "mean_token_accuracy": 0.17183898240327836,
      "num_tokens": 30859410.0,
      "step": 16730
    },
    {
      "entropy": 5.730327177047729,
      "epoch": 1.4059651333753413,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004805525998218447,
      "loss": 5.4243,
      "mean_token_accuracy": 0.16271052360534669,
      "num_tokens": 30868048.0,
      "step": 16735
    },
    {
      "entropy": 5.665938663482666,
      "epoch": 1.4063852131905061,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048054041909884873,
      "loss": 5.4339,
      "mean_token_accuracy": 0.16629239469766616,
      "num_tokens": 30876785.0,
      "step": 16740
    },
    {
      "entropy": 5.724353694915772,
      "epoch": 1.4068052930056711,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048052823473481007,
      "loss": 5.4949,
      "mean_token_accuracy": 0.16633669435977935,
      "num_tokens": 30886158.0,
      "step": 16745
    },
    {
      "entropy": 5.70852746963501,
      "epoch": 1.407225372820836,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048051604672994446,
      "loss": 5.3455,
      "mean_token_accuracy": 0.16650964617729186,
      "num_tokens": 30895283.0,
      "step": 16750
    },
    {
      "entropy": 5.654937887191773,
      "epoch": 1.4076454526360007,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048050385508446804,
      "loss": 5.3971,
      "mean_token_accuracy": 0.16910174041986464,
      "num_tokens": 30905514.0,
      "step": 16755
    },
    {
      "entropy": 5.6511979579925535,
      "epoch": 1.4080655324511657,
      "grad_norm": 2.6875,
      "learning_rate": 0.00048049165979859655,
      "loss": 5.3137,
      "mean_token_accuracy": 0.1823897883296013,
      "num_tokens": 30914794.0,
      "step": 16760
    },
    {
      "entropy": 5.641238451004028,
      "epoch": 1.4084856122663307,
      "grad_norm": 1.6875,
      "learning_rate": 0.00048047946087254615,
      "loss": 5.3444,
      "mean_token_accuracy": 0.16592412441968918,
      "num_tokens": 30923823.0,
      "step": 16765
    },
    {
      "entropy": 5.5764069080352785,
      "epoch": 1.4089056920814955,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048046725830653295,
      "loss": 5.4609,
      "mean_token_accuracy": 0.16102780997753144,
      "num_tokens": 30932738.0,
      "step": 16770
    },
    {
      "entropy": 5.649274396896362,
      "epoch": 1.4093257718966603,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048045505210077304,
      "loss": 5.4608,
      "mean_token_accuracy": 0.16147117018699647,
      "num_tokens": 30942302.0,
      "step": 16775
    },
    {
      "entropy": 5.7223867893219,
      "epoch": 1.4097458517118253,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004804428422554826,
      "loss": 5.3816,
      "mean_token_accuracy": 0.16938356906175614,
      "num_tokens": 30951662.0,
      "step": 16780
    },
    {
      "entropy": 5.656762599945068,
      "epoch": 1.41016593152699,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004804306287708782,
      "loss": 5.399,
      "mean_token_accuracy": 0.1710711047053337,
      "num_tokens": 30960475.0,
      "step": 16785
    },
    {
      "entropy": 5.609744453430176,
      "epoch": 1.410586011342155,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048041841164717574,
      "loss": 5.2329,
      "mean_token_accuracy": 0.17638496905565262,
      "num_tokens": 30969075.0,
      "step": 16790
    },
    {
      "entropy": 5.563787126541138,
      "epoch": 1.41100609115732,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004804061908845921,
      "loss": 5.3195,
      "mean_token_accuracy": 0.17508750706911086,
      "num_tokens": 30978030.0,
      "step": 16795
    },
    {
      "entropy": 5.605288314819336,
      "epoch": 1.4114261709724847,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00048039396648334346,
      "loss": 5.297,
      "mean_token_accuracy": 0.17446033656597137,
      "num_tokens": 30985639.0,
      "step": 16800
    },
    {
      "entropy": 5.695837736129761,
      "epoch": 1.4118462507876497,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004803817384436465,
      "loss": 5.4342,
      "mean_token_accuracy": 0.1659676179289818,
      "num_tokens": 30994811.0,
      "step": 16805
    },
    {
      "entropy": 5.71638503074646,
      "epoch": 1.4122663306028145,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004803695067657178,
      "loss": 5.4134,
      "mean_token_accuracy": 0.17041690796613693,
      "num_tokens": 31003813.0,
      "step": 16810
    },
    {
      "entropy": 5.652092409133911,
      "epoch": 1.4126864104179795,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000480357271449774,
      "loss": 5.3548,
      "mean_token_accuracy": 0.17870038002729416,
      "num_tokens": 31012488.0,
      "step": 16815
    },
    {
      "entropy": 5.671490907669067,
      "epoch": 1.4131064902331443,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004803450324960318,
      "loss": 5.3671,
      "mean_token_accuracy": 0.16869506388902664,
      "num_tokens": 31021089.0,
      "step": 16820
    },
    {
      "entropy": 5.631192970275879,
      "epoch": 1.413526570048309,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048033278990470825,
      "loss": 5.382,
      "mean_token_accuracy": 0.16601116210222244,
      "num_tokens": 31029903.0,
      "step": 16825
    },
    {
      "entropy": 5.624605464935303,
      "epoch": 1.413946649863474,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00048032054367601996,
      "loss": 5.3891,
      "mean_token_accuracy": 0.16890241354703903,
      "num_tokens": 31039207.0,
      "step": 16830
    },
    {
      "entropy": 5.647729778289795,
      "epoch": 1.414366729678639,
      "grad_norm": 1.5,
      "learning_rate": 0.00048030829381018396,
      "loss": 5.4254,
      "mean_token_accuracy": 0.16352934539318084,
      "num_tokens": 31048190.0,
      "step": 16835
    },
    {
      "entropy": 5.704311561584473,
      "epoch": 1.4147868094938039,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004802960403074173,
      "loss": 5.5011,
      "mean_token_accuracy": 0.16595527827739714,
      "num_tokens": 31058769.0,
      "step": 16840
    },
    {
      "entropy": 5.671142196655273,
      "epoch": 1.4152068893089687,
      "grad_norm": 1.625,
      "learning_rate": 0.00048028378316793705,
      "loss": 5.4486,
      "mean_token_accuracy": 0.16351300925016404,
      "num_tokens": 31066830.0,
      "step": 16845
    },
    {
      "entropy": 5.732022857666015,
      "epoch": 1.4156269691241337,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004802715223919602,
      "loss": 5.4978,
      "mean_token_accuracy": 0.16644495725631714,
      "num_tokens": 31077205.0,
      "step": 16850
    },
    {
      "entropy": 5.704752731323242,
      "epoch": 1.4160470489392984,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048025925797970403,
      "loss": 5.4246,
      "mean_token_accuracy": 0.17204785197973252,
      "num_tokens": 31087327.0,
      "step": 16855
    },
    {
      "entropy": 5.615525102615356,
      "epoch": 1.4164671287544635,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048024698993138587,
      "loss": 5.3676,
      "mean_token_accuracy": 0.168218894302845,
      "num_tokens": 31096501.0,
      "step": 16860
    },
    {
      "entropy": 5.723089742660522,
      "epoch": 1.4168872085696282,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048023471824722294,
      "loss": 5.5522,
      "mean_token_accuracy": 0.15669072568416595,
      "num_tokens": 31105949.0,
      "step": 16865
    },
    {
      "entropy": 5.754108715057373,
      "epoch": 1.417307288384793,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048022244292743256,
      "loss": 5.4554,
      "mean_token_accuracy": 0.15739416480064392,
      "num_tokens": 31115482.0,
      "step": 16870
    },
    {
      "entropy": 5.677294015884399,
      "epoch": 1.417727368199958,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048021016397223234,
      "loss": 5.3916,
      "mean_token_accuracy": 0.17019754350185395,
      "num_tokens": 31124758.0,
      "step": 16875
    },
    {
      "entropy": 5.607404279708862,
      "epoch": 1.4181474480151228,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048019788138183977,
      "loss": 5.2828,
      "mean_token_accuracy": 0.17816756069660186,
      "num_tokens": 31134114.0,
      "step": 16880
    },
    {
      "entropy": 5.5718903064727785,
      "epoch": 1.4185675278302878,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048018559515647244,
      "loss": 5.3193,
      "mean_token_accuracy": 0.16984036862850188,
      "num_tokens": 31142667.0,
      "step": 16885
    },
    {
      "entropy": 5.651354122161865,
      "epoch": 1.4189876076454526,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048017330529634785,
      "loss": 5.427,
      "mean_token_accuracy": 0.162412828207016,
      "num_tokens": 31152105.0,
      "step": 16890
    },
    {
      "entropy": 5.681749629974365,
      "epoch": 1.4194076874606174,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048016101180168376,
      "loss": 5.3926,
      "mean_token_accuracy": 0.17753778100013734,
      "num_tokens": 31160277.0,
      "step": 16895
    },
    {
      "entropy": 5.787999868392944,
      "epoch": 1.4198277672757824,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048014871467269804,
      "loss": 5.6082,
      "mean_token_accuracy": 0.15824506431818008,
      "num_tokens": 31170677.0,
      "step": 16900
    },
    {
      "entropy": 5.663241958618164,
      "epoch": 1.4202478470909472,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048013641390960856,
      "loss": 5.3841,
      "mean_token_accuracy": 0.16147300899028777,
      "num_tokens": 31179298.0,
      "step": 16905
    },
    {
      "entropy": 5.6704291820526125,
      "epoch": 1.4206679269061122,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004801241095126331,
      "loss": 5.4128,
      "mean_token_accuracy": 0.16424123346805572,
      "num_tokens": 31188547.0,
      "step": 16910
    },
    {
      "entropy": 5.667984914779663,
      "epoch": 1.421088006721277,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004801118014819896,
      "loss": 5.3993,
      "mean_token_accuracy": 0.16866909414529802,
      "num_tokens": 31197680.0,
      "step": 16915
    },
    {
      "entropy": 5.615919971466065,
      "epoch": 1.421508086536442,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004800994898178962,
      "loss": 5.355,
      "mean_token_accuracy": 0.17297782152891159,
      "num_tokens": 31206351.0,
      "step": 16920
    },
    {
      "entropy": 5.640166568756103,
      "epoch": 1.4219281663516068,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004800871745205708,
      "loss": 5.5521,
      "mean_token_accuracy": 0.16399392560124398,
      "num_tokens": 31216478.0,
      "step": 16925
    },
    {
      "entropy": 5.783535957336426,
      "epoch": 1.4223482461667718,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048007485559023195,
      "loss": 5.5224,
      "mean_token_accuracy": 0.16302904337644578,
      "num_tokens": 31225920.0,
      "step": 16930
    },
    {
      "entropy": 5.646339702606201,
      "epoch": 1.4227683259819366,
      "grad_norm": 1.875,
      "learning_rate": 0.0004800625330270975,
      "loss": 5.3783,
      "mean_token_accuracy": 0.17351202368736268,
      "num_tokens": 31235061.0,
      "step": 16935
    },
    {
      "entropy": 5.580986547470093,
      "epoch": 1.4231884057971014,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004800502068313859,
      "loss": 5.3557,
      "mean_token_accuracy": 0.16914222687482833,
      "num_tokens": 31243448.0,
      "step": 16940
    },
    {
      "entropy": 5.717418241500854,
      "epoch": 1.4236084856122664,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004800378770033154,
      "loss": 5.4732,
      "mean_token_accuracy": 0.1651209145784378,
      "num_tokens": 31252569.0,
      "step": 16945
    },
    {
      "entropy": 5.6657874584198,
      "epoch": 1.4240285654274312,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004800255435431046,
      "loss": 5.3537,
      "mean_token_accuracy": 0.16902960389852523,
      "num_tokens": 31261905.0,
      "step": 16950
    },
    {
      "entropy": 5.585388231277466,
      "epoch": 1.4244486452425962,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048001320645097177,
      "loss": 5.3409,
      "mean_token_accuracy": 0.172618305683136,
      "num_tokens": 31271203.0,
      "step": 16955
    },
    {
      "entropy": 5.582739210128784,
      "epoch": 1.424868725057761,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048000086572713566,
      "loss": 5.3308,
      "mean_token_accuracy": 0.1773141622543335,
      "num_tokens": 31279812.0,
      "step": 16960
    },
    {
      "entropy": 5.618875360488891,
      "epoch": 1.4252888048729258,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004799885213718147,
      "loss": 5.3742,
      "mean_token_accuracy": 0.16394567787647246,
      "num_tokens": 31289615.0,
      "step": 16965
    },
    {
      "entropy": 5.646689176559448,
      "epoch": 1.4257088846880908,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00047997617338522763,
      "loss": 5.3175,
      "mean_token_accuracy": 0.1717886134982109,
      "num_tokens": 31298947.0,
      "step": 16970
    },
    {
      "entropy": 5.650730991363526,
      "epoch": 1.4261289645032555,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047996382176759324,
      "loss": 5.3205,
      "mean_token_accuracy": 0.16948004364967345,
      "num_tokens": 31307465.0,
      "step": 16975
    },
    {
      "entropy": 5.618907785415649,
      "epoch": 1.4265490443184206,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004799514665191303,
      "loss": 5.4339,
      "mean_token_accuracy": 0.16848465651273728,
      "num_tokens": 31317682.0,
      "step": 16980
    },
    {
      "entropy": 5.731729412078858,
      "epoch": 1.4269691241335853,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004799391076400576,
      "loss": 5.429,
      "mean_token_accuracy": 0.16680052876472473,
      "num_tokens": 31326113.0,
      "step": 16985
    },
    {
      "entropy": 5.770191287994384,
      "epoch": 1.4273892039487501,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047992674513059415,
      "loss": 5.4598,
      "mean_token_accuracy": 0.170396788418293,
      "num_tokens": 31335263.0,
      "step": 16990
    },
    {
      "entropy": 5.6245646476745605,
      "epoch": 1.4278092837639151,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047991437899095896,
      "loss": 5.4127,
      "mean_token_accuracy": 0.1713246524333954,
      "num_tokens": 31344503.0,
      "step": 16995
    },
    {
      "entropy": 5.618230819702148,
      "epoch": 1.4282293635790801,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047990200922137105,
      "loss": 5.468,
      "mean_token_accuracy": 0.16611219495534896,
      "num_tokens": 31354530.0,
      "step": 17000
    },
    {
      "entropy": 5.615735769271851,
      "epoch": 1.428649443394245,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004798896358220496,
      "loss": 5.2714,
      "mean_token_accuracy": 0.1760880783200264,
      "num_tokens": 31362761.0,
      "step": 17005
    },
    {
      "entropy": 5.610041666030884,
      "epoch": 1.4290695232094097,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004798772587932137,
      "loss": 5.2965,
      "mean_token_accuracy": 0.16842328161001205,
      "num_tokens": 31372933.0,
      "step": 17010
    },
    {
      "entropy": 5.739865112304687,
      "epoch": 1.4294896030245747,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004798648781350826,
      "loss": 5.5222,
      "mean_token_accuracy": 0.1676481083035469,
      "num_tokens": 31382651.0,
      "step": 17015
    },
    {
      "entropy": 5.670643711090088,
      "epoch": 1.4299096828397395,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004798524938478758,
      "loss": 5.4387,
      "mean_token_accuracy": 0.16613756790757178,
      "num_tokens": 31392272.0,
      "step": 17020
    },
    {
      "entropy": 5.660736322402954,
      "epoch": 1.4303297626549045,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004798401059318124,
      "loss": 5.3502,
      "mean_token_accuracy": 0.16613146811723709,
      "num_tokens": 31400684.0,
      "step": 17025
    },
    {
      "entropy": 5.678120946884155,
      "epoch": 1.4307498424700693,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004798277143871122,
      "loss": 5.3353,
      "mean_token_accuracy": 0.17574092447757722,
      "num_tokens": 31409082.0,
      "step": 17030
    },
    {
      "entropy": 5.60793628692627,
      "epoch": 1.431169922285234,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004798153192139944,
      "loss": 5.3171,
      "mean_token_accuracy": 0.17482832223176956,
      "num_tokens": 31417415.0,
      "step": 17035
    },
    {
      "entropy": 5.6694670677185055,
      "epoch": 1.431590002100399,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004798029204126786,
      "loss": 5.468,
      "mean_token_accuracy": 0.16680483222007753,
      "num_tokens": 31427510.0,
      "step": 17040
    },
    {
      "entropy": 5.648876094818116,
      "epoch": 1.432010081915564,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004797905179833847,
      "loss": 5.322,
      "mean_token_accuracy": 0.1751502901315689,
      "num_tokens": 31436187.0,
      "step": 17045
    },
    {
      "entropy": 5.639628219604492,
      "epoch": 1.432430161730729,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004797781119263321,
      "loss": 5.3442,
      "mean_token_accuracy": 0.1648264065384865,
      "num_tokens": 31445179.0,
      "step": 17050
    },
    {
      "entropy": 5.7023780822753904,
      "epoch": 1.4328502415458937,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004797657022417408,
      "loss": 5.4128,
      "mean_token_accuracy": 0.16612943559885024,
      "num_tokens": 31454434.0,
      "step": 17055
    },
    {
      "entropy": 5.661962509155273,
      "epoch": 1.4332703213610585,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047975328892983045,
      "loss": 5.411,
      "mean_token_accuracy": 0.1663669228553772,
      "num_tokens": 31464202.0,
      "step": 17060
    },
    {
      "entropy": 5.563780021667481,
      "epoch": 1.4336904011762235,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047974087199082095,
      "loss": 5.3018,
      "mean_token_accuracy": 0.1784432828426361,
      "num_tokens": 31473158.0,
      "step": 17065
    },
    {
      "entropy": 5.61002836227417,
      "epoch": 1.4341104809913885,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00047972845142493244,
      "loss": 5.3377,
      "mean_token_accuracy": 0.15915263295173646,
      "num_tokens": 31482643.0,
      "step": 17070
    },
    {
      "entropy": 5.617830038070679,
      "epoch": 1.4345305608065533,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004797160272323848,
      "loss": 5.3904,
      "mean_token_accuracy": 0.17026168406009673,
      "num_tokens": 31492080.0,
      "step": 17075
    },
    {
      "entropy": 5.6294958114624025,
      "epoch": 1.434950640621718,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047970359941339815,
      "loss": 5.3662,
      "mean_token_accuracy": 0.1749309167265892,
      "num_tokens": 31501990.0,
      "step": 17080
    },
    {
      "entropy": 5.637498474121093,
      "epoch": 1.435370720436883,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004796911679681926,
      "loss": 5.4127,
      "mean_token_accuracy": 0.16695185154676437,
      "num_tokens": 31510548.0,
      "step": 17085
    },
    {
      "entropy": 5.697969961166382,
      "epoch": 1.4357908002520479,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047967873289698847,
      "loss": 5.367,
      "mean_token_accuracy": 0.16952662467956542,
      "num_tokens": 31518695.0,
      "step": 17090
    },
    {
      "entropy": 5.767078542709351,
      "epoch": 1.4362108800672129,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047966629420000595,
      "loss": 5.5195,
      "mean_token_accuracy": 0.16411047503352166,
      "num_tokens": 31528021.0,
      "step": 17095
    },
    {
      "entropy": 5.7415252208709715,
      "epoch": 1.4366309598823777,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004796538518774654,
      "loss": 5.4941,
      "mean_token_accuracy": 0.16071161776781082,
      "num_tokens": 31537786.0,
      "step": 17100
    },
    {
      "entropy": 5.671219158172607,
      "epoch": 1.4370510396975424,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047964140592958725,
      "loss": 5.4374,
      "mean_token_accuracy": 0.16812409907579423,
      "num_tokens": 31548006.0,
      "step": 17105
    },
    {
      "entropy": 5.671098279953003,
      "epoch": 1.4374711195127075,
      "grad_norm": 2.359375,
      "learning_rate": 0.000479628956356592,
      "loss": 5.3847,
      "mean_token_accuracy": 0.1713287577033043,
      "num_tokens": 31557042.0,
      "step": 17110
    },
    {
      "entropy": 5.748289775848389,
      "epoch": 1.4378911993278722,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004796165031587001,
      "loss": 5.4216,
      "mean_token_accuracy": 0.16564884930849075,
      "num_tokens": 31566661.0,
      "step": 17115
    },
    {
      "entropy": 5.710162782669068,
      "epoch": 1.4383112791430372,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004796040463361323,
      "loss": 5.3926,
      "mean_token_accuracy": 0.18157869428396226,
      "num_tokens": 31575724.0,
      "step": 17120
    },
    {
      "entropy": 5.659053802490234,
      "epoch": 1.438731358958202,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004795915858891091,
      "loss": 5.4726,
      "mean_token_accuracy": 0.17062633037567138,
      "num_tokens": 31585068.0,
      "step": 17125
    },
    {
      "entropy": 5.701984643936157,
      "epoch": 1.4391514387733668,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004795791218178514,
      "loss": 5.4932,
      "mean_token_accuracy": 0.16348476186394692,
      "num_tokens": 31594629.0,
      "step": 17130
    },
    {
      "entropy": 5.653209495544433,
      "epoch": 1.4395715185885318,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047956665412257984,
      "loss": 5.3583,
      "mean_token_accuracy": 0.16999612152576446,
      "num_tokens": 31603469.0,
      "step": 17135
    },
    {
      "entropy": 5.641184759140015,
      "epoch": 1.4399915984036968,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047955418280351526,
      "loss": 5.3148,
      "mean_token_accuracy": 0.17733933478593827,
      "num_tokens": 31611674.0,
      "step": 17140
    },
    {
      "entropy": 5.773440551757813,
      "epoch": 1.4404116782188616,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004795417078608788,
      "loss": 5.5871,
      "mean_token_accuracy": 0.15769852250814437,
      "num_tokens": 31621863.0,
      "step": 17145
    },
    {
      "entropy": 5.721550559997558,
      "epoch": 1.4408317580340264,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047952922929489126,
      "loss": 5.4341,
      "mean_token_accuracy": 0.16333554387092591,
      "num_tokens": 31630968.0,
      "step": 17150
    },
    {
      "entropy": 5.631154489517212,
      "epoch": 1.4412518378491914,
      "grad_norm": 1.765625,
      "learning_rate": 0.00047951674710577366,
      "loss": 5.4246,
      "mean_token_accuracy": 0.1666700839996338,
      "num_tokens": 31640643.0,
      "step": 17155
    },
    {
      "entropy": 5.550993490219116,
      "epoch": 1.4416719176643562,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047950426129374723,
      "loss": 5.314,
      "mean_token_accuracy": 0.17611225098371505,
      "num_tokens": 31648941.0,
      "step": 17160
    },
    {
      "entropy": 5.663345623016357,
      "epoch": 1.4420919974795212,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047949177185903314,
      "loss": 5.4271,
      "mean_token_accuracy": 0.17190447002649306,
      "num_tokens": 31658019.0,
      "step": 17165
    },
    {
      "entropy": 5.756283473968506,
      "epoch": 1.442512077294686,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004794792788018526,
      "loss": 5.4751,
      "mean_token_accuracy": 0.1569846749305725,
      "num_tokens": 31668050.0,
      "step": 17170
    },
    {
      "entropy": 5.691919469833374,
      "epoch": 1.4429321571098508,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000479466782122427,
      "loss": 5.3313,
      "mean_token_accuracy": 0.16783742904663085,
      "num_tokens": 31676727.0,
      "step": 17175
    },
    {
      "entropy": 5.665456914901734,
      "epoch": 1.4433522369250158,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047945428182097756,
      "loss": 5.4192,
      "mean_token_accuracy": 0.1627032019197941,
      "num_tokens": 31686205.0,
      "step": 17180
    },
    {
      "entropy": 5.676385736465454,
      "epoch": 1.4437723167401806,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047944177789772583,
      "loss": 5.4461,
      "mean_token_accuracy": 0.16626082956790925,
      "num_tokens": 31695521.0,
      "step": 17185
    },
    {
      "entropy": 5.83642258644104,
      "epoch": 1.4441923965553456,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004794292703528932,
      "loss": 5.4862,
      "mean_token_accuracy": 0.1566154882311821,
      "num_tokens": 31706606.0,
      "step": 17190
    },
    {
      "entropy": 5.726326513290405,
      "epoch": 1.4446124763705104,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047941675918670133,
      "loss": 5.5735,
      "mean_token_accuracy": 0.1578015848994255,
      "num_tokens": 31716881.0,
      "step": 17195
    },
    {
      "entropy": 5.613589954376221,
      "epoch": 1.4450325561856752,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004794042443993719,
      "loss": 5.3549,
      "mean_token_accuracy": 0.16491276174783706,
      "num_tokens": 31725878.0,
      "step": 17200
    },
    {
      "entropy": 5.658415508270264,
      "epoch": 1.4454526360008402,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004793917259911265,
      "loss": 5.4521,
      "mean_token_accuracy": 0.1677870571613312,
      "num_tokens": 31735033.0,
      "step": 17205
    },
    {
      "entropy": 5.629816246032715,
      "epoch": 1.445872715816005,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004793792039621869,
      "loss": 5.3883,
      "mean_token_accuracy": 0.17036439329385758,
      "num_tokens": 31744887.0,
      "step": 17210
    },
    {
      "entropy": 5.6929848194122314,
      "epoch": 1.44629279563117,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047936667831277504,
      "loss": 5.4597,
      "mean_token_accuracy": 0.15823331326246262,
      "num_tokens": 31754137.0,
      "step": 17215
    },
    {
      "entropy": 5.663027572631836,
      "epoch": 1.4467128754463348,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004793541490431126,
      "loss": 5.2765,
      "mean_token_accuracy": 0.1757225662469864,
      "num_tokens": 31763394.0,
      "step": 17220
    },
    {
      "entropy": 5.700362491607666,
      "epoch": 1.4471329552614998,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004793416161534216,
      "loss": 5.4095,
      "mean_token_accuracy": 0.1688271403312683,
      "num_tokens": 31771905.0,
      "step": 17225
    },
    {
      "entropy": 5.561145162582397,
      "epoch": 1.4475530350766646,
      "grad_norm": 1.5,
      "learning_rate": 0.00047932907964392423,
      "loss": 5.2381,
      "mean_token_accuracy": 0.17822422385215758,
      "num_tokens": 31780788.0,
      "step": 17230
    },
    {
      "entropy": 5.63405475616455,
      "epoch": 1.4479731148918296,
      "grad_norm": 1.5,
      "learning_rate": 0.00047931653951484234,
      "loss": 5.4126,
      "mean_token_accuracy": 0.16318247765302657,
      "num_tokens": 31790198.0,
      "step": 17235
    },
    {
      "entropy": 5.668708610534668,
      "epoch": 1.4483931947069943,
      "grad_norm": 1.5,
      "learning_rate": 0.00047930399576639815,
      "loss": 5.4155,
      "mean_token_accuracy": 0.17210718393325805,
      "num_tokens": 31799396.0,
      "step": 17240
    },
    {
      "entropy": 5.60841851234436,
      "epoch": 1.4488132745221591,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047929144839881386,
      "loss": 5.265,
      "mean_token_accuracy": 0.18170080929994584,
      "num_tokens": 31807680.0,
      "step": 17245
    },
    {
      "entropy": 5.692173480987549,
      "epoch": 1.4492333543373241,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047927889741231186,
      "loss": 5.411,
      "mean_token_accuracy": 0.16371279507875441,
      "num_tokens": 31817406.0,
      "step": 17250
    },
    {
      "entropy": 5.670788717269898,
      "epoch": 1.449653434152489,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047926634280711435,
      "loss": 5.3861,
      "mean_token_accuracy": 0.170447413623333,
      "num_tokens": 31826518.0,
      "step": 17255
    },
    {
      "entropy": 5.658002042770386,
      "epoch": 1.450073513967654,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004792537845834437,
      "loss": 5.4782,
      "mean_token_accuracy": 0.16245739459991454,
      "num_tokens": 31835538.0,
      "step": 17260
    },
    {
      "entropy": 5.616776990890503,
      "epoch": 1.4504935937828187,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004792412227415224,
      "loss": 5.3303,
      "mean_token_accuracy": 0.17447643131017684,
      "num_tokens": 31844899.0,
      "step": 17265
    },
    {
      "entropy": 5.608181142807007,
      "epoch": 1.4509136735979835,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047922865728157314,
      "loss": 5.3794,
      "mean_token_accuracy": 0.17140736877918245,
      "num_tokens": 31854322.0,
      "step": 17270
    },
    {
      "entropy": 5.621095943450928,
      "epoch": 1.4513337534131485,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004792160882038183,
      "loss": 5.3578,
      "mean_token_accuracy": 0.1676871508359909,
      "num_tokens": 31863657.0,
      "step": 17275
    },
    {
      "entropy": 5.666088962554932,
      "epoch": 1.4517538332283133,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004792035155084806,
      "loss": 5.3349,
      "mean_token_accuracy": 0.16719566136598588,
      "num_tokens": 31873468.0,
      "step": 17280
    },
    {
      "entropy": 5.64121379852295,
      "epoch": 1.4521739130434783,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047919093919578283,
      "loss": 5.4419,
      "mean_token_accuracy": 0.16885169148445128,
      "num_tokens": 31882391.0,
      "step": 17285
    },
    {
      "entropy": 5.628076696395874,
      "epoch": 1.452593992858643,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004791783592659476,
      "loss": 5.4375,
      "mean_token_accuracy": 0.16824038773775102,
      "num_tokens": 31891370.0,
      "step": 17290
    },
    {
      "entropy": 5.633961296081543,
      "epoch": 1.4530140726738079,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000479165775719198,
      "loss": 5.3757,
      "mean_token_accuracy": 0.1686269760131836,
      "num_tokens": 31900688.0,
      "step": 17295
    },
    {
      "entropy": 5.613438320159912,
      "epoch": 1.453434152488973,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00047915318855575674,
      "loss": 5.3925,
      "mean_token_accuracy": 0.1753884881734848,
      "num_tokens": 31909359.0,
      "step": 17300
    },
    {
      "entropy": 5.581719303131104,
      "epoch": 1.453854232304138,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047914059777584686,
      "loss": 5.3597,
      "mean_token_accuracy": 0.1750309631228447,
      "num_tokens": 31918529.0,
      "step": 17305
    },
    {
      "entropy": 5.633323049545288,
      "epoch": 1.4542743121193027,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047912800337969144,
      "loss": 5.4445,
      "mean_token_accuracy": 0.16532712876796724,
      "num_tokens": 31928310.0,
      "step": 17310
    },
    {
      "entropy": 5.642775583267212,
      "epoch": 1.4546943919344675,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047911540536751355,
      "loss": 5.3436,
      "mean_token_accuracy": 0.16816910803318025,
      "num_tokens": 31937077.0,
      "step": 17315
    },
    {
      "entropy": 5.733968687057495,
      "epoch": 1.4551144717496325,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004791028037395363,
      "loss": 5.4156,
      "mean_token_accuracy": 0.16309573352336884,
      "num_tokens": 31946023.0,
      "step": 17320
    },
    {
      "entropy": 5.550899887084961,
      "epoch": 1.4555345515647973,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047909019849598305,
      "loss": 5.2634,
      "mean_token_accuracy": 0.17997643649578093,
      "num_tokens": 31954741.0,
      "step": 17325
    },
    {
      "entropy": 5.593026351928711,
      "epoch": 1.4559546313799623,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047907758963707696,
      "loss": 5.3695,
      "mean_token_accuracy": 0.16652967631816865,
      "num_tokens": 31963516.0,
      "step": 17330
    },
    {
      "entropy": 5.681953620910645,
      "epoch": 1.456374711195127,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00047906497716304153,
      "loss": 5.3995,
      "mean_token_accuracy": 0.17053551524877547,
      "num_tokens": 31971917.0,
      "step": 17335
    },
    {
      "entropy": 5.67693305015564,
      "epoch": 1.4567947910102919,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004790523610741001,
      "loss": 5.4388,
      "mean_token_accuracy": 0.16347652077674865,
      "num_tokens": 31980718.0,
      "step": 17340
    },
    {
      "entropy": 5.707173824310303,
      "epoch": 1.4572148708254569,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047903974137047614,
      "loss": 5.3882,
      "mean_token_accuracy": 0.16744230091571807,
      "num_tokens": 31988664.0,
      "step": 17345
    },
    {
      "entropy": 5.699107837677002,
      "epoch": 1.4576349506406217,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047902711805239325,
      "loss": 5.4518,
      "mean_token_accuracy": 0.1658826529979706,
      "num_tokens": 31998415.0,
      "step": 17350
    },
    {
      "entropy": 5.696130228042603,
      "epoch": 1.4580550304557867,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047901449112007494,
      "loss": 5.4633,
      "mean_token_accuracy": 0.16892063170671462,
      "num_tokens": 32007915.0,
      "step": 17355
    },
    {
      "entropy": 5.620400428771973,
      "epoch": 1.4584751102709514,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047900186057374514,
      "loss": 5.4009,
      "mean_token_accuracy": 0.16876232475042344,
      "num_tokens": 32016582.0,
      "step": 17360
    },
    {
      "entropy": 5.607079744338989,
      "epoch": 1.4588951900861162,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047898922641362724,
      "loss": 5.3947,
      "mean_token_accuracy": 0.1683703899383545,
      "num_tokens": 32026008.0,
      "step": 17365
    },
    {
      "entropy": 5.710273742675781,
      "epoch": 1.4593152699012812,
      "grad_norm": 1.625,
      "learning_rate": 0.0004789765886399453,
      "loss": 5.426,
      "mean_token_accuracy": 0.16709780246019362,
      "num_tokens": 32034554.0,
      "step": 17370
    },
    {
      "entropy": 5.744694328308105,
      "epoch": 1.4597353497164463,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047896394725292313,
      "loss": 5.418,
      "mean_token_accuracy": 0.1785625696182251,
      "num_tokens": 32044003.0,
      "step": 17375
    },
    {
      "entropy": 5.630892181396485,
      "epoch": 1.460155429531611,
      "grad_norm": 1.875,
      "learning_rate": 0.00047895130225278473,
      "loss": 5.4049,
      "mean_token_accuracy": 0.16883346140384675,
      "num_tokens": 32053753.0,
      "step": 17380
    },
    {
      "entropy": 5.625681734085083,
      "epoch": 1.4605755093467758,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004789386536397539,
      "loss": 5.405,
      "mean_token_accuracy": 0.16639624238014222,
      "num_tokens": 32062459.0,
      "step": 17385
    },
    {
      "entropy": 5.734336090087891,
      "epoch": 1.4609955891619408,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004789260014140549,
      "loss": 5.504,
      "mean_token_accuracy": 0.16596467792987823,
      "num_tokens": 32072544.0,
      "step": 17390
    },
    {
      "entropy": 5.740262985229492,
      "epoch": 1.4614156689771056,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047891334557591177,
      "loss": 5.4336,
      "mean_token_accuracy": 0.16199401617050171,
      "num_tokens": 32082015.0,
      "step": 17395
    },
    {
      "entropy": 5.633042907714843,
      "epoch": 1.4618357487922706,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004789006861255488,
      "loss": 5.3592,
      "mean_token_accuracy": 0.17149775624275207,
      "num_tokens": 32091622.0,
      "step": 17400
    },
    {
      "entropy": 5.715030241012573,
      "epoch": 1.4622558286074354,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004788880230631901,
      "loss": 5.5433,
      "mean_token_accuracy": 0.1612006053328514,
      "num_tokens": 32102716.0,
      "step": 17405
    },
    {
      "entropy": 5.709442043304444,
      "epoch": 1.4626759084226002,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047887535638906005,
      "loss": 5.2897,
      "mean_token_accuracy": 0.1801255911588669,
      "num_tokens": 32111051.0,
      "step": 17410
    },
    {
      "entropy": 5.573637962341309,
      "epoch": 1.4630959882377652,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000478862686103383,
      "loss": 5.322,
      "mean_token_accuracy": 0.1755004644393921,
      "num_tokens": 32119781.0,
      "step": 17415
    },
    {
      "entropy": 5.622990322113037,
      "epoch": 1.46351606805293,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047885001220638354,
      "loss": 5.4177,
      "mean_token_accuracy": 0.17067472338676454,
      "num_tokens": 32128849.0,
      "step": 17420
    },
    {
      "entropy": 5.679988384246826,
      "epoch": 1.463936147868095,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047883733469828604,
      "loss": 5.4526,
      "mean_token_accuracy": 0.17174424827098847,
      "num_tokens": 32138046.0,
      "step": 17425
    },
    {
      "entropy": 5.790944051742554,
      "epoch": 1.4643562276832598,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047882465357931516,
      "loss": 5.5027,
      "mean_token_accuracy": 0.16203498095273972,
      "num_tokens": 32147994.0,
      "step": 17430
    },
    {
      "entropy": 5.733061122894287,
      "epoch": 1.4647763074984246,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004788119688496954,
      "loss": 5.4445,
      "mean_token_accuracy": 0.1703865870833397,
      "num_tokens": 32156835.0,
      "step": 17435
    },
    {
      "entropy": 5.662545680999756,
      "epoch": 1.4651963873135896,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004787992805096516,
      "loss": 5.3803,
      "mean_token_accuracy": 0.16614514589309692,
      "num_tokens": 32166751.0,
      "step": 17440
    },
    {
      "entropy": 5.646206092834473,
      "epoch": 1.4656164671287546,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047878658855940855,
      "loss": 5.4779,
      "mean_token_accuracy": 0.1631500840187073,
      "num_tokens": 32175705.0,
      "step": 17445
    },
    {
      "entropy": 5.7637755393981935,
      "epoch": 1.4660365469439194,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004787738929991909,
      "loss": 5.538,
      "mean_token_accuracy": 0.1581514447927475,
      "num_tokens": 32185404.0,
      "step": 17450
    },
    {
      "entropy": 5.700724649429321,
      "epoch": 1.4664566267590842,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047876119382922374,
      "loss": 5.3948,
      "mean_token_accuracy": 0.17208822518587114,
      "num_tokens": 32194054.0,
      "step": 17455
    },
    {
      "entropy": 5.664627313613892,
      "epoch": 1.4668767065742492,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047874849104973194,
      "loss": 5.4727,
      "mean_token_accuracy": 0.15883335024118422,
      "num_tokens": 32204080.0,
      "step": 17460
    },
    {
      "entropy": 5.637804985046387,
      "epoch": 1.467296786389414,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047873578466094054,
      "loss": 5.3778,
      "mean_token_accuracy": 0.16466535180807113,
      "num_tokens": 32213279.0,
      "step": 17465
    },
    {
      "entropy": 5.642923879623413,
      "epoch": 1.467716866204579,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004787230746630746,
      "loss": 5.3774,
      "mean_token_accuracy": 0.16785975694656372,
      "num_tokens": 32221668.0,
      "step": 17470
    },
    {
      "entropy": 5.698230314254761,
      "epoch": 1.4681369460197438,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004787103610563593,
      "loss": 5.3143,
      "mean_token_accuracy": 0.174575175344944,
      "num_tokens": 32229683.0,
      "step": 17475
    },
    {
      "entropy": 5.638741779327392,
      "epoch": 1.4685570258349085,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047869764384101993,
      "loss": 5.3827,
      "mean_token_accuracy": 0.17056026607751845,
      "num_tokens": 32238948.0,
      "step": 17480
    },
    {
      "entropy": 5.646110200881958,
      "epoch": 1.4689771056500736,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047868492301728164,
      "loss": 5.4054,
      "mean_token_accuracy": 0.16506677567958833,
      "num_tokens": 32248079.0,
      "step": 17485
    },
    {
      "entropy": 5.6019964694976805,
      "epoch": 1.4693971854652383,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047867219858536975,
      "loss": 5.2652,
      "mean_token_accuracy": 0.17940953373908997,
      "num_tokens": 32256413.0,
      "step": 17490
    },
    {
      "entropy": 5.627816963195801,
      "epoch": 1.4698172652804034,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004786594705455098,
      "loss": 5.4085,
      "mean_token_accuracy": 0.1593916341662407,
      "num_tokens": 32265954.0,
      "step": 17495
    },
    {
      "entropy": 5.602740812301636,
      "epoch": 1.4702373450955681,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004786467388979272,
      "loss": 5.3315,
      "mean_token_accuracy": 0.1727946951985359,
      "num_tokens": 32273817.0,
      "step": 17500
    },
    {
      "entropy": 5.579733514785767,
      "epoch": 1.470657424910733,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047863400364284744,
      "loss": 5.3736,
      "mean_token_accuracy": 0.17278320342302322,
      "num_tokens": 32283025.0,
      "step": 17505
    },
    {
      "entropy": 5.666515493392945,
      "epoch": 1.471077504725898,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047862126478049623,
      "loss": 5.359,
      "mean_token_accuracy": 0.16888631582260133,
      "num_tokens": 32292321.0,
      "step": 17510
    },
    {
      "entropy": 5.761226415634155,
      "epoch": 1.4714975845410627,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047860852231109915,
      "loss": 5.4662,
      "mean_token_accuracy": 0.15754509419202806,
      "num_tokens": 32302203.0,
      "step": 17515
    },
    {
      "entropy": 5.576655435562134,
      "epoch": 1.4719176643562277,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004785957762348819,
      "loss": 5.2889,
      "mean_token_accuracy": 0.17498506158590316,
      "num_tokens": 32310893.0,
      "step": 17520
    },
    {
      "entropy": 5.576455879211426,
      "epoch": 1.4723377441713925,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004785830265520703,
      "loss": 5.3622,
      "mean_token_accuracy": 0.1675017073750496,
      "num_tokens": 32320320.0,
      "step": 17525
    },
    {
      "entropy": 5.55831127166748,
      "epoch": 1.4727578239865575,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047857027326289023,
      "loss": 5.2658,
      "mean_token_accuracy": 0.1819165274500847,
      "num_tokens": 32329196.0,
      "step": 17530
    },
    {
      "entropy": 5.643089294433594,
      "epoch": 1.4731779038017223,
      "grad_norm": 1.90625,
      "learning_rate": 0.00047855751636756763,
      "loss": 5.3915,
      "mean_token_accuracy": 0.16785504668951035,
      "num_tokens": 32338529.0,
      "step": 17535
    },
    {
      "entropy": 5.68659725189209,
      "epoch": 1.4735979836168873,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004785447558663284,
      "loss": 5.3826,
      "mean_token_accuracy": 0.17284310460090638,
      "num_tokens": 32347114.0,
      "step": 17540
    },
    {
      "entropy": 5.692071056365966,
      "epoch": 1.474018063432052,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047853199175939865,
      "loss": 5.5471,
      "mean_token_accuracy": 0.1611107259988785,
      "num_tokens": 32356765.0,
      "step": 17545
    },
    {
      "entropy": 5.759571981430054,
      "epoch": 1.474438143247217,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004785192240470045,
      "loss": 5.5072,
      "mean_token_accuracy": 0.1675972655415535,
      "num_tokens": 32366175.0,
      "step": 17550
    },
    {
      "entropy": 5.627310848236084,
      "epoch": 1.474858223062382,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000478506452729372,
      "loss": 5.2843,
      "mean_token_accuracy": 0.17344675362110137,
      "num_tokens": 32375063.0,
      "step": 17555
    },
    {
      "entropy": 5.650757598876953,
      "epoch": 1.4752783028775467,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047849367780672755,
      "loss": 5.3975,
      "mean_token_accuracy": 0.16767994165420533,
      "num_tokens": 32384596.0,
      "step": 17560
    },
    {
      "entropy": 5.557038164138794,
      "epoch": 1.4756983826927117,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004784808992792974,
      "loss": 5.3405,
      "mean_token_accuracy": 0.1690414607524872,
      "num_tokens": 32393489.0,
      "step": 17565
    },
    {
      "entropy": 5.655299949645996,
      "epoch": 1.4761184625078765,
      "grad_norm": 1.375,
      "learning_rate": 0.0004784681171473079,
      "loss": 5.3329,
      "mean_token_accuracy": 0.1704338937997818,
      "num_tokens": 32402192.0,
      "step": 17570
    },
    {
      "entropy": 5.729859638214111,
      "epoch": 1.4765385423230413,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047845533141098543,
      "loss": 5.4221,
      "mean_token_accuracy": 0.16427381634712218,
      "num_tokens": 32411317.0,
      "step": 17575
    },
    {
      "entropy": 5.717471647262573,
      "epoch": 1.4769586221382063,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004784425420705565,
      "loss": 5.4609,
      "mean_token_accuracy": 0.1668254777789116,
      "num_tokens": 32420308.0,
      "step": 17580
    },
    {
      "entropy": 5.572137641906738,
      "epoch": 1.477378701953371,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004784297491262477,
      "loss": 5.4004,
      "mean_token_accuracy": 0.1696450561285019,
      "num_tokens": 32429532.0,
      "step": 17585
    },
    {
      "entropy": 5.634917783737182,
      "epoch": 1.477798781768536,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004784169525782858,
      "loss": 5.3775,
      "mean_token_accuracy": 0.1638658955693245,
      "num_tokens": 32439382.0,
      "step": 17590
    },
    {
      "entropy": 5.697785520553589,
      "epoch": 1.4782188615837009,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004784041524268971,
      "loss": 5.3715,
      "mean_token_accuracy": 0.178119695186615,
      "num_tokens": 32447893.0,
      "step": 17595
    },
    {
      "entropy": 5.63444676399231,
      "epoch": 1.4786389413988656,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047839134867230874,
      "loss": 5.3719,
      "mean_token_accuracy": 0.16978124529123306,
      "num_tokens": 32457770.0,
      "step": 17600
    },
    {
      "entropy": 5.659803771972657,
      "epoch": 1.4790590212140307,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047837854131474726,
      "loss": 5.4804,
      "mean_token_accuracy": 0.17077279835939407,
      "num_tokens": 32467247.0,
      "step": 17605
    },
    {
      "entropy": 5.714891862869263,
      "epoch": 1.4794791010291957,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047836573035443976,
      "loss": 5.4552,
      "mean_token_accuracy": 0.16792477369308473,
      "num_tokens": 32477453.0,
      "step": 17610
    },
    {
      "entropy": 5.707805252075195,
      "epoch": 1.4798991808443605,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047835291579161293,
      "loss": 5.4173,
      "mean_token_accuracy": 0.1741449236869812,
      "num_tokens": 32486278.0,
      "step": 17615
    },
    {
      "entropy": 5.6040425300598145,
      "epoch": 1.4803192606595252,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004783400976264941,
      "loss": 5.3651,
      "mean_token_accuracy": 0.17205783426761628,
      "num_tokens": 32495523.0,
      "step": 17620
    },
    {
      "entropy": 5.662339067459106,
      "epoch": 1.4807393404746902,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047832727585930997,
      "loss": 5.3948,
      "mean_token_accuracy": 0.16870251148939133,
      "num_tokens": 32504952.0,
      "step": 17625
    },
    {
      "entropy": 5.629589462280274,
      "epoch": 1.481159420289855,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004783144504902879,
      "loss": 5.3797,
      "mean_token_accuracy": 0.16395614445209503,
      "num_tokens": 32515620.0,
      "step": 17630
    },
    {
      "entropy": 5.592870140075684,
      "epoch": 1.48157950010502,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000478301621519655,
      "loss": 5.3457,
      "mean_token_accuracy": 0.17192930728197098,
      "num_tokens": 32524549.0,
      "step": 17635
    },
    {
      "entropy": 5.675258922576904,
      "epoch": 1.4819995799201848,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004782887889476386,
      "loss": 5.2369,
      "mean_token_accuracy": 0.1833137184381485,
      "num_tokens": 32533043.0,
      "step": 17640
    },
    {
      "entropy": 5.631205272674561,
      "epoch": 1.4824196597353496,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000478275952774466,
      "loss": 5.3423,
      "mean_token_accuracy": 0.1733717292547226,
      "num_tokens": 32541679.0,
      "step": 17645
    },
    {
      "entropy": 5.625184726715088,
      "epoch": 1.4828397395505146,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004782631130003646,
      "loss": 5.4748,
      "mean_token_accuracy": 0.17073367685079574,
      "num_tokens": 32550922.0,
      "step": 17650
    },
    {
      "entropy": 5.683590936660766,
      "epoch": 1.4832598193656794,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004782502696255617,
      "loss": 5.4669,
      "mean_token_accuracy": 0.16418150514364244,
      "num_tokens": 32560063.0,
      "step": 17655
    },
    {
      "entropy": 5.602748537063599,
      "epoch": 1.4836798991808444,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047823742265028495,
      "loss": 5.3326,
      "mean_token_accuracy": 0.17246636748313904,
      "num_tokens": 32569476.0,
      "step": 17660
    },
    {
      "entropy": 5.674150800704956,
      "epoch": 1.4840999789960092,
      "grad_norm": 1.578125,
      "learning_rate": 0.000478224572074762,
      "loss": 5.3998,
      "mean_token_accuracy": 0.17552062273025512,
      "num_tokens": 32578552.0,
      "step": 17665
    },
    {
      "entropy": 5.681234836578369,
      "epoch": 1.484520058811174,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004782117178992203,
      "loss": 5.4021,
      "mean_token_accuracy": 0.1674213707447052,
      "num_tokens": 32589074.0,
      "step": 17670
    },
    {
      "entropy": 5.639297246932983,
      "epoch": 1.484940138626339,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004781988601238878,
      "loss": 5.417,
      "mean_token_accuracy": 0.17062274515628814,
      "num_tokens": 32599288.0,
      "step": 17675
    },
    {
      "entropy": 5.759956312179566,
      "epoch": 1.485360218441504,
      "grad_norm": 1.9921875,
      "learning_rate": 0.000478185998748992,
      "loss": 5.4804,
      "mean_token_accuracy": 0.15948504209518433,
      "num_tokens": 32609430.0,
      "step": 17680
    },
    {
      "entropy": 5.624119091033935,
      "epoch": 1.4857802982566688,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047817313377476083,
      "loss": 5.3206,
      "mean_token_accuracy": 0.17070451378822327,
      "num_tokens": 32617763.0,
      "step": 17685
    },
    {
      "entropy": 5.566957235336304,
      "epoch": 1.4862003780718336,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047816026520142234,
      "loss": 5.4057,
      "mean_token_accuracy": 0.16364279687404631,
      "num_tokens": 32627465.0,
      "step": 17690
    },
    {
      "entropy": 5.725888967514038,
      "epoch": 1.4866204578869986,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004781473930292043,
      "loss": 5.2953,
      "mean_token_accuracy": 0.17939362078905105,
      "num_tokens": 32635984.0,
      "step": 17695
    },
    {
      "entropy": 5.580030059814453,
      "epoch": 1.4870405377021634,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004781345172583348,
      "loss": 5.2499,
      "mean_token_accuracy": 0.17612322568893432,
      "num_tokens": 32644346.0,
      "step": 17700
    },
    {
      "entropy": 5.565918779373169,
      "epoch": 1.4874606175173284,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047812163788904196,
      "loss": 5.3925,
      "mean_token_accuracy": 0.16759877502918244,
      "num_tokens": 32654118.0,
      "step": 17705
    },
    {
      "entropy": 5.685973119735718,
      "epoch": 1.4878806973324932,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047810875492155386,
      "loss": 5.4155,
      "mean_token_accuracy": 0.17071556597948073,
      "num_tokens": 32664258.0,
      "step": 17710
    },
    {
      "entropy": 5.656961488723755,
      "epoch": 1.488300777147658,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004780958683560987,
      "loss": 5.4621,
      "mean_token_accuracy": 0.16185254529118537,
      "num_tokens": 32673672.0,
      "step": 17715
    },
    {
      "entropy": 5.653824186325073,
      "epoch": 1.488720856962823,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004780829781929049,
      "loss": 5.412,
      "mean_token_accuracy": 0.16046522855758666,
      "num_tokens": 32682901.0,
      "step": 17720
    },
    {
      "entropy": 5.707868957519532,
      "epoch": 1.4891409367779878,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004780700844322007,
      "loss": 5.3688,
      "mean_token_accuracy": 0.17320224791765212,
      "num_tokens": 32691384.0,
      "step": 17725
    },
    {
      "entropy": 5.608836221694946,
      "epoch": 1.4895610165931528,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047805718707421446,
      "loss": 5.4056,
      "mean_token_accuracy": 0.16906026899814605,
      "num_tokens": 32700758.0,
      "step": 17730
    },
    {
      "entropy": 5.6796036720275875,
      "epoch": 1.4899810964083176,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047804428611917475,
      "loss": 5.5089,
      "mean_token_accuracy": 0.1667666420340538,
      "num_tokens": 32709676.0,
      "step": 17735
    },
    {
      "entropy": 5.722919750213623,
      "epoch": 1.4904011762234823,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047803138156731,
      "loss": 5.4146,
      "mean_token_accuracy": 0.17104718834161758,
      "num_tokens": 32718102.0,
      "step": 17740
    },
    {
      "entropy": 5.714565277099609,
      "epoch": 1.4908212560386473,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047801847341884897,
      "loss": 5.3923,
      "mean_token_accuracy": 0.17259609401226045,
      "num_tokens": 32727356.0,
      "step": 17745
    },
    {
      "entropy": 5.625113821029663,
      "epoch": 1.4912413358538124,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004780055616740202,
      "loss": 5.3845,
      "mean_token_accuracy": 0.16478749513626098,
      "num_tokens": 32736605.0,
      "step": 17750
    },
    {
      "entropy": 5.595635271072387,
      "epoch": 1.4916614156689771,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004779926463330524,
      "loss": 5.3394,
      "mean_token_accuracy": 0.16678551584482193,
      "num_tokens": 32745573.0,
      "step": 17755
    },
    {
      "entropy": 5.646907806396484,
      "epoch": 1.492081495484142,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004779797273961744,
      "loss": 5.3867,
      "mean_token_accuracy": 0.1738370679318905,
      "num_tokens": 32755695.0,
      "step": 17760
    },
    {
      "entropy": 5.642058420181274,
      "epoch": 1.492501575299307,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004779668048636151,
      "loss": 5.3149,
      "mean_token_accuracy": 0.17506082057952882,
      "num_tokens": 32763570.0,
      "step": 17765
    },
    {
      "entropy": 5.653747844696045,
      "epoch": 1.4929216551144717,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047795387873560336,
      "loss": 5.41,
      "mean_token_accuracy": 0.1683773159980774,
      "num_tokens": 32772006.0,
      "step": 17770
    },
    {
      "entropy": 5.698104476928711,
      "epoch": 1.4933417349296367,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004779409490123681,
      "loss": 5.3515,
      "mean_token_accuracy": 0.16677836030721666,
      "num_tokens": 32781080.0,
      "step": 17775
    },
    {
      "entropy": 5.627619647979737,
      "epoch": 1.4937618147448015,
      "grad_norm": 1.875,
      "learning_rate": 0.0004779280156941384,
      "loss": 5.3282,
      "mean_token_accuracy": 0.17010657787322997,
      "num_tokens": 32789880.0,
      "step": 17780
    },
    {
      "entropy": 5.694786787033081,
      "epoch": 1.4941818945599663,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047791507878114354,
      "loss": 5.3724,
      "mean_token_accuracy": 0.17173658013343812,
      "num_tokens": 32799222.0,
      "step": 17785
    },
    {
      "entropy": 5.607821369171143,
      "epoch": 1.4946019743751313,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004779021382736124,
      "loss": 5.3594,
      "mean_token_accuracy": 0.1683207020163536,
      "num_tokens": 32808945.0,
      "step": 17790
    },
    {
      "entropy": 5.570729064941406,
      "epoch": 1.495022054190296,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004778891941717745,
      "loss": 5.2748,
      "mean_token_accuracy": 0.18244078308343886,
      "num_tokens": 32818386.0,
      "step": 17795
    },
    {
      "entropy": 5.5826162815094,
      "epoch": 1.495442134005461,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004778762464758589,
      "loss": 5.358,
      "mean_token_accuracy": 0.1659644290804863,
      "num_tokens": 32828364.0,
      "step": 17800
    },
    {
      "entropy": 5.7570044040679935,
      "epoch": 1.495862213820626,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047786329518609505,
      "loss": 5.4816,
      "mean_token_accuracy": 0.1639697015285492,
      "num_tokens": 32837399.0,
      "step": 17805
    },
    {
      "entropy": 5.6532985210418705,
      "epoch": 1.4962822936357907,
      "grad_norm": 1.875,
      "learning_rate": 0.00047785034030271243,
      "loss": 5.3086,
      "mean_token_accuracy": 0.17564183622598648,
      "num_tokens": 32846111.0,
      "step": 17810
    },
    {
      "entropy": 5.629184007644653,
      "epoch": 1.4967023734509557,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004778373818259404,
      "loss": 5.2166,
      "mean_token_accuracy": 0.1854969322681427,
      "num_tokens": 32855839.0,
      "step": 17815
    },
    {
      "entropy": 5.668022871017456,
      "epoch": 1.4971224532661207,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047782441975600866,
      "loss": 5.5116,
      "mean_token_accuracy": 0.16896526962518693,
      "num_tokens": 32865946.0,
      "step": 17820
    },
    {
      "entropy": 5.65639214515686,
      "epoch": 1.4975425330812855,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004778114540931468,
      "loss": 5.486,
      "mean_token_accuracy": 0.1631571039557457,
      "num_tokens": 32875310.0,
      "step": 17825
    },
    {
      "entropy": 5.665874910354614,
      "epoch": 1.4979626128964503,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047779848483758445,
      "loss": 5.4006,
      "mean_token_accuracy": 0.1713417261838913,
      "num_tokens": 32885315.0,
      "step": 17830
    },
    {
      "entropy": 5.653257369995117,
      "epoch": 1.4983826927116153,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047778551198955133,
      "loss": 5.3773,
      "mean_token_accuracy": 0.17557456493377685,
      "num_tokens": 32894055.0,
      "step": 17835
    },
    {
      "entropy": 5.647949886322022,
      "epoch": 1.49880277252678,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004777725355492773,
      "loss": 5.3876,
      "mean_token_accuracy": 0.17691041976213456,
      "num_tokens": 32903030.0,
      "step": 17840
    },
    {
      "entropy": 5.648596525192261,
      "epoch": 1.499222852341945,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004777595555169922,
      "loss": 5.3233,
      "mean_token_accuracy": 0.16855899542570113,
      "num_tokens": 32911562.0,
      "step": 17845
    },
    {
      "entropy": 5.697796630859375,
      "epoch": 1.4996429321571099,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000477746571892926,
      "loss": 5.4345,
      "mean_token_accuracy": 0.16200840920209886,
      "num_tokens": 32920376.0,
      "step": 17850
    },
    {
      "entropy": 5.708303642272949,
      "epoch": 1.5000630119722747,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004777335846773087,
      "loss": 5.3703,
      "mean_token_accuracy": 0.16397346258163453,
      "num_tokens": 32929374.0,
      "step": 17855
    },
    {
      "entropy": 5.5458550453186035,
      "epoch": 1.5004830917874397,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047772059387037025,
      "loss": 5.3177,
      "mean_token_accuracy": 0.16871764361858368,
      "num_tokens": 32938695.0,
      "step": 17860
    },
    {
      "entropy": 5.636627531051635,
      "epoch": 1.5009031716026044,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004777075994723409,
      "loss": 5.3895,
      "mean_token_accuracy": 0.17524078339338303,
      "num_tokens": 32947725.0,
      "step": 17865
    },
    {
      "entropy": 5.710057210922241,
      "epoch": 1.5013232514177695,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047769460148345085,
      "loss": 5.3901,
      "mean_token_accuracy": 0.16596524715423583,
      "num_tokens": 32957017.0,
      "step": 17870
    },
    {
      "entropy": 5.648824882507324,
      "epoch": 1.5017433312329342,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004776815999039303,
      "loss": 5.3555,
      "mean_token_accuracy": 0.1691620171070099,
      "num_tokens": 32965944.0,
      "step": 17875
    },
    {
      "entropy": 5.619991302490234,
      "epoch": 1.502163411048099,
      "grad_norm": 1.5,
      "learning_rate": 0.0004776685947340096,
      "loss": 5.3598,
      "mean_token_accuracy": 0.17564940303564072,
      "num_tokens": 32975368.0,
      "step": 17880
    },
    {
      "entropy": 5.644674015045166,
      "epoch": 1.502583490863264,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004776555859739191,
      "loss": 5.4302,
      "mean_token_accuracy": 0.1631500840187073,
      "num_tokens": 32984603.0,
      "step": 17885
    },
    {
      "entropy": 5.661406755447388,
      "epoch": 1.503003570678429,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047764257362388913,
      "loss": 5.3997,
      "mean_token_accuracy": 0.1676112085580826,
      "num_tokens": 32993621.0,
      "step": 17890
    },
    {
      "entropy": 5.620744848251343,
      "epoch": 1.5034236504935938,
      "grad_norm": 1.625,
      "learning_rate": 0.0004776295576841504,
      "loss": 5.3777,
      "mean_token_accuracy": 0.16989209353923798,
      "num_tokens": 33002637.0,
      "step": 17895
    },
    {
      "entropy": 5.6318518161773685,
      "epoch": 1.5038437303087586,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047761653815493337,
      "loss": 5.3219,
      "mean_token_accuracy": 0.17872236669063568,
      "num_tokens": 33011964.0,
      "step": 17900
    },
    {
      "entropy": 5.6476317882537845,
      "epoch": 1.5042638101239234,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047760351503646877,
      "loss": 5.3804,
      "mean_token_accuracy": 0.16790363639593125,
      "num_tokens": 33020626.0,
      "step": 17905
    },
    {
      "entropy": 5.655825614929199,
      "epoch": 1.5046838899390884,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004775904883289871,
      "loss": 5.3375,
      "mean_token_accuracy": 0.1692615494132042,
      "num_tokens": 33029212.0,
      "step": 17910
    },
    {
      "entropy": 5.656114673614502,
      "epoch": 1.5051039697542534,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047757745803271936,
      "loss": 5.4202,
      "mean_token_accuracy": 0.1647963285446167,
      "num_tokens": 33038893.0,
      "step": 17915
    },
    {
      "entropy": 5.6254799365997314,
      "epoch": 1.5055240495694182,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004775644241478962,
      "loss": 5.3922,
      "mean_token_accuracy": 0.16385914236307145,
      "num_tokens": 33048058.0,
      "step": 17920
    },
    {
      "entropy": 5.555276250839233,
      "epoch": 1.505944129384583,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047755138667474864,
      "loss": 5.3031,
      "mean_token_accuracy": 0.17976188957691192,
      "num_tokens": 33057106.0,
      "step": 17925
    },
    {
      "entropy": 5.631603145599366,
      "epoch": 1.506364209199748,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004775383456135075,
      "loss": 5.4527,
      "mean_token_accuracy": 0.1698027804493904,
      "num_tokens": 33066400.0,
      "step": 17930
    },
    {
      "entropy": 5.660645914077759,
      "epoch": 1.5067842890149128,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004775253009644038,
      "loss": 5.3018,
      "mean_token_accuracy": 0.17767325341701506,
      "num_tokens": 33075357.0,
      "step": 17935
    },
    {
      "entropy": 5.714733505249024,
      "epoch": 1.5072043688300778,
      "grad_norm": 1.375,
      "learning_rate": 0.00047751225272766885,
      "loss": 5.4089,
      "mean_token_accuracy": 0.16591506600379943,
      "num_tokens": 33085707.0,
      "step": 17940
    },
    {
      "entropy": 5.770922994613647,
      "epoch": 1.5076244486452426,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004774992009035335,
      "loss": 5.5196,
      "mean_token_accuracy": 0.16405797749757767,
      "num_tokens": 33095825.0,
      "step": 17945
    },
    {
      "entropy": 5.612726068496704,
      "epoch": 1.5080445284604074,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004774861454922291,
      "loss": 5.318,
      "mean_token_accuracy": 0.17482093423604966,
      "num_tokens": 33105130.0,
      "step": 17950
    },
    {
      "entropy": 5.598492860794067,
      "epoch": 1.5084646082755724,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004774730864939869,
      "loss": 5.3389,
      "mean_token_accuracy": 0.17147504091262816,
      "num_tokens": 33113226.0,
      "step": 17955
    },
    {
      "entropy": 5.667261028289795,
      "epoch": 1.5088846880907374,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047746002390903824,
      "loss": 5.3534,
      "mean_token_accuracy": 0.17695703953504563,
      "num_tokens": 33120824.0,
      "step": 17960
    },
    {
      "entropy": 5.680776929855346,
      "epoch": 1.5093047679059022,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004774469577376145,
      "loss": 5.3328,
      "mean_token_accuracy": 0.1755007192492485,
      "num_tokens": 33129503.0,
      "step": 17965
    },
    {
      "entropy": 5.515224266052246,
      "epoch": 1.509724847721067,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047743388797994715,
      "loss": 5.2584,
      "mean_token_accuracy": 0.17618102580308914,
      "num_tokens": 33138838.0,
      "step": 17970
    },
    {
      "entropy": 5.607498073577881,
      "epoch": 1.5101449275362318,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047742081463626767,
      "loss": 5.3641,
      "mean_token_accuracy": 0.17117352187633514,
      "num_tokens": 33148142.0,
      "step": 17975
    },
    {
      "entropy": 5.659677505493164,
      "epoch": 1.5105650073513968,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004774077377068078,
      "loss": 5.3632,
      "mean_token_accuracy": 0.17830976098775864,
      "num_tokens": 33156750.0,
      "step": 17980
    },
    {
      "entropy": 5.732007122039795,
      "epoch": 1.5109850871665618,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000477394657191799,
      "loss": 5.5028,
      "mean_token_accuracy": 0.16068897396326065,
      "num_tokens": 33166511.0,
      "step": 17985
    },
    {
      "entropy": 5.734124994277954,
      "epoch": 1.5114051669817266,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047738157309147307,
      "loss": 5.4564,
      "mean_token_accuracy": 0.17238975465297698,
      "num_tokens": 33175812.0,
      "step": 17990
    },
    {
      "entropy": 5.629014778137207,
      "epoch": 1.5118252467968913,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047736848540606174,
      "loss": 5.312,
      "mean_token_accuracy": 0.16764700263738633,
      "num_tokens": 33185201.0,
      "step": 17995
    },
    {
      "entropy": 5.629927110671997,
      "epoch": 1.5122453266120561,
      "grad_norm": 1.484375,
      "learning_rate": 0.000477355394135797,
      "loss": 5.3136,
      "mean_token_accuracy": 0.17039200961589812,
      "num_tokens": 33195151.0,
      "step": 18000
    },
    {
      "epoch": 1.5122453266120561,
      "eval_entropy": 5.459788374511875,
      "eval_loss": 5.455646514892578,
      "eval_mean_token_accuracy": 0.1750611171024281,
      "eval_num_tokens": 33195151.0,
      "eval_runtime": 27.0726,
      "eval_samples_per_second": 1380.213,
      "eval_steps_per_second": 172.536,
      "step": 18000
    },
    {
      "entropy": 5.674426651000976,
      "epoch": 1.5126654064272211,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004773422992809106,
      "loss": 5.3449,
      "mean_token_accuracy": 0.17582328766584396,
      "num_tokens": 33204800.0,
      "step": 18005
    },
    {
      "entropy": 5.641755676269531,
      "epoch": 1.5130854862423861,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004773292008416346,
      "loss": 5.4032,
      "mean_token_accuracy": 0.1712186098098755,
      "num_tokens": 33214529.0,
      "step": 18010
    },
    {
      "entropy": 5.686544275283813,
      "epoch": 1.513505566057551,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047731609881820095,
      "loss": 5.4118,
      "mean_token_accuracy": 0.16349627077579498,
      "num_tokens": 33224522.0,
      "step": 18015
    },
    {
      "entropy": 5.73136887550354,
      "epoch": 1.5139256458727157,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047730299321084173,
      "loss": 5.3984,
      "mean_token_accuracy": 0.17047033011913298,
      "num_tokens": 33233220.0,
      "step": 18020
    },
    {
      "entropy": 5.655951595306396,
      "epoch": 1.5143457256878807,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047728988401978916,
      "loss": 5.3258,
      "mean_token_accuracy": 0.1736277535557747,
      "num_tokens": 33242277.0,
      "step": 18025
    },
    {
      "entropy": 5.648154401779175,
      "epoch": 1.5147658055030457,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004772767712452756,
      "loss": 5.3858,
      "mean_token_accuracy": 0.1800223171710968,
      "num_tokens": 33251113.0,
      "step": 18030
    },
    {
      "entropy": 5.632388877868652,
      "epoch": 1.5151858853182105,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047726365488753305,
      "loss": 5.5291,
      "mean_token_accuracy": 0.16324126422405244,
      "num_tokens": 33261055.0,
      "step": 18035
    },
    {
      "entropy": 5.738013982772827,
      "epoch": 1.5156059651333753,
      "grad_norm": 1.625,
      "learning_rate": 0.00047725053494679403,
      "loss": 5.4864,
      "mean_token_accuracy": 0.16342642456293105,
      "num_tokens": 33270981.0,
      "step": 18040
    },
    {
      "entropy": 5.768500185012817,
      "epoch": 1.51602604494854,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047723741142329104,
      "loss": 5.4397,
      "mean_token_accuracy": 0.16923352479934692,
      "num_tokens": 33279516.0,
      "step": 18045
    },
    {
      "entropy": 5.5703369140625,
      "epoch": 1.516446124763705,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047722428431725637,
      "loss": 5.3518,
      "mean_token_accuracy": 0.17937315553426741,
      "num_tokens": 33288300.0,
      "step": 18050
    },
    {
      "entropy": 5.634455728530884,
      "epoch": 1.5168662045788701,
      "grad_norm": 1.75,
      "learning_rate": 0.0004772111536289226,
      "loss": 5.401,
      "mean_token_accuracy": 0.16815452575683593,
      "num_tokens": 33299059.0,
      "step": 18055
    },
    {
      "entropy": 5.678378868103027,
      "epoch": 1.517286284394035,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047719801935852235,
      "loss": 5.4347,
      "mean_token_accuracy": 0.16574823260307311,
      "num_tokens": 33308879.0,
      "step": 18060
    },
    {
      "entropy": 5.758377027511597,
      "epoch": 1.5177063642091997,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004771848815062883,
      "loss": 5.5428,
      "mean_token_accuracy": 0.16148000359535217,
      "num_tokens": 33318615.0,
      "step": 18065
    },
    {
      "entropy": 5.742393970489502,
      "epoch": 1.5181264440243645,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004771717400724532,
      "loss": 5.5484,
      "mean_token_accuracy": 0.16010773554444313,
      "num_tokens": 33328748.0,
      "step": 18070
    },
    {
      "entropy": 5.733602523803711,
      "epoch": 1.5185465238395295,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004771585950572499,
      "loss": 5.365,
      "mean_token_accuracy": 0.16677297055721282,
      "num_tokens": 33338350.0,
      "step": 18075
    },
    {
      "entropy": 5.652308416366577,
      "epoch": 1.5189666036546945,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004771454464609111,
      "loss": 5.3776,
      "mean_token_accuracy": 0.16500310599803925,
      "num_tokens": 33348202.0,
      "step": 18080
    },
    {
      "entropy": 5.582103490829468,
      "epoch": 1.5193866834698593,
      "grad_norm": 2.0,
      "learning_rate": 0.0004771322942836699,
      "loss": 5.379,
      "mean_token_accuracy": 0.16805653274059296,
      "num_tokens": 33356996.0,
      "step": 18085
    },
    {
      "entropy": 5.797093152999878,
      "epoch": 1.519806763285024,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004771191385257592,
      "loss": 5.5094,
      "mean_token_accuracy": 0.15904798954725266,
      "num_tokens": 33366173.0,
      "step": 18090
    },
    {
      "entropy": 5.70246376991272,
      "epoch": 1.520226843100189,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004771059791874119,
      "loss": 5.4095,
      "mean_token_accuracy": 0.16094013452529907,
      "num_tokens": 33375921.0,
      "step": 18095
    },
    {
      "entropy": 5.551356458663941,
      "epoch": 1.520646922915354,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004770928162688613,
      "loss": 5.3922,
      "mean_token_accuracy": 0.17148933112621306,
      "num_tokens": 33385538.0,
      "step": 18100
    },
    {
      "entropy": 5.603475856781006,
      "epoch": 1.5210670027305189,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047707964977034055,
      "loss": 5.293,
      "mean_token_accuracy": 0.1830568492412567,
      "num_tokens": 33393728.0,
      "step": 18105
    },
    {
      "entropy": 5.759542036056518,
      "epoch": 1.5214870825456837,
      "grad_norm": 2.375,
      "learning_rate": 0.0004770664796920828,
      "loss": 5.4061,
      "mean_token_accuracy": 0.16567544937133788,
      "num_tokens": 33402540.0,
      "step": 18110
    },
    {
      "entropy": 5.624875116348266,
      "epoch": 1.5219071623608484,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004770533060343215,
      "loss": 5.3697,
      "mean_token_accuracy": 0.17069965749979019,
      "num_tokens": 33411706.0,
      "step": 18115
    },
    {
      "entropy": 5.566268110275269,
      "epoch": 1.5223272421760135,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004770401287972899,
      "loss": 5.3314,
      "mean_token_accuracy": 0.1724267154932022,
      "num_tokens": 33420604.0,
      "step": 18120
    },
    {
      "entropy": 5.570205879211426,
      "epoch": 1.5227473219911785,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047702694798122143,
      "loss": 5.2871,
      "mean_token_accuracy": 0.18330391943454744,
      "num_tokens": 33429558.0,
      "step": 18125
    },
    {
      "entropy": 5.7841990947723385,
      "epoch": 1.5231674018063432,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047701376358634957,
      "loss": 5.5114,
      "mean_token_accuracy": 0.1609061524271965,
      "num_tokens": 33439620.0,
      "step": 18130
    },
    {
      "entropy": 5.718216800689698,
      "epoch": 1.523587481621508,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047700057561290797,
      "loss": 5.4643,
      "mean_token_accuracy": 0.16582092493772507,
      "num_tokens": 33449067.0,
      "step": 18135
    },
    {
      "entropy": 5.585053205490112,
      "epoch": 1.5240075614366728,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004769873840611302,
      "loss": 5.3521,
      "mean_token_accuracy": 0.17460926324129106,
      "num_tokens": 33458089.0,
      "step": 18140
    },
    {
      "entropy": 5.645829582214356,
      "epoch": 1.5244276412518378,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004769741889312499,
      "loss": 5.4824,
      "mean_token_accuracy": 0.16593504399061204,
      "num_tokens": 33466883.0,
      "step": 18145
    },
    {
      "entropy": 5.703077697753907,
      "epoch": 1.5248477210670028,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047696099022350087,
      "loss": 5.4897,
      "mean_token_accuracy": 0.1618822008371353,
      "num_tokens": 33476649.0,
      "step": 18150
    },
    {
      "entropy": 5.781108331680298,
      "epoch": 1.5252678008821676,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047694778793811685,
      "loss": 5.4763,
      "mean_token_accuracy": 0.16275810748338698,
      "num_tokens": 33486274.0,
      "step": 18155
    },
    {
      "entropy": 5.6584916591644285,
      "epoch": 1.5256878806973324,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047693458207533177,
      "loss": 5.354,
      "mean_token_accuracy": 0.17072590291500092,
      "num_tokens": 33494950.0,
      "step": 18160
    },
    {
      "entropy": 5.595110654830933,
      "epoch": 1.5261079605124974,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004769213726353795,
      "loss": 5.3853,
      "mean_token_accuracy": 0.1706066131591797,
      "num_tokens": 33503545.0,
      "step": 18165
    },
    {
      "entropy": 5.643012619018554,
      "epoch": 1.5265280403276622,
      "grad_norm": 1.75,
      "learning_rate": 0.00047690815961849416,
      "loss": 5.4278,
      "mean_token_accuracy": 0.16880149990320206,
      "num_tokens": 33512871.0,
      "step": 18170
    },
    {
      "entropy": 5.648615884780884,
      "epoch": 1.5269481201428272,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004768949430249097,
      "loss": 5.344,
      "mean_token_accuracy": 0.1674499899148941,
      "num_tokens": 33521933.0,
      "step": 18175
    },
    {
      "entropy": 5.603993940353393,
      "epoch": 1.527368199957992,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004768817228548603,
      "loss": 5.3417,
      "mean_token_accuracy": 0.1691083163022995,
      "num_tokens": 33531370.0,
      "step": 18180
    },
    {
      "entropy": 5.7255950450897215,
      "epoch": 1.5277882797731568,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004768684991085802,
      "loss": 5.409,
      "mean_token_accuracy": 0.16762564182281495,
      "num_tokens": 33540310.0,
      "step": 18185
    },
    {
      "entropy": 5.693144083023071,
      "epoch": 1.5282083595883218,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047685527178630347,
      "loss": 5.4399,
      "mean_token_accuracy": 0.1699402615427971,
      "num_tokens": 33549943.0,
      "step": 18190
    },
    {
      "entropy": 5.709833097457886,
      "epoch": 1.5286284394034868,
      "grad_norm": 1.75,
      "learning_rate": 0.0004768420408882646,
      "loss": 5.5036,
      "mean_token_accuracy": 0.1646678313612938,
      "num_tokens": 33560167.0,
      "step": 18195
    },
    {
      "entropy": 5.747394323348999,
      "epoch": 1.5290485192186516,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047682880641469787,
      "loss": 5.3777,
      "mean_token_accuracy": 0.16801611781120301,
      "num_tokens": 33569604.0,
      "step": 18200
    },
    {
      "entropy": 5.669314241409301,
      "epoch": 1.5294685990338164,
      "grad_norm": 1.5,
      "learning_rate": 0.0004768155683658378,
      "loss": 5.3642,
      "mean_token_accuracy": 0.17004830688238143,
      "num_tokens": 33578400.0,
      "step": 18205
    },
    {
      "entropy": 5.609042596817017,
      "epoch": 1.5298886788489812,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004768023267419188,
      "loss": 5.3469,
      "mean_token_accuracy": 0.1695615068078041,
      "num_tokens": 33587527.0,
      "step": 18210
    },
    {
      "entropy": 5.629545259475708,
      "epoch": 1.5303087586641462,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004767890815431756,
      "loss": 5.2727,
      "mean_token_accuracy": 0.1764537826180458,
      "num_tokens": 33596026.0,
      "step": 18215
    },
    {
      "entropy": 5.630014228820801,
      "epoch": 1.5307288384793112,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047677583276984264,
      "loss": 5.3784,
      "mean_token_accuracy": 0.17321739494800567,
      "num_tokens": 33605906.0,
      "step": 18220
    },
    {
      "entropy": 5.627338838577271,
      "epoch": 1.531148918294476,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004767625804221548,
      "loss": 5.3286,
      "mean_token_accuracy": 0.16919002383947374,
      "num_tokens": 33615758.0,
      "step": 18225
    },
    {
      "entropy": 5.625001525878906,
      "epoch": 1.5315689981096408,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004767493245003466,
      "loss": 5.4025,
      "mean_token_accuracy": 0.17981577515602112,
      "num_tokens": 33625486.0,
      "step": 18230
    },
    {
      "entropy": 5.62530198097229,
      "epoch": 1.5319890779248058,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047673606500465315,
      "loss": 5.3323,
      "mean_token_accuracy": 0.17970189601182937,
      "num_tokens": 33633954.0,
      "step": 18235
    },
    {
      "entropy": 5.6397469997406,
      "epoch": 1.5324091577399706,
      "grad_norm": 1.671875,
      "learning_rate": 0.000476722801935309,
      "loss": 5.4214,
      "mean_token_accuracy": 0.16800543367862703,
      "num_tokens": 33642478.0,
      "step": 18240
    },
    {
      "entropy": 5.604885625839233,
      "epoch": 1.5328292375551356,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004767095352925495,
      "loss": 5.348,
      "mean_token_accuracy": 0.1728253498673439,
      "num_tokens": 33650785.0,
      "step": 18245
    },
    {
      "entropy": 5.623634243011475,
      "epoch": 1.5332493173703003,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004766962650766093,
      "loss": 5.298,
      "mean_token_accuracy": 0.17767170667648316,
      "num_tokens": 33659677.0,
      "step": 18250
    },
    {
      "entropy": 5.698609399795532,
      "epoch": 1.5336693971854651,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047668299128772365,
      "loss": 5.4872,
      "mean_token_accuracy": 0.15881659388542174,
      "num_tokens": 33669493.0,
      "step": 18255
    },
    {
      "entropy": 5.724362850189209,
      "epoch": 1.5340894770006301,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004766697139261277,
      "loss": 5.4498,
      "mean_token_accuracy": 0.16795236021280288,
      "num_tokens": 33678446.0,
      "step": 18260
    },
    {
      "entropy": 5.6707254409790036,
      "epoch": 1.5345095568157952,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004766564329920566,
      "loss": 5.3239,
      "mean_token_accuracy": 0.18016877472400666,
      "num_tokens": 33687647.0,
      "step": 18265
    },
    {
      "entropy": 5.671639156341553,
      "epoch": 1.53492963663096,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004766431484857456,
      "loss": 5.4058,
      "mean_token_accuracy": 0.1716668576002121,
      "num_tokens": 33697395.0,
      "step": 18270
    },
    {
      "entropy": 5.6308736324310305,
      "epoch": 1.5353497164461247,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047662986040743004,
      "loss": 5.3981,
      "mean_token_accuracy": 0.17508726418018342,
      "num_tokens": 33706779.0,
      "step": 18275
    },
    {
      "entropy": 5.608095645904541,
      "epoch": 1.5357697962612895,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004766165687573454,
      "loss": 5.3683,
      "mean_token_accuracy": 0.168911375105381,
      "num_tokens": 33714828.0,
      "step": 18280
    },
    {
      "entropy": 5.676533889770508,
      "epoch": 1.5361898760764545,
      "grad_norm": 1.484375,
      "learning_rate": 0.000476603273535727,
      "loss": 5.382,
      "mean_token_accuracy": 0.1679120436310768,
      "num_tokens": 33724730.0,
      "step": 18285
    },
    {
      "entropy": 5.724509286880493,
      "epoch": 1.5366099558916195,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004765899747428104,
      "loss": 5.4585,
      "mean_token_accuracy": 0.16771758794784547,
      "num_tokens": 33734374.0,
      "step": 18290
    },
    {
      "entropy": 5.744474029541015,
      "epoch": 1.5370300357067843,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047657667237883125,
      "loss": 5.4238,
      "mean_token_accuracy": 0.1712176039814949,
      "num_tokens": 33743395.0,
      "step": 18295
    },
    {
      "entropy": 5.710663461685181,
      "epoch": 1.537450115521949,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047656336644402513,
      "loss": 5.4788,
      "mean_token_accuracy": 0.16940985769033431,
      "num_tokens": 33752526.0,
      "step": 18300
    },
    {
      "entropy": 5.695776033401489,
      "epoch": 1.5378701953371139,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004765500569386278,
      "loss": 5.4193,
      "mean_token_accuracy": 0.17909296303987504,
      "num_tokens": 33761310.0,
      "step": 18305
    },
    {
      "entropy": 5.641575622558594,
      "epoch": 1.538290275152279,
      "grad_norm": 1.515625,
      "learning_rate": 0.000476536743862875,
      "loss": 5.3412,
      "mean_token_accuracy": 0.17165313214063643,
      "num_tokens": 33770870.0,
      "step": 18310
    },
    {
      "entropy": 5.54035587310791,
      "epoch": 1.538710354967444,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047652342721700246,
      "loss": 5.2801,
      "mean_token_accuracy": 0.17212190181016923,
      "num_tokens": 33779648.0,
      "step": 18315
    },
    {
      "entropy": 5.639022302627564,
      "epoch": 1.5391304347826087,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004765101070012462,
      "loss": 5.4694,
      "mean_token_accuracy": 0.16077025532722472,
      "num_tokens": 33789172.0,
      "step": 18320
    },
    {
      "entropy": 5.7464159488677975,
      "epoch": 1.5395505145977735,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047649678321584214,
      "loss": 5.4439,
      "mean_token_accuracy": 0.1607149749994278,
      "num_tokens": 33798069.0,
      "step": 18325
    },
    {
      "entropy": 5.705628442764282,
      "epoch": 1.5399705944129385,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047648345586102643,
      "loss": 5.4069,
      "mean_token_accuracy": 0.1719604402780533,
      "num_tokens": 33806214.0,
      "step": 18330
    },
    {
      "entropy": 5.67223048210144,
      "epoch": 1.5403906742281035,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000476470124937035,
      "loss": 5.3932,
      "mean_token_accuracy": 0.17444093525409698,
      "num_tokens": 33815365.0,
      "step": 18335
    },
    {
      "entropy": 5.744498062133789,
      "epoch": 1.5408107540432683,
      "grad_norm": 1.34375,
      "learning_rate": 0.000476456790444104,
      "loss": 5.3428,
      "mean_token_accuracy": 0.1765643909573555,
      "num_tokens": 33825204.0,
      "step": 18340
    },
    {
      "entropy": 5.692826271057129,
      "epoch": 1.541230833858433,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004764434523824697,
      "loss": 5.4314,
      "mean_token_accuracy": 0.1726793810725212,
      "num_tokens": 33834439.0,
      "step": 18345
    },
    {
      "entropy": 5.5473486423492435,
      "epoch": 1.5416509136735979,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047643011075236845,
      "loss": 5.4164,
      "mean_token_accuracy": 0.16454253643751143,
      "num_tokens": 33843959.0,
      "step": 18350
    },
    {
      "entropy": 5.74797101020813,
      "epoch": 1.5420709934887629,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047641676555403646,
      "loss": 5.4772,
      "mean_token_accuracy": 0.16578357815742492,
      "num_tokens": 33853234.0,
      "step": 18355
    },
    {
      "entropy": 5.688867378234863,
      "epoch": 1.5424910733039279,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004764034167877102,
      "loss": 5.3755,
      "mean_token_accuracy": 0.16894726753234862,
      "num_tokens": 33861755.0,
      "step": 18360
    },
    {
      "entropy": 5.648456668853759,
      "epoch": 1.5429111531190927,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047639006445362607,
      "loss": 5.4635,
      "mean_token_accuracy": 0.1700041949748993,
      "num_tokens": 33870956.0,
      "step": 18365
    },
    {
      "entropy": 5.599646472930909,
      "epoch": 1.5433312329342574,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004763767085520207,
      "loss": 5.3291,
      "mean_token_accuracy": 0.16942460536956788,
      "num_tokens": 33880568.0,
      "step": 18370
    },
    {
      "entropy": 5.743152856826782,
      "epoch": 1.5437513127494222,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004763633490831306,
      "loss": 5.5217,
      "mean_token_accuracy": 0.15770011246204377,
      "num_tokens": 33890145.0,
      "step": 18375
    },
    {
      "entropy": 5.648756408691407,
      "epoch": 1.5441713925645872,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004763499860471925,
      "loss": 5.3668,
      "mean_token_accuracy": 0.1693769097328186,
      "num_tokens": 33899155.0,
      "step": 18380
    },
    {
      "entropy": 5.666848087310791,
      "epoch": 1.5445914723797523,
      "grad_norm": 2.09375,
      "learning_rate": 0.000476336619444443,
      "loss": 5.413,
      "mean_token_accuracy": 0.16453282237052919,
      "num_tokens": 33909410.0,
      "step": 18385
    },
    {
      "entropy": 5.56054277420044,
      "epoch": 1.545011552194917,
      "grad_norm": 1.515625,
      "learning_rate": 0.000476323249275119,
      "loss": 5.2953,
      "mean_token_accuracy": 0.17509517818689346,
      "num_tokens": 33918451.0,
      "step": 18390
    },
    {
      "entropy": 5.5916523933410645,
      "epoch": 1.5454316320100818,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004763098755394573,
      "loss": 5.3374,
      "mean_token_accuracy": 0.171656733751297,
      "num_tokens": 33928317.0,
      "step": 18395
    },
    {
      "entropy": 5.697586011886597,
      "epoch": 1.5458517118252468,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004762964982376949,
      "loss": 5.4917,
      "mean_token_accuracy": 0.16770461499691008,
      "num_tokens": 33938010.0,
      "step": 18400
    },
    {
      "entropy": 5.679467391967774,
      "epoch": 1.5462717916404118,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047628311737006856,
      "loss": 5.309,
      "mean_token_accuracy": 0.1681760534644127,
      "num_tokens": 33946964.0,
      "step": 18405
    },
    {
      "entropy": 5.624345302581787,
      "epoch": 1.5466918714555766,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047626973293681555,
      "loss": 5.3301,
      "mean_token_accuracy": 0.1661735489964485,
      "num_tokens": 33956026.0,
      "step": 18410
    },
    {
      "entropy": 5.6073212146759035,
      "epoch": 1.5471119512707414,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004762563449381728,
      "loss": 5.3586,
      "mean_token_accuracy": 0.1675393611192703,
      "num_tokens": 33965787.0,
      "step": 18415
    },
    {
      "entropy": 5.650176954269409,
      "epoch": 1.5475320310859062,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047624295337437753,
      "loss": 5.3941,
      "mean_token_accuracy": 0.17182754427194596,
      "num_tokens": 33974178.0,
      "step": 18420
    },
    {
      "entropy": 5.591141080856323,
      "epoch": 1.5479521109010712,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004762295582456669,
      "loss": 5.2596,
      "mean_token_accuracy": 0.17376982569694518,
      "num_tokens": 33983652.0,
      "step": 18425
    },
    {
      "entropy": 5.680289840698242,
      "epoch": 1.5483721907162362,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047621615955227835,
      "loss": 5.3368,
      "mean_token_accuracy": 0.1786057621240616,
      "num_tokens": 33991938.0,
      "step": 18430
    },
    {
      "entropy": 5.592361164093018,
      "epoch": 1.548792270531401,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004762027572944491,
      "loss": 5.3305,
      "mean_token_accuracy": 0.17347856611013412,
      "num_tokens": 33999918.0,
      "step": 18435
    },
    {
      "entropy": 5.570802354812622,
      "epoch": 1.5492123503465658,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047618935147241667,
      "loss": 5.3394,
      "mean_token_accuracy": 0.1787159413099289,
      "num_tokens": 34008416.0,
      "step": 18440
    },
    {
      "entropy": 5.68155517578125,
      "epoch": 1.5496324301617306,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004761759420864184,
      "loss": 5.431,
      "mean_token_accuracy": 0.16568011343479155,
      "num_tokens": 34017616.0,
      "step": 18445
    },
    {
      "entropy": 5.665131998062134,
      "epoch": 1.5500525099768956,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000476162529136692,
      "loss": 5.3626,
      "mean_token_accuracy": 0.17261845469474793,
      "num_tokens": 34026064.0,
      "step": 18450
    },
    {
      "entropy": 5.614729022979736,
      "epoch": 1.5504725897920606,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004761491126234749,
      "loss": 5.2637,
      "mean_token_accuracy": 0.17827146798372268,
      "num_tokens": 34035378.0,
      "step": 18455
    },
    {
      "entropy": 5.60989990234375,
      "epoch": 1.5508926696072254,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004761356925470049,
      "loss": 5.3104,
      "mean_token_accuracy": 0.16938712447881699,
      "num_tokens": 34044600.0,
      "step": 18460
    },
    {
      "entropy": 5.62862114906311,
      "epoch": 1.5513127494223902,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047612226890751956,
      "loss": 5.3891,
      "mean_token_accuracy": 0.1683188259601593,
      "num_tokens": 34054680.0,
      "step": 18465
    },
    {
      "entropy": 5.5783733367919925,
      "epoch": 1.5517328292375552,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047610884170525697,
      "loss": 5.3333,
      "mean_token_accuracy": 0.17493952959775924,
      "num_tokens": 34063034.0,
      "step": 18470
    },
    {
      "entropy": 5.623970699310303,
      "epoch": 1.55215290905272,
      "grad_norm": 1.5,
      "learning_rate": 0.0004760954109404547,
      "loss": 5.3189,
      "mean_token_accuracy": 0.1793431058526039,
      "num_tokens": 34072122.0,
      "step": 18475
    },
    {
      "entropy": 5.64604082107544,
      "epoch": 1.552572988867885,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004760819766133508,
      "loss": 5.3283,
      "mean_token_accuracy": 0.17005451023578644,
      "num_tokens": 34081493.0,
      "step": 18480
    },
    {
      "entropy": 5.659594774246216,
      "epoch": 1.5529930686830498,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047606853872418317,
      "loss": 5.4279,
      "mean_token_accuracy": 0.16368640214204788,
      "num_tokens": 34090872.0,
      "step": 18485
    },
    {
      "entropy": 5.584814262390137,
      "epoch": 1.5534131484982145,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004760550972731899,
      "loss": 5.2813,
      "mean_token_accuracy": 0.17194133400917053,
      "num_tokens": 34100729.0,
      "step": 18490
    },
    {
      "entropy": 5.536647844314575,
      "epoch": 1.5538332283133796,
      "grad_norm": 1.75,
      "learning_rate": 0.0004760416522606092,
      "loss": 5.2504,
      "mean_token_accuracy": 0.18017490804195405,
      "num_tokens": 34109492.0,
      "step": 18495
    },
    {
      "entropy": 5.523468780517578,
      "epoch": 1.5542533081285446,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004760282036866791,
      "loss": 5.3556,
      "mean_token_accuracy": 0.17621608078479767,
      "num_tokens": 34119529.0,
      "step": 18500
    },
    {
      "entropy": 5.731783056259156,
      "epoch": 1.5546733879437094,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004760147515516379,
      "loss": 5.4073,
      "mean_token_accuracy": 0.1692374214529991,
      "num_tokens": 34128261.0,
      "step": 18505
    },
    {
      "entropy": 5.6497862339019775,
      "epoch": 1.5550934677588741,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047600129585572386,
      "loss": 5.4039,
      "mean_token_accuracy": 0.16976870894432067,
      "num_tokens": 34136916.0,
      "step": 18510
    },
    {
      "entropy": 5.609410810470581,
      "epoch": 1.555513547574039,
      "grad_norm": 1.375,
      "learning_rate": 0.0004759878365991754,
      "loss": 5.3207,
      "mean_token_accuracy": 0.17621056884527206,
      "num_tokens": 34146400.0,
      "step": 18515
    },
    {
      "entropy": 5.583196067810059,
      "epoch": 1.555933627389204,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004759743737822309,
      "loss": 5.38,
      "mean_token_accuracy": 0.17592145204544068,
      "num_tokens": 34155611.0,
      "step": 18520
    },
    {
      "entropy": 5.687038278579712,
      "epoch": 1.556353707204369,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047596090740512884,
      "loss": 5.3917,
      "mean_token_accuracy": 0.17226926237344742,
      "num_tokens": 34165301.0,
      "step": 18525
    },
    {
      "entropy": 5.642127132415771,
      "epoch": 1.5567737870195337,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047594743746810786,
      "loss": 5.3733,
      "mean_token_accuracy": 0.16970995664596558,
      "num_tokens": 34174655.0,
      "step": 18530
    },
    {
      "entropy": 5.751519250869751,
      "epoch": 1.5571938668346985,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047593396397140644,
      "loss": 5.5275,
      "mean_token_accuracy": 0.1625624492764473,
      "num_tokens": 34184293.0,
      "step": 18535
    },
    {
      "entropy": 5.685022068023682,
      "epoch": 1.5576139466498635,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004759204869152632,
      "loss": 5.4077,
      "mean_token_accuracy": 0.16249970123171806,
      "num_tokens": 34193025.0,
      "step": 18540
    },
    {
      "entropy": 5.5956226825714115,
      "epoch": 1.5580340264650283,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004759070062999171,
      "loss": 5.3153,
      "mean_token_accuracy": 0.17234051376581191,
      "num_tokens": 34201082.0,
      "step": 18545
    },
    {
      "entropy": 5.693072509765625,
      "epoch": 1.5584541062801933,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004758935221256069,
      "loss": 5.4665,
      "mean_token_accuracy": 0.1661413475871086,
      "num_tokens": 34211210.0,
      "step": 18550
    },
    {
      "entropy": 5.663186168670654,
      "epoch": 1.558874186095358,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047588003439257134,
      "loss": 5.395,
      "mean_token_accuracy": 0.17036228477954865,
      "num_tokens": 34220309.0,
      "step": 18555
    },
    {
      "entropy": 5.679746532440186,
      "epoch": 1.559294265910523,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047586654310104946,
      "loss": 5.3977,
      "mean_token_accuracy": 0.16031052619218827,
      "num_tokens": 34229532.0,
      "step": 18560
    },
    {
      "entropy": 5.772851848602295,
      "epoch": 1.559714345725688,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004758530482512801,
      "loss": 5.6332,
      "mean_token_accuracy": 0.15485470741987228,
      "num_tokens": 34239543.0,
      "step": 18565
    },
    {
      "entropy": 5.688569211959839,
      "epoch": 1.560134425540853,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004758395498435024,
      "loss": 5.4115,
      "mean_token_accuracy": 0.16483961641788483,
      "num_tokens": 34248654.0,
      "step": 18570
    },
    {
      "entropy": 5.635505485534668,
      "epoch": 1.5605545053560177,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047582604787795555,
      "loss": 5.4089,
      "mean_token_accuracy": 0.16171129792928696,
      "num_tokens": 34258757.0,
      "step": 18575
    },
    {
      "entropy": 5.671541738510132,
      "epoch": 1.5609745851711825,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004758125423548787,
      "loss": 5.4048,
      "mean_token_accuracy": 0.17029386013746262,
      "num_tokens": 34268253.0,
      "step": 18580
    },
    {
      "entropy": 5.706531238555908,
      "epoch": 1.5613946649863473,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047579903327451097,
      "loss": 5.4684,
      "mean_token_accuracy": 0.16566625684499742,
      "num_tokens": 34277361.0,
      "step": 18585
    },
    {
      "entropy": 5.5921790599823,
      "epoch": 1.5618147448015123,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004757855206370919,
      "loss": 5.3349,
      "mean_token_accuracy": 0.17001636028289796,
      "num_tokens": 34285923.0,
      "step": 18590
    },
    {
      "entropy": 5.567906761169434,
      "epoch": 1.5622348246166773,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047577200444286064,
      "loss": 5.3412,
      "mean_token_accuracy": 0.167476150393486,
      "num_tokens": 34296300.0,
      "step": 18595
    },
    {
      "entropy": 5.721004152297974,
      "epoch": 1.562654904431842,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004757584846920567,
      "loss": 5.4073,
      "mean_token_accuracy": 0.16413081288337708,
      "num_tokens": 34305757.0,
      "step": 18600
    },
    {
      "entropy": 5.654394674301147,
      "epoch": 1.5630749842470069,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004757449613849196,
      "loss": 5.4196,
      "mean_token_accuracy": 0.16548021137714386,
      "num_tokens": 34314714.0,
      "step": 18605
    },
    {
      "entropy": 5.674863195419311,
      "epoch": 1.5634950640621716,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047573143452168883,
      "loss": 5.4369,
      "mean_token_accuracy": 0.17374185994267463,
      "num_tokens": 34323501.0,
      "step": 18610
    },
    {
      "entropy": 5.645120096206665,
      "epoch": 1.5639151438773367,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047571790410260405,
      "loss": 5.3779,
      "mean_token_accuracy": 0.1740238681435585,
      "num_tokens": 34331752.0,
      "step": 18615
    },
    {
      "entropy": 5.680461835861206,
      "epoch": 1.5643352236925017,
      "grad_norm": 2.46875,
      "learning_rate": 0.000475704370127905,
      "loss": 5.436,
      "mean_token_accuracy": 0.16010896265506744,
      "num_tokens": 34341479.0,
      "step": 18620
    },
    {
      "entropy": 5.651295614242554,
      "epoch": 1.5647553035076665,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004756908325978314,
      "loss": 5.4327,
      "mean_token_accuracy": 0.1683880478143692,
      "num_tokens": 34350991.0,
      "step": 18625
    },
    {
      "entropy": 5.654466438293457,
      "epoch": 1.5651753833228312,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047567729151262305,
      "loss": 5.3597,
      "mean_token_accuracy": 0.1722327247262001,
      "num_tokens": 34360089.0,
      "step": 18630
    },
    {
      "entropy": 5.592139482498169,
      "epoch": 1.5655954631379962,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004756637468725198,
      "loss": 5.3313,
      "mean_token_accuracy": 0.16946941018104553,
      "num_tokens": 34370352.0,
      "step": 18635
    },
    {
      "entropy": 5.566221141815186,
      "epoch": 1.5660155429531613,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004756501986777616,
      "loss": 5.3057,
      "mean_token_accuracy": 0.16700929552316665,
      "num_tokens": 34378958.0,
      "step": 18640
    },
    {
      "entropy": 5.583894443511963,
      "epoch": 1.566435622768326,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047563664692858843,
      "loss": 5.284,
      "mean_token_accuracy": 0.17487050145864486,
      "num_tokens": 34387723.0,
      "step": 18645
    },
    {
      "entropy": 5.690687847137451,
      "epoch": 1.5668557025834908,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004756230916252404,
      "loss": 5.4078,
      "mean_token_accuracy": 0.18035681545734406,
      "num_tokens": 34397089.0,
      "step": 18650
    },
    {
      "entropy": 5.7349998474121096,
      "epoch": 1.5672757823986556,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047560953276795756,
      "loss": 5.4243,
      "mean_token_accuracy": 0.17100331783294678,
      "num_tokens": 34406278.0,
      "step": 18655
    },
    {
      "entropy": 5.692036485671997,
      "epoch": 1.5676958622138206,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047559597035698014,
      "loss": 5.3913,
      "mean_token_accuracy": 0.16650496870279313,
      "num_tokens": 34415404.0,
      "step": 18660
    },
    {
      "entropy": 5.637453556060791,
      "epoch": 1.5681159420289856,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004755824043925485,
      "loss": 5.4261,
      "mean_token_accuracy": 0.17367287278175353,
      "num_tokens": 34425036.0,
      "step": 18665
    },
    {
      "entropy": 5.641345071792602,
      "epoch": 1.5685360218441504,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004755688348749027,
      "loss": 5.3343,
      "mean_token_accuracy": 0.17022797614336013,
      "num_tokens": 34434246.0,
      "step": 18670
    },
    {
      "entropy": 5.590941476821899,
      "epoch": 1.5689561016593152,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004755552618042834,
      "loss": 5.3347,
      "mean_token_accuracy": 0.17202802896499633,
      "num_tokens": 34444189.0,
      "step": 18675
    },
    {
      "entropy": 5.637591981887818,
      "epoch": 1.56937618147448,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004755416851809308,
      "loss": 5.3401,
      "mean_token_accuracy": 0.17225669771432878,
      "num_tokens": 34453727.0,
      "step": 18680
    },
    {
      "entropy": 5.569540357589721,
      "epoch": 1.569796261289645,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004755281050050856,
      "loss": 5.3308,
      "mean_token_accuracy": 0.174957799911499,
      "num_tokens": 34462835.0,
      "step": 18685
    },
    {
      "entropy": 5.586309862136841,
      "epoch": 1.57021634110481,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004755145212769882,
      "loss": 5.3857,
      "mean_token_accuracy": 0.1714427351951599,
      "num_tokens": 34471642.0,
      "step": 18690
    },
    {
      "entropy": 5.6509918689727785,
      "epoch": 1.5706364209199748,
      "grad_norm": 1.5,
      "learning_rate": 0.00047550093399687936,
      "loss": 5.3534,
      "mean_token_accuracy": 0.1714189037680626,
      "num_tokens": 34480468.0,
      "step": 18695
    },
    {
      "entropy": 5.7392994403839115,
      "epoch": 1.5710565007351396,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004754873431649997,
      "loss": 5.3911,
      "mean_token_accuracy": 0.16480306535959244,
      "num_tokens": 34490299.0,
      "step": 18700
    },
    {
      "entropy": 5.64626841545105,
      "epoch": 1.5714765805503046,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047547374878159003,
      "loss": 5.4117,
      "mean_token_accuracy": 0.16512584537267685,
      "num_tokens": 34498831.0,
      "step": 18705
    },
    {
      "entropy": 5.619808053970337,
      "epoch": 1.5718966603654696,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004754601508468911,
      "loss": 5.3846,
      "mean_token_accuracy": 0.17046747952699662,
      "num_tokens": 34508048.0,
      "step": 18710
    },
    {
      "entropy": 5.6205531597137455,
      "epoch": 1.5723167401806344,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004754465493611438,
      "loss": 5.469,
      "mean_token_accuracy": 0.16530200242996215,
      "num_tokens": 34517070.0,
      "step": 18715
    },
    {
      "entropy": 5.585184478759766,
      "epoch": 1.5727368199957992,
      "grad_norm": 1.5,
      "learning_rate": 0.00047543294432458904,
      "loss": 5.2721,
      "mean_token_accuracy": 0.17481959313154222,
      "num_tokens": 34525934.0,
      "step": 18720
    },
    {
      "entropy": 5.727499008178711,
      "epoch": 1.573156899810964,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000475419335737468,
      "loss": 5.4823,
      "mean_token_accuracy": 0.17422295212745667,
      "num_tokens": 34534222.0,
      "step": 18725
    },
    {
      "entropy": 5.701776647567749,
      "epoch": 1.573576979626129,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047540572360002157,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1694585531949997,
      "num_tokens": 34543291.0,
      "step": 18730
    },
    {
      "entropy": 5.697884798049927,
      "epoch": 1.573997059441294,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047539210791249095,
      "loss": 5.3251,
      "mean_token_accuracy": 0.17326029986143113,
      "num_tokens": 34552383.0,
      "step": 18735
    },
    {
      "entropy": 5.6322752952575685,
      "epoch": 1.5744171392564588,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004753784886751173,
      "loss": 5.3279,
      "mean_token_accuracy": 0.17536226212978362,
      "num_tokens": 34560311.0,
      "step": 18740
    },
    {
      "entropy": 5.608207654953003,
      "epoch": 1.5748372190716236,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004753648658881419,
      "loss": 5.3587,
      "mean_token_accuracy": 0.1810239687561989,
      "num_tokens": 34569903.0,
      "step": 18745
    },
    {
      "entropy": 5.672663879394531,
      "epoch": 1.5752572988867883,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047535123955180607,
      "loss": 5.368,
      "mean_token_accuracy": 0.16771269589662552,
      "num_tokens": 34579735.0,
      "step": 18750
    },
    {
      "entropy": 5.706464767456055,
      "epoch": 1.5756773787019533,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004753376096663512,
      "loss": 5.4061,
      "mean_token_accuracy": 0.16758844405412673,
      "num_tokens": 34589105.0,
      "step": 18755
    },
    {
      "entropy": 5.573053598403931,
      "epoch": 1.5760974585171184,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047532397623201877,
      "loss": 5.3497,
      "mean_token_accuracy": 0.17709609121084213,
      "num_tokens": 34597883.0,
      "step": 18760
    },
    {
      "entropy": 5.634762716293335,
      "epoch": 1.5765175383322831,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047531033924905024,
      "loss": 5.3166,
      "mean_token_accuracy": 0.17477548271417617,
      "num_tokens": 34606666.0,
      "step": 18765
    },
    {
      "entropy": 5.723498058319092,
      "epoch": 1.576937618147448,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004752966987176873,
      "loss": 5.4369,
      "mean_token_accuracy": 0.17316569834947587,
      "num_tokens": 34616547.0,
      "step": 18770
    },
    {
      "entropy": 5.6272937774658205,
      "epoch": 1.577357697962613,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004752830546381713,
      "loss": 5.4293,
      "mean_token_accuracy": 0.1722670242190361,
      "num_tokens": 34625679.0,
      "step": 18775
    },
    {
      "entropy": 5.585916328430176,
      "epoch": 1.5777777777777777,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004752694070107442,
      "loss": 5.357,
      "mean_token_accuracy": 0.17428914755582808,
      "num_tokens": 34635633.0,
      "step": 18780
    },
    {
      "entropy": 5.679674100875855,
      "epoch": 1.5781978575929427,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004752557558356476,
      "loss": 5.3892,
      "mean_token_accuracy": 0.17079094052314758,
      "num_tokens": 34645206.0,
      "step": 18785
    },
    {
      "entropy": 5.653341150283813,
      "epoch": 1.5786179374081075,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004752421011131234,
      "loss": 5.407,
      "mean_token_accuracy": 0.16707225739955903,
      "num_tokens": 34653884.0,
      "step": 18790
    },
    {
      "entropy": 5.589267587661743,
      "epoch": 1.5790380172232723,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047522844284341364,
      "loss": 5.2546,
      "mean_token_accuracy": 0.17837179750204085,
      "num_tokens": 34662170.0,
      "step": 18795
    },
    {
      "entropy": 5.629407215118408,
      "epoch": 1.5794580970384373,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004752147810267601,
      "loss": 5.4155,
      "mean_token_accuracy": 0.1689947798848152,
      "num_tokens": 34672548.0,
      "step": 18800
    },
    {
      "entropy": 5.712817621231079,
      "epoch": 1.5798781768536023,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047520111566340465,
      "loss": 5.4077,
      "mean_token_accuracy": 0.16957559436559677,
      "num_tokens": 34680972.0,
      "step": 18805
    },
    {
      "entropy": 5.617722415924073,
      "epoch": 1.580298256668767,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047518744675358965,
      "loss": 5.2817,
      "mean_token_accuracy": 0.1753813788294792,
      "num_tokens": 34689589.0,
      "step": 18810
    },
    {
      "entropy": 5.578068208694458,
      "epoch": 1.580718336483932,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004751737742975571,
      "loss": 5.3069,
      "mean_token_accuracy": 0.17108375132083892,
      "num_tokens": 34698747.0,
      "step": 18815
    },
    {
      "entropy": 5.618763208389282,
      "epoch": 1.5811384162990967,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047516009829554913,
      "loss": 5.3712,
      "mean_token_accuracy": 0.16741059124469757,
      "num_tokens": 34707502.0,
      "step": 18820
    },
    {
      "entropy": 5.573977041244507,
      "epoch": 1.5815584961142617,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047514641874780815,
      "loss": 5.285,
      "mean_token_accuracy": 0.17877837717533113,
      "num_tokens": 34715879.0,
      "step": 18825
    },
    {
      "entropy": 5.636180353164673,
      "epoch": 1.5819785759294267,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047513273565457644,
      "loss": 5.4762,
      "mean_token_accuracy": 0.16692524328827857,
      "num_tokens": 34726090.0,
      "step": 18830
    },
    {
      "entropy": 5.730469369888306,
      "epoch": 1.5823986557445915,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004751190490160964,
      "loss": 5.4484,
      "mean_token_accuracy": 0.16253113001585007,
      "num_tokens": 34736014.0,
      "step": 18835
    },
    {
      "entropy": 5.674402475357056,
      "epoch": 1.5828187355597563,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047510535883261035,
      "loss": 5.3765,
      "mean_token_accuracy": 0.17151375561952592,
      "num_tokens": 34745648.0,
      "step": 18840
    },
    {
      "entropy": 5.661863565444946,
      "epoch": 1.5832388153749213,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000475091665104361,
      "loss": 5.3693,
      "mean_token_accuracy": 0.1738445997238159,
      "num_tokens": 34753908.0,
      "step": 18845
    },
    {
      "entropy": 5.6560314178466795,
      "epoch": 1.583658895190086,
      "grad_norm": 1.875,
      "learning_rate": 0.0004750779678315908,
      "loss": 5.2255,
      "mean_token_accuracy": 0.18110145032405853,
      "num_tokens": 34762303.0,
      "step": 18850
    },
    {
      "entropy": 5.580573463439942,
      "epoch": 1.584078975005251,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004750642670145424,
      "loss": 5.3952,
      "mean_token_accuracy": 0.16657186895608903,
      "num_tokens": 34771463.0,
      "step": 18855
    },
    {
      "entropy": 5.725437164306641,
      "epoch": 1.5844990548204159,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004750505626534585,
      "loss": 5.4872,
      "mean_token_accuracy": 0.16719557642936705,
      "num_tokens": 34780704.0,
      "step": 18860
    },
    {
      "entropy": 5.574211645126343,
      "epoch": 1.5849191346355807,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047503685474858194,
      "loss": 5.3001,
      "mean_token_accuracy": 0.1722961261868477,
      "num_tokens": 34790262.0,
      "step": 18865
    },
    {
      "entropy": 5.660973405838012,
      "epoch": 1.5853392144507457,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004750231433001555,
      "loss": 5.3416,
      "mean_token_accuracy": 0.17414891868829727,
      "num_tokens": 34799450.0,
      "step": 18870
    },
    {
      "entropy": 5.697402715682983,
      "epoch": 1.5857592942659107,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004750094283084221,
      "loss": 5.3937,
      "mean_token_accuracy": 0.16056714206933975,
      "num_tokens": 34808220.0,
      "step": 18875
    },
    {
      "entropy": 5.722061729431152,
      "epoch": 1.5861793740810755,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047499570977362467,
      "loss": 5.404,
      "mean_token_accuracy": 0.16384544521570205,
      "num_tokens": 34817846.0,
      "step": 18880
    },
    {
      "entropy": 5.64260458946228,
      "epoch": 1.5865994538962402,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047498198769600617,
      "loss": 5.4186,
      "mean_token_accuracy": 0.1719417929649353,
      "num_tokens": 34826962.0,
      "step": 18885
    },
    {
      "entropy": 5.592233324050904,
      "epoch": 1.587019533711405,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004749682620758097,
      "loss": 5.3572,
      "mean_token_accuracy": 0.1710098758339882,
      "num_tokens": 34837170.0,
      "step": 18890
    },
    {
      "entropy": 5.613478660583496,
      "epoch": 1.58743961352657,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047495453291327854,
      "loss": 5.3568,
      "mean_token_accuracy": 0.1741234615445137,
      "num_tokens": 34845336.0,
      "step": 18895
    },
    {
      "entropy": 5.602258491516113,
      "epoch": 1.587859693341735,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047494080020865577,
      "loss": 5.3359,
      "mean_token_accuracy": 0.17452080994844438,
      "num_tokens": 34854613.0,
      "step": 18900
    },
    {
      "entropy": 5.6866683006286625,
      "epoch": 1.5882797731568998,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004749270639621846,
      "loss": 5.4126,
      "mean_token_accuracy": 0.17136956751346588,
      "num_tokens": 34864254.0,
      "step": 18905
    },
    {
      "entropy": 5.682665205001831,
      "epoch": 1.5886998529720646,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004749133241741085,
      "loss": 5.4735,
      "mean_token_accuracy": 0.16655561476945877,
      "num_tokens": 34874380.0,
      "step": 18910
    },
    {
      "entropy": 5.6765281677246096,
      "epoch": 1.5891199327872296,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004748995808446708,
      "loss": 5.3949,
      "mean_token_accuracy": 0.16749880760908126,
      "num_tokens": 34883688.0,
      "step": 18915
    },
    {
      "entropy": 5.593254899978637,
      "epoch": 1.5895400126023944,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047488583397411495,
      "loss": 5.2666,
      "mean_token_accuracy": 0.17997846901416778,
      "num_tokens": 34892831.0,
      "step": 18920
    },
    {
      "entropy": 5.602783346176148,
      "epoch": 1.5899600924175594,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047487208356268454,
      "loss": 5.3657,
      "mean_token_accuracy": 0.17542098313570023,
      "num_tokens": 34901517.0,
      "step": 18925
    },
    {
      "entropy": 5.672917652130127,
      "epoch": 1.5903801722327242,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047485832961062296,
      "loss": 5.3777,
      "mean_token_accuracy": 0.17032893896102905,
      "num_tokens": 34910765.0,
      "step": 18930
    },
    {
      "entropy": 5.723647260665894,
      "epoch": 1.590800252047889,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047484457211817405,
      "loss": 5.4147,
      "mean_token_accuracy": 0.1626910910010338,
      "num_tokens": 34919799.0,
      "step": 18935
    },
    {
      "entropy": 5.57265453338623,
      "epoch": 1.591220331863054,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047483081108558143,
      "loss": 5.2938,
      "mean_token_accuracy": 0.17613977789878846,
      "num_tokens": 34928199.0,
      "step": 18940
    },
    {
      "entropy": 5.641741323471069,
      "epoch": 1.591640411678219,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000474817046513089,
      "loss": 5.4383,
      "mean_token_accuracy": 0.16948165893554687,
      "num_tokens": 34937751.0,
      "step": 18945
    },
    {
      "entropy": 5.724054002761841,
      "epoch": 1.5920604914933838,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004748032784009403,
      "loss": 5.3584,
      "mean_token_accuracy": 0.17121144831180574,
      "num_tokens": 34946052.0,
      "step": 18950
    },
    {
      "entropy": 5.673901796340942,
      "epoch": 1.5924805713085486,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004747895067493796,
      "loss": 5.3581,
      "mean_token_accuracy": 0.1705708011984825,
      "num_tokens": 34954932.0,
      "step": 18955
    },
    {
      "entropy": 5.661267900466919,
      "epoch": 1.5929006511237134,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004747757315586505,
      "loss": 5.3471,
      "mean_token_accuracy": 0.17376241683959961,
      "num_tokens": 34963581.0,
      "step": 18960
    },
    {
      "entropy": 5.4893430233001705,
      "epoch": 1.5933207309388784,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047476195282899727,
      "loss": 5.1488,
      "mean_token_accuracy": 0.18746469616889955,
      "num_tokens": 34972844.0,
      "step": 18965
    },
    {
      "entropy": 5.622801446914673,
      "epoch": 1.5937408107540434,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047474817056066383,
      "loss": 5.3613,
      "mean_token_accuracy": 0.1791597992181778,
      "num_tokens": 34981998.0,
      "step": 18970
    },
    {
      "entropy": 5.609399318695068,
      "epoch": 1.5941608905692082,
      "grad_norm": 1.671875,
      "learning_rate": 0.00047473438475389453,
      "loss": 5.3003,
      "mean_token_accuracy": 0.17580795288085938,
      "num_tokens": 34990552.0,
      "step": 18975
    },
    {
      "entropy": 5.669076728820801,
      "epoch": 1.594580970384373,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004747205954089333,
      "loss": 5.3297,
      "mean_token_accuracy": 0.1735967993736267,
      "num_tokens": 35000259.0,
      "step": 18980
    },
    {
      "entropy": 5.687023115158081,
      "epoch": 1.5950010501995378,
      "grad_norm": 1.75,
      "learning_rate": 0.0004747068025260247,
      "loss": 5.3965,
      "mean_token_accuracy": 0.16323356330394745,
      "num_tokens": 35009592.0,
      "step": 18985
    },
    {
      "entropy": 5.691207504272461,
      "epoch": 1.5954211300147028,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004746930061054129,
      "loss": 5.4479,
      "mean_token_accuracy": 0.1601082503795624,
      "num_tokens": 35019356.0,
      "step": 18990
    },
    {
      "entropy": 5.6604314804077145,
      "epoch": 1.5958412098298678,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047467920614734224,
      "loss": 5.3466,
      "mean_token_accuracy": 0.17569188624620438,
      "num_tokens": 35028764.0,
      "step": 18995
    },
    {
      "entropy": 5.629007577896118,
      "epoch": 1.5962612896450326,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004746654026520573,
      "loss": 5.3705,
      "mean_token_accuracy": 0.1699652537703514,
      "num_tokens": 35037903.0,
      "step": 19000
    },
    {
      "entropy": 5.597415447235107,
      "epoch": 1.5966813694601973,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004746515956198026,
      "loss": 5.2896,
      "mean_token_accuracy": 0.17737337201833725,
      "num_tokens": 35046326.0,
      "step": 19005
    },
    {
      "entropy": 5.744392395019531,
      "epoch": 1.5971014492753624,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047463778505082266,
      "loss": 5.5036,
      "mean_token_accuracy": 0.16676781475543975,
      "num_tokens": 35055551.0,
      "step": 19010
    },
    {
      "entropy": 5.613027286529541,
      "epoch": 1.5975215290905274,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004746239709453621,
      "loss": 5.2786,
      "mean_token_accuracy": 0.183333557844162,
      "num_tokens": 35065595.0,
      "step": 19015
    },
    {
      "entropy": 5.679959154129028,
      "epoch": 1.5979416089056921,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004746101533036658,
      "loss": 5.3933,
      "mean_token_accuracy": 0.16878037750720978,
      "num_tokens": 35075097.0,
      "step": 19020
    },
    {
      "entropy": 5.745026302337647,
      "epoch": 1.598361688720857,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047459633212597834,
      "loss": 5.4682,
      "mean_token_accuracy": 0.16506182700395583,
      "num_tokens": 35084092.0,
      "step": 19025
    },
    {
      "entropy": 5.628228950500488,
      "epoch": 1.5987817685360217,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004745825074125447,
      "loss": 5.3593,
      "mean_token_accuracy": 0.16868028193712234,
      "num_tokens": 35093007.0,
      "step": 19030
    },
    {
      "entropy": 5.712028741836548,
      "epoch": 1.5992018483511867,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004745686791636097,
      "loss": 5.4238,
      "mean_token_accuracy": 0.16526060700416564,
      "num_tokens": 35103094.0,
      "step": 19035
    },
    {
      "entropy": 5.5950596809387205,
      "epoch": 1.5996219281663517,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047455484737941823,
      "loss": 5.2461,
      "mean_token_accuracy": 0.18176242113113403,
      "num_tokens": 35112561.0,
      "step": 19040
    },
    {
      "entropy": 5.565781784057617,
      "epoch": 1.6000420079815165,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004745410120602155,
      "loss": 5.3641,
      "mean_token_accuracy": 0.16723383218050003,
      "num_tokens": 35121718.0,
      "step": 19045
    },
    {
      "entropy": 5.631525993347168,
      "epoch": 1.6004620877966813,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047452717320624647,
      "loss": 5.3111,
      "mean_token_accuracy": 0.18230061531066893,
      "num_tokens": 35130073.0,
      "step": 19050
    },
    {
      "entropy": 5.623306941986084,
      "epoch": 1.600882167611846,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004745133308177562,
      "loss": 5.3555,
      "mean_token_accuracy": 0.16863761693239213,
      "num_tokens": 35138876.0,
      "step": 19055
    },
    {
      "entropy": 5.611622285842896,
      "epoch": 1.601302247427011,
      "grad_norm": 1.75,
      "learning_rate": 0.00047449948489499007,
      "loss": 5.3558,
      "mean_token_accuracy": 0.16690559834241867,
      "num_tokens": 35147750.0,
      "step": 19060
    },
    {
      "entropy": 5.63360447883606,
      "epoch": 1.6017223272421761,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047448563543819335,
      "loss": 5.3749,
      "mean_token_accuracy": 0.17310378402471543,
      "num_tokens": 35156955.0,
      "step": 19065
    },
    {
      "entropy": 5.617711114883423,
      "epoch": 1.602142407057341,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004744717824476112,
      "loss": 5.4058,
      "mean_token_accuracy": 0.1691771298646927,
      "num_tokens": 35166542.0,
      "step": 19070
    },
    {
      "entropy": 5.650089073181152,
      "epoch": 1.6025624868725057,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047445792592348926,
      "loss": 5.352,
      "mean_token_accuracy": 0.17186888605356215,
      "num_tokens": 35175258.0,
      "step": 19075
    },
    {
      "entropy": 5.669472885131836,
      "epoch": 1.6029825666876707,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004744440658660729,
      "loss": 5.3651,
      "mean_token_accuracy": 0.16814722567796708,
      "num_tokens": 35184970.0,
      "step": 19080
    },
    {
      "entropy": 5.640087366104126,
      "epoch": 1.6034026465028357,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004744302022756075,
      "loss": 5.3454,
      "mean_token_accuracy": 0.16680478900671006,
      "num_tokens": 35193948.0,
      "step": 19085
    },
    {
      "entropy": 5.513348531723023,
      "epoch": 1.6038227263180005,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047441633515233874,
      "loss": 5.3153,
      "mean_token_accuracy": 0.17404042780399323,
      "num_tokens": 35203792.0,
      "step": 19090
    },
    {
      "entropy": 5.611915922164917,
      "epoch": 1.6042428061331653,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004744024644965123,
      "loss": 5.4697,
      "mean_token_accuracy": 0.1673903599381447,
      "num_tokens": 35212684.0,
      "step": 19095
    },
    {
      "entropy": 5.649331140518188,
      "epoch": 1.60466288594833,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047438859030837397,
      "loss": 5.285,
      "mean_token_accuracy": 0.17957256138324737,
      "num_tokens": 35220830.0,
      "step": 19100
    },
    {
      "entropy": 5.6428052425384525,
      "epoch": 1.605082965763495,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047437471258816936,
      "loss": 5.3373,
      "mean_token_accuracy": 0.17119317799806594,
      "num_tokens": 35230171.0,
      "step": 19105
    },
    {
      "entropy": 5.585913228988647,
      "epoch": 1.60550304557866,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00047436083133614446,
      "loss": 5.2691,
      "mean_token_accuracy": 0.18000213354825972,
      "num_tokens": 35239022.0,
      "step": 19110
    },
    {
      "entropy": 5.604876136779785,
      "epoch": 1.6059231253938249,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047434694655254495,
      "loss": 5.2944,
      "mean_token_accuracy": 0.1716891884803772,
      "num_tokens": 35247564.0,
      "step": 19115
    },
    {
      "entropy": 5.604823255538941,
      "epoch": 1.6063432052089897,
      "grad_norm": 1.546875,
      "learning_rate": 0.000474333058237617,
      "loss": 5.3308,
      "mean_token_accuracy": 0.16862958818674087,
      "num_tokens": 35256175.0,
      "step": 19120
    },
    {
      "entropy": 5.737623691558838,
      "epoch": 1.6067632850241544,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047431916639160656,
      "loss": 5.4738,
      "mean_token_accuracy": 0.16881804317235946,
      "num_tokens": 35265278.0,
      "step": 19125
    },
    {
      "entropy": 5.555996465682983,
      "epoch": 1.6071833648393195,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004743052710147598,
      "loss": 5.2156,
      "mean_token_accuracy": 0.1811077132821083,
      "num_tokens": 35274715.0,
      "step": 19130
    },
    {
      "entropy": 5.501038646697998,
      "epoch": 1.6076034446544845,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047429137210732266,
      "loss": 5.3167,
      "mean_token_accuracy": 0.17403266429901124,
      "num_tokens": 35285450.0,
      "step": 19135
    },
    {
      "entropy": 5.629172325134277,
      "epoch": 1.6080235244696492,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004742774696695415,
      "loss": 5.3447,
      "mean_token_accuracy": 0.16400485038757323,
      "num_tokens": 35294531.0,
      "step": 19140
    },
    {
      "entropy": 5.699636173248291,
      "epoch": 1.608443604284814,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047426356370166266,
      "loss": 5.3749,
      "mean_token_accuracy": 0.16152792572975158,
      "num_tokens": 35303749.0,
      "step": 19145
    },
    {
      "entropy": 5.576769399642944,
      "epoch": 1.608863684099979,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004742496542039324,
      "loss": 5.3381,
      "mean_token_accuracy": 0.16586049050092697,
      "num_tokens": 35312994.0,
      "step": 19150
    },
    {
      "entropy": 5.610832977294922,
      "epoch": 1.6092837639151438,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047423574117659703,
      "loss": 5.3136,
      "mean_token_accuracy": 0.17269802391529082,
      "num_tokens": 35322533.0,
      "step": 19155
    },
    {
      "entropy": 5.599225378036499,
      "epoch": 1.6097038437303088,
      "grad_norm": 1.828125,
      "learning_rate": 0.00047422182461990316,
      "loss": 5.3504,
      "mean_token_accuracy": 0.17488384395837783,
      "num_tokens": 35331872.0,
      "step": 19160
    },
    {
      "entropy": 5.577617979049682,
      "epoch": 1.6101239235454736,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047420790453409724,
      "loss": 5.3898,
      "mean_token_accuracy": 0.16724039763212203,
      "num_tokens": 35341517.0,
      "step": 19165
    },
    {
      "entropy": 5.606807851791382,
      "epoch": 1.6105440033606384,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004741939809194258,
      "loss": 5.2978,
      "mean_token_accuracy": 0.17621794044971467,
      "num_tokens": 35350291.0,
      "step": 19170
    },
    {
      "entropy": 5.702043342590332,
      "epoch": 1.6109640831758034,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047418005377613566,
      "loss": 5.4582,
      "mean_token_accuracy": 0.1616254672408104,
      "num_tokens": 35360711.0,
      "step": 19175
    },
    {
      "entropy": 5.675919246673584,
      "epoch": 1.6113841629909684,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004741661231044733,
      "loss": 5.3824,
      "mean_token_accuracy": 0.17175389528274537,
      "num_tokens": 35370069.0,
      "step": 19180
    },
    {
      "entropy": 5.677516508102417,
      "epoch": 1.6118042428061332,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047415218890468577,
      "loss": 5.3513,
      "mean_token_accuracy": 0.17706242352724075,
      "num_tokens": 35380389.0,
      "step": 19185
    },
    {
      "entropy": 5.586080074310303,
      "epoch": 1.612224322621298,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004741382511770197,
      "loss": 5.3494,
      "mean_token_accuracy": 0.17001866102218627,
      "num_tokens": 35389420.0,
      "step": 19190
    },
    {
      "entropy": 5.655878019332886,
      "epoch": 1.6126444024364628,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047412430992172205,
      "loss": 5.4568,
      "mean_token_accuracy": 0.1621379017829895,
      "num_tokens": 35399418.0,
      "step": 19195
    },
    {
      "entropy": 5.576916694641113,
      "epoch": 1.6130644822516278,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047411036513903974,
      "loss": 5.3542,
      "mean_token_accuracy": 0.17415417730808258,
      "num_tokens": 35408717.0,
      "step": 19200
    },
    {
      "entropy": 5.611598634719849,
      "epoch": 1.6134845620667928,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047409641682921987,
      "loss": 5.282,
      "mean_token_accuracy": 0.1839701771736145,
      "num_tokens": 35417118.0,
      "step": 19205
    },
    {
      "entropy": 5.657357597351075,
      "epoch": 1.6139046418819576,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004740824649925096,
      "loss": 5.395,
      "mean_token_accuracy": 0.17100959569215773,
      "num_tokens": 35425526.0,
      "step": 19210
    },
    {
      "entropy": 5.535145854949951,
      "epoch": 1.6143247216971224,
      "grad_norm": 1.625,
      "learning_rate": 0.0004740685096291559,
      "loss": 5.3747,
      "mean_token_accuracy": 0.16799048632383345,
      "num_tokens": 35434932.0,
      "step": 19215
    },
    {
      "entropy": 5.677419805526734,
      "epoch": 1.6147448015122874,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047405455073940597,
      "loss": 5.4143,
      "mean_token_accuracy": 0.16868570894002916,
      "num_tokens": 35443909.0,
      "step": 19220
    },
    {
      "entropy": 5.735181903839111,
      "epoch": 1.6151648813274522,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004740405883235072,
      "loss": 5.3894,
      "mean_token_accuracy": 0.17232444435358046,
      "num_tokens": 35454082.0,
      "step": 19225
    },
    {
      "entropy": 5.7452147006988525,
      "epoch": 1.6155849611426172,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047402662238170694,
      "loss": 5.4488,
      "mean_token_accuracy": 0.16738691329956054,
      "num_tokens": 35464547.0,
      "step": 19230
    },
    {
      "entropy": 5.588961219787597,
      "epoch": 1.616005040957782,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004740126529142526,
      "loss": 5.3173,
      "mean_token_accuracy": 0.1764468416571617,
      "num_tokens": 35473310.0,
      "step": 19235
    },
    {
      "entropy": 5.6116711616516115,
      "epoch": 1.6164251207729468,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004739986799213915,
      "loss": 5.4343,
      "mean_token_accuracy": 0.17493599504232407,
      "num_tokens": 35483502.0,
      "step": 19240
    },
    {
      "entropy": 5.710875844955444,
      "epoch": 1.6168452005881118,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004739847034033713,
      "loss": 5.3991,
      "mean_token_accuracy": 0.1670544594526291,
      "num_tokens": 35493063.0,
      "step": 19245
    },
    {
      "entropy": 5.629021024703979,
      "epoch": 1.6172652804032768,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047397072336043957,
      "loss": 5.3669,
      "mean_token_accuracy": 0.1677888661623001,
      "num_tokens": 35501829.0,
      "step": 19250
    },
    {
      "entropy": 5.621792793273926,
      "epoch": 1.6176853602184416,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047395673979284383,
      "loss": 5.3581,
      "mean_token_accuracy": 0.1690702572464943,
      "num_tokens": 35510411.0,
      "step": 19255
    },
    {
      "entropy": 5.663351249694824,
      "epoch": 1.6181054400336063,
      "grad_norm": 1.265625,
      "learning_rate": 0.000473942752700832,
      "loss": 5.3899,
      "mean_token_accuracy": 0.1687271237373352,
      "num_tokens": 35519571.0,
      "step": 19260
    },
    {
      "entropy": 5.607766580581665,
      "epoch": 1.6185255198487711,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00047392876208465166,
      "loss": 5.3333,
      "mean_token_accuracy": 0.165669484436512,
      "num_tokens": 35527306.0,
      "step": 19265
    },
    {
      "entropy": 5.607378530502319,
      "epoch": 1.6189455996639361,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004739147679445508,
      "loss": 5.3205,
      "mean_token_accuracy": 0.16435217559337617,
      "num_tokens": 35536126.0,
      "step": 19270
    },
    {
      "entropy": 5.642214250564575,
      "epoch": 1.6193656794791011,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004739007702807773,
      "loss": 5.382,
      "mean_token_accuracy": 0.1744124636054039,
      "num_tokens": 35545593.0,
      "step": 19275
    },
    {
      "entropy": 5.555519723892212,
      "epoch": 1.619785759294266,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047388676909357894,
      "loss": 5.3203,
      "mean_token_accuracy": 0.169613254070282,
      "num_tokens": 35554780.0,
      "step": 19280
    },
    {
      "entropy": 5.635881948471069,
      "epoch": 1.6202058391094307,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047387276438320394,
      "loss": 5.3188,
      "mean_token_accuracy": 0.18244855999946594,
      "num_tokens": 35562982.0,
      "step": 19285
    },
    {
      "entropy": 5.692273807525635,
      "epoch": 1.6206259189245955,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004738587561499003,
      "loss": 5.4306,
      "mean_token_accuracy": 0.16878276020288469,
      "num_tokens": 35571528.0,
      "step": 19290
    },
    {
      "entropy": 5.585696363449097,
      "epoch": 1.6210459987397605,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047384474439391615,
      "loss": 5.2765,
      "mean_token_accuracy": 0.1792723834514618,
      "num_tokens": 35580386.0,
      "step": 19295
    },
    {
      "entropy": 5.583816576004028,
      "epoch": 1.6214660785549255,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004738307291154998,
      "loss": 5.2437,
      "mean_token_accuracy": 0.17867328524589537,
      "num_tokens": 35589456.0,
      "step": 19300
    },
    {
      "entropy": 5.620837068557739,
      "epoch": 1.6218861583700903,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004738167103148995,
      "loss": 5.3389,
      "mean_token_accuracy": 0.1736226886510849,
      "num_tokens": 35598116.0,
      "step": 19305
    },
    {
      "entropy": 5.62102165222168,
      "epoch": 1.622306238185255,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047380268799236355,
      "loss": 5.3318,
      "mean_token_accuracy": 0.17170940786600114,
      "num_tokens": 35606481.0,
      "step": 19310
    },
    {
      "entropy": 5.604389429092407,
      "epoch": 1.62272631800042,
      "grad_norm": 1.375,
      "learning_rate": 0.00047378866214814024,
      "loss": 5.33,
      "mean_token_accuracy": 0.17124983221292495,
      "num_tokens": 35615517.0,
      "step": 19315
    },
    {
      "entropy": 5.6375732421875,
      "epoch": 1.6231463978155851,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047377463278247827,
      "loss": 5.3614,
      "mean_token_accuracy": 0.16923375874757768,
      "num_tokens": 35625100.0,
      "step": 19320
    },
    {
      "entropy": 5.640396308898926,
      "epoch": 1.62356647763075,
      "grad_norm": 1.46875,
      "learning_rate": 0.000473760599895626,
      "loss": 5.2859,
      "mean_token_accuracy": 0.1676229327917099,
      "num_tokens": 35634572.0,
      "step": 19325
    },
    {
      "entropy": 5.606831455230713,
      "epoch": 1.6239865574459147,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000473746563487832,
      "loss": 5.323,
      "mean_token_accuracy": 0.17448222637176514,
      "num_tokens": 35643883.0,
      "step": 19330
    },
    {
      "entropy": 5.6158154010772705,
      "epoch": 1.6244066372610795,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047373252355934506,
      "loss": 5.3979,
      "mean_token_accuracy": 0.1719308689236641,
      "num_tokens": 35652527.0,
      "step": 19335
    },
    {
      "entropy": 5.722070121765137,
      "epoch": 1.6248267170762445,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047371848011041375,
      "loss": 5.426,
      "mean_token_accuracy": 0.17274598628282548,
      "num_tokens": 35662436.0,
      "step": 19340
    },
    {
      "entropy": 5.631454086303711,
      "epoch": 1.6252467968914095,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047370443314128687,
      "loss": 5.3225,
      "mean_token_accuracy": 0.17204887717962264,
      "num_tokens": 35672302.0,
      "step": 19345
    },
    {
      "entropy": 5.633196592330933,
      "epoch": 1.6256668767065743,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004736903826522132,
      "loss": 5.3749,
      "mean_token_accuracy": 0.16764545142650605,
      "num_tokens": 35680852.0,
      "step": 19350
    },
    {
      "entropy": 5.650068759918213,
      "epoch": 1.626086956521739,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004736763286434419,
      "loss": 5.3466,
      "mean_token_accuracy": 0.17409290075302125,
      "num_tokens": 35690159.0,
      "step": 19355
    },
    {
      "entropy": 5.607654190063476,
      "epoch": 1.6265070363369039,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004736622711152216,
      "loss": 5.2887,
      "mean_token_accuracy": 0.17731603235006332,
      "num_tokens": 35699165.0,
      "step": 19360
    },
    {
      "entropy": 5.60712890625,
      "epoch": 1.6269271161520689,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004736482100678015,
      "loss": 5.3436,
      "mean_token_accuracy": 0.17809633165597916,
      "num_tokens": 35708910.0,
      "step": 19365
    },
    {
      "entropy": 5.646894550323486,
      "epoch": 1.6273471959672339,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047363414550143063,
      "loss": 5.4227,
      "mean_token_accuracy": 0.16512314975261688,
      "num_tokens": 35718218.0,
      "step": 19370
    },
    {
      "entropy": 5.63840684890747,
      "epoch": 1.6277672757823987,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047362007741635816,
      "loss": 5.3452,
      "mean_token_accuracy": 0.17157889306545257,
      "num_tokens": 35727076.0,
      "step": 19375
    },
    {
      "entropy": 5.62888469696045,
      "epoch": 1.6281873555975634,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004736060058128333,
      "loss": 5.4552,
      "mean_token_accuracy": 0.16568203270435333,
      "num_tokens": 35736316.0,
      "step": 19380
    },
    {
      "entropy": 5.649870681762695,
      "epoch": 1.6286074354127285,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047359193069110533,
      "loss": 5.3996,
      "mean_token_accuracy": 0.1711649164557457,
      "num_tokens": 35745747.0,
      "step": 19385
    },
    {
      "entropy": 5.7266498565673825,
      "epoch": 1.6290275152278935,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047357785205142354,
      "loss": 5.3638,
      "mean_token_accuracy": 0.17583349645137786,
      "num_tokens": 35754825.0,
      "step": 19390
    },
    {
      "entropy": 5.61067762374878,
      "epoch": 1.6294475950430583,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004735637698940374,
      "loss": 5.3287,
      "mean_token_accuracy": 0.17602441608905792,
      "num_tokens": 35764504.0,
      "step": 19395
    },
    {
      "entropy": 5.705884027481079,
      "epoch": 1.629867674858223,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004735496842191963,
      "loss": 5.4114,
      "mean_token_accuracy": 0.17331966012716293,
      "num_tokens": 35774195.0,
      "step": 19400
    },
    {
      "entropy": 5.554973030090332,
      "epoch": 1.6302877546733878,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047353559502714976,
      "loss": 5.2968,
      "mean_token_accuracy": 0.1762244775891304,
      "num_tokens": 35783721.0,
      "step": 19405
    },
    {
      "entropy": 5.58658390045166,
      "epoch": 1.6307078344885528,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004735215023181474,
      "loss": 5.3687,
      "mean_token_accuracy": 0.17109697461128234,
      "num_tokens": 35792821.0,
      "step": 19410
    },
    {
      "entropy": 5.662140369415283,
      "epoch": 1.6311279143037178,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047350740609243883,
      "loss": 5.3866,
      "mean_token_accuracy": 0.16959262937307357,
      "num_tokens": 35802746.0,
      "step": 19415
    },
    {
      "entropy": 5.701506185531616,
      "epoch": 1.6315479941188826,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004734933063502738,
      "loss": 5.3849,
      "mean_token_accuracy": 0.17653521597385408,
      "num_tokens": 35811196.0,
      "step": 19420
    },
    {
      "entropy": 5.775771045684815,
      "epoch": 1.6319680739340474,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047347920309190203,
      "loss": 5.4324,
      "mean_token_accuracy": 0.16661548614501953,
      "num_tokens": 35820787.0,
      "step": 19425
    },
    {
      "entropy": 5.600856828689575,
      "epoch": 1.6323881537492122,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004734650963175734,
      "loss": 5.409,
      "mean_token_accuracy": 0.1677461639046669,
      "num_tokens": 35831247.0,
      "step": 19430
    },
    {
      "entropy": 5.605106353759766,
      "epoch": 1.6328082335643772,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047345098602753777,
      "loss": 5.4335,
      "mean_token_accuracy": 0.166945444047451,
      "num_tokens": 35840759.0,
      "step": 19435
    },
    {
      "entropy": 5.655753040313721,
      "epoch": 1.6332283133795422,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004734368722220451,
      "loss": 5.3812,
      "mean_token_accuracy": 0.16615464687347412,
      "num_tokens": 35850137.0,
      "step": 19440
    },
    {
      "entropy": 5.589809322357178,
      "epoch": 1.633648393194707,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004734227549013455,
      "loss": 5.2166,
      "mean_token_accuracy": 0.1793386995792389,
      "num_tokens": 35858412.0,
      "step": 19445
    },
    {
      "entropy": 5.602662038803101,
      "epoch": 1.6340684730098718,
      "grad_norm": 1.75,
      "learning_rate": 0.0004734086340656889,
      "loss": 5.2948,
      "mean_token_accuracy": 0.1745496153831482,
      "num_tokens": 35868202.0,
      "step": 19450
    },
    {
      "entropy": 5.641941833496094,
      "epoch": 1.6344885528250368,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004733945097153255,
      "loss": 5.3745,
      "mean_token_accuracy": 0.17630792111158372,
      "num_tokens": 35877237.0,
      "step": 19455
    },
    {
      "entropy": 5.54696478843689,
      "epoch": 1.6349086326402016,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004733803818505055,
      "loss": 5.2427,
      "mean_token_accuracy": 0.17969369441270827,
      "num_tokens": 35887016.0,
      "step": 19460
    },
    {
      "entropy": 5.637591075897217,
      "epoch": 1.6353287124553666,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047336625047147924,
      "loss": 5.312,
      "mean_token_accuracy": 0.18022883385419847,
      "num_tokens": 35896393.0,
      "step": 19465
    },
    {
      "entropy": 5.59985933303833,
      "epoch": 1.6357487922705314,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047335211557849693,
      "loss": 5.3729,
      "mean_token_accuracy": 0.16767509430646896,
      "num_tokens": 35905237.0,
      "step": 19470
    },
    {
      "entropy": 5.611186122894287,
      "epoch": 1.6361688720856962,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004733379771718092,
      "loss": 5.3883,
      "mean_token_accuracy": 0.172068952023983,
      "num_tokens": 35914352.0,
      "step": 19475
    },
    {
      "entropy": 5.657137441635132,
      "epoch": 1.6365889519008612,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004733238352516661,
      "loss": 5.4514,
      "mean_token_accuracy": 0.17099396884441376,
      "num_tokens": 35923785.0,
      "step": 19480
    },
    {
      "entropy": 5.716256523132325,
      "epoch": 1.6370090317160262,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047330968981831856,
      "loss": 5.3622,
      "mean_token_accuracy": 0.1690943643450737,
      "num_tokens": 35932495.0,
      "step": 19485
    },
    {
      "entropy": 5.6488525390625,
      "epoch": 1.637429111531191,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047329554087201687,
      "loss": 5.3728,
      "mean_token_accuracy": 0.17575292885303498,
      "num_tokens": 35941745.0,
      "step": 19490
    },
    {
      "entropy": 5.666727590560913,
      "epoch": 1.6378491913463558,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047328138841301186,
      "loss": 5.4183,
      "mean_token_accuracy": 0.16949606239795684,
      "num_tokens": 35950281.0,
      "step": 19495
    },
    {
      "entropy": 5.640449714660645,
      "epoch": 1.6382692711615205,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004732672324415541,
      "loss": 5.3684,
      "mean_token_accuracy": 0.17443180382251738,
      "num_tokens": 35959531.0,
      "step": 19500
    },
    {
      "entropy": 5.660059309005737,
      "epoch": 1.6386893509766856,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004732530729578945,
      "loss": 5.4016,
      "mean_token_accuracy": 0.1761951193213463,
      "num_tokens": 35969462.0,
      "step": 19505
    },
    {
      "entropy": 5.618856143951416,
      "epoch": 1.6391094307918506,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004732389099622837,
      "loss": 5.3821,
      "mean_token_accuracy": 0.17058353871107101,
      "num_tokens": 35978022.0,
      "step": 19510
    },
    {
      "entropy": 5.682546234130859,
      "epoch": 1.6395295106070154,
      "grad_norm": 1.890625,
      "learning_rate": 0.00047322474345497267,
      "loss": 5.3855,
      "mean_token_accuracy": 0.1683097392320633,
      "num_tokens": 35988193.0,
      "step": 19515
    },
    {
      "entropy": 5.683224821090699,
      "epoch": 1.6399495904221801,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047321057343621247,
      "loss": 5.4046,
      "mean_token_accuracy": 0.16349054276943206,
      "num_tokens": 35997404.0,
      "step": 19520
    },
    {
      "entropy": 5.575731086730957,
      "epoch": 1.6403696702373451,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047319639990625395,
      "loss": 5.2799,
      "mean_token_accuracy": 0.17922523021697997,
      "num_tokens": 36005356.0,
      "step": 19525
    },
    {
      "entropy": 5.709725046157837,
      "epoch": 1.64078975005251,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047318222286534824,
      "loss": 5.5431,
      "mean_token_accuracy": 0.16441298499703408,
      "num_tokens": 36015305.0,
      "step": 19530
    },
    {
      "entropy": 5.691441440582276,
      "epoch": 1.641209829867675,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047316804231374663,
      "loss": 5.3842,
      "mean_token_accuracy": 0.1662529468536377,
      "num_tokens": 36024278.0,
      "step": 19535
    },
    {
      "entropy": 5.579236650466919,
      "epoch": 1.6416299096828397,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004731538582517001,
      "loss": 5.2069,
      "mean_token_accuracy": 0.1806870624423027,
      "num_tokens": 36032870.0,
      "step": 19540
    },
    {
      "entropy": 5.536693048477173,
      "epoch": 1.6420499894980045,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047313967067945996,
      "loss": 5.2663,
      "mean_token_accuracy": 0.18009178936481476,
      "num_tokens": 36041725.0,
      "step": 19545
    },
    {
      "entropy": 5.641549587249756,
      "epoch": 1.6424700693131695,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004731254795972777,
      "loss": 5.4045,
      "mean_token_accuracy": 0.16794433146715165,
      "num_tokens": 36050929.0,
      "step": 19550
    },
    {
      "entropy": 5.6989109992980955,
      "epoch": 1.6428901491283345,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004731112850054045,
      "loss": 5.3995,
      "mean_token_accuracy": 0.17151074707508088,
      "num_tokens": 36060059.0,
      "step": 19555
    },
    {
      "entropy": 5.632898187637329,
      "epoch": 1.6433102289434993,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004730970869040919,
      "loss": 5.307,
      "mean_token_accuracy": 0.18627825677394866,
      "num_tokens": 36069445.0,
      "step": 19560
    },
    {
      "entropy": 5.647307300567627,
      "epoch": 1.643730308758664,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047308288529359147,
      "loss": 5.4637,
      "mean_token_accuracy": 0.16950182914733886,
      "num_tokens": 36079129.0,
      "step": 19565
    },
    {
      "entropy": 5.687177848815918,
      "epoch": 1.644150388573829,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004730686801741547,
      "loss": 5.3334,
      "mean_token_accuracy": 0.17340679913759233,
      "num_tokens": 36088320.0,
      "step": 19570
    },
    {
      "entropy": 5.630806398391724,
      "epoch": 1.644570468388994,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004730544715460332,
      "loss": 5.3894,
      "mean_token_accuracy": 0.17694738507270813,
      "num_tokens": 36097728.0,
      "step": 19575
    },
    {
      "entropy": 5.66889214515686,
      "epoch": 1.644990548204159,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047304025940947875,
      "loss": 5.3879,
      "mean_token_accuracy": 0.17116793617606163,
      "num_tokens": 36106566.0,
      "step": 19580
    },
    {
      "entropy": 5.648784923553467,
      "epoch": 1.6454106280193237,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047302604376474306,
      "loss": 5.3379,
      "mean_token_accuracy": 0.1719046488404274,
      "num_tokens": 36115475.0,
      "step": 19585
    },
    {
      "entropy": 5.606548929214478,
      "epoch": 1.6458307078344885,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047301182461207807,
      "loss": 5.4482,
      "mean_token_accuracy": 0.17226337492465973,
      "num_tokens": 36124404.0,
      "step": 19590
    },
    {
      "entropy": 5.58370943069458,
      "epoch": 1.6462507876496533,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00047299760195173554,
      "loss": 5.2991,
      "mean_token_accuracy": 0.1753391683101654,
      "num_tokens": 36132987.0,
      "step": 19595
    },
    {
      "entropy": 5.650071668624878,
      "epoch": 1.6466708674648183,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004729833757839673,
      "loss": 5.4392,
      "mean_token_accuracy": 0.1720758482813835,
      "num_tokens": 36142163.0,
      "step": 19600
    },
    {
      "entropy": 5.641680431365967,
      "epoch": 1.6470909472799833,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047296914610902565,
      "loss": 5.4081,
      "mean_token_accuracy": 0.16769561767578126,
      "num_tokens": 36152561.0,
      "step": 19605
    },
    {
      "entropy": 5.653435897827149,
      "epoch": 1.647511027095148,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047295491292716245,
      "loss": 5.3205,
      "mean_token_accuracy": 0.1730746880173683,
      "num_tokens": 36161877.0,
      "step": 19610
    },
    {
      "entropy": 5.675718498229981,
      "epoch": 1.6479311069103129,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047294067623862996,
      "loss": 5.3841,
      "mean_token_accuracy": 0.16492767482995987,
      "num_tokens": 36171523.0,
      "step": 19615
    },
    {
      "entropy": 5.593134927749634,
      "epoch": 1.6483511867254779,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047292643604368025,
      "loss": 5.3149,
      "mean_token_accuracy": 0.1766111820936203,
      "num_tokens": 36180339.0,
      "step": 19620
    },
    {
      "entropy": 5.6570531845092775,
      "epoch": 1.6487712665406429,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004729121923425657,
      "loss": 5.4118,
      "mean_token_accuracy": 0.16898104399442673,
      "num_tokens": 36191584.0,
      "step": 19625
    },
    {
      "entropy": 5.7628068923950195,
      "epoch": 1.6491913463558077,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004728979451355385,
      "loss": 5.4211,
      "mean_token_accuracy": 0.17653830349445343,
      "num_tokens": 36200738.0,
      "step": 19630
    },
    {
      "entropy": 5.610785245895386,
      "epoch": 1.6496114261709725,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047288369442285115,
      "loss": 5.2565,
      "mean_token_accuracy": 0.1821356624364853,
      "num_tokens": 36209394.0,
      "step": 19635
    },
    {
      "entropy": 5.603229665756226,
      "epoch": 1.6500315059861372,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047286944020475606,
      "loss": 5.3709,
      "mean_token_accuracy": 0.17178926318883897,
      "num_tokens": 36218268.0,
      "step": 19640
    },
    {
      "entropy": 5.611238861083985,
      "epoch": 1.6504515858013022,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004728551824815057,
      "loss": 5.3123,
      "mean_token_accuracy": 0.17810044586658477,
      "num_tokens": 36226974.0,
      "step": 19645
    },
    {
      "entropy": 5.55068359375,
      "epoch": 1.6508716656164673,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047284092125335277,
      "loss": 5.2788,
      "mean_token_accuracy": 0.17575833350419998,
      "num_tokens": 36235892.0,
      "step": 19650
    },
    {
      "entropy": 5.547413349151611,
      "epoch": 1.651291745431632,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004728266565205497,
      "loss": 5.2892,
      "mean_token_accuracy": 0.17635726779699326,
      "num_tokens": 36244750.0,
      "step": 19655
    },
    {
      "entropy": 5.648470115661621,
      "epoch": 1.6517118252467968,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047281238828334924,
      "loss": 5.3497,
      "mean_token_accuracy": 0.17414160966873168,
      "num_tokens": 36254902.0,
      "step": 19660
    },
    {
      "entropy": 5.66616415977478,
      "epoch": 1.6521319050619616,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004727981165420042,
      "loss": 5.4005,
      "mean_token_accuracy": 0.17089224606752396,
      "num_tokens": 36265546.0,
      "step": 19665
    },
    {
      "entropy": 5.531275129318237,
      "epoch": 1.6525519848771266,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004727838412967674,
      "loss": 5.3116,
      "mean_token_accuracy": 0.16871818602085115,
      "num_tokens": 36273978.0,
      "step": 19670
    },
    {
      "entropy": 5.6874082565307615,
      "epoch": 1.6529720646922916,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004727695625478917,
      "loss": 5.3488,
      "mean_token_accuracy": 0.17191920876502992,
      "num_tokens": 36283117.0,
      "step": 19675
    },
    {
      "entropy": 5.751308298110962,
      "epoch": 1.6533921445074564,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047275528029562996,
      "loss": 5.3334,
      "mean_token_accuracy": 0.16691725105047225,
      "num_tokens": 36293031.0,
      "step": 19680
    },
    {
      "entropy": 5.5786652088165285,
      "epoch": 1.6538122243226212,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047274099454023535,
      "loss": 5.3332,
      "mean_token_accuracy": 0.17999812811613083,
      "num_tokens": 36302080.0,
      "step": 19685
    },
    {
      "entropy": 5.574540090560913,
      "epoch": 1.6542323041377862,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047272670528196084,
      "loss": 5.3551,
      "mean_token_accuracy": 0.17055315971374513,
      "num_tokens": 36311077.0,
      "step": 19690
    },
    {
      "entropy": 5.591321849822998,
      "epoch": 1.6546523839529512,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004727124125210595,
      "loss": 5.2989,
      "mean_token_accuracy": 0.17874737232923507,
      "num_tokens": 36320300.0,
      "step": 19695
    },
    {
      "entropy": 5.636029148101807,
      "epoch": 1.655072463768116,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047269811625778456,
      "loss": 5.3553,
      "mean_token_accuracy": 0.17314265221357344,
      "num_tokens": 36330184.0,
      "step": 19700
    },
    {
      "entropy": 5.58466100692749,
      "epoch": 1.6554925435832808,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004726838164923893,
      "loss": 5.3684,
      "mean_token_accuracy": 0.17024628221988677,
      "num_tokens": 36339526.0,
      "step": 19705
    },
    {
      "entropy": 5.598156499862671,
      "epoch": 1.6559126233984456,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00047266951322512716,
      "loss": 5.3573,
      "mean_token_accuracy": 0.17232748866081238,
      "num_tokens": 36348849.0,
      "step": 19710
    },
    {
      "entropy": 5.675782585144043,
      "epoch": 1.6563327032136106,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047265520645625123,
      "loss": 5.3391,
      "mean_token_accuracy": 0.1653357744216919,
      "num_tokens": 36358924.0,
      "step": 19715
    },
    {
      "entropy": 5.729450988769531,
      "epoch": 1.6567527830287756,
      "grad_norm": 3.4375,
      "learning_rate": 0.00047264089618601513,
      "loss": 5.3878,
      "mean_token_accuracy": 0.17455374002456664,
      "num_tokens": 36367130.0,
      "step": 19720
    },
    {
      "entropy": 5.59416561126709,
      "epoch": 1.6571728628439404,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004726265824146724,
      "loss": 5.3355,
      "mean_token_accuracy": 0.17004982084035875,
      "num_tokens": 36376575.0,
      "step": 19725
    },
    {
      "entropy": 5.50741171836853,
      "epoch": 1.6575929426591052,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004726122651424764,
      "loss": 5.2622,
      "mean_token_accuracy": 0.17389485687017442,
      "num_tokens": 36385010.0,
      "step": 19730
    },
    {
      "entropy": 5.552058601379395,
      "epoch": 1.65801302247427,
      "grad_norm": 1.359375,
      "learning_rate": 0.000472597944369681,
      "loss": 5.073,
      "mean_token_accuracy": 0.18783630281686783,
      "num_tokens": 36393574.0,
      "step": 19735
    },
    {
      "entropy": 5.642769813537598,
      "epoch": 1.658433102289435,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047258362009653965,
      "loss": 5.2894,
      "mean_token_accuracy": 0.17900355607271196,
      "num_tokens": 36401992.0,
      "step": 19740
    },
    {
      "entropy": 5.658441543579102,
      "epoch": 1.6588531821046,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047256929232330624,
      "loss": 5.4346,
      "mean_token_accuracy": 0.16492825895547866,
      "num_tokens": 36411712.0,
      "step": 19745
    },
    {
      "entropy": 5.565474510192871,
      "epoch": 1.6592732619197648,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004725549610502346,
      "loss": 5.2528,
      "mean_token_accuracy": 0.17570376843214036,
      "num_tokens": 36420240.0,
      "step": 19750
    },
    {
      "entropy": 5.635782623291016,
      "epoch": 1.6596933417349296,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047254062627757854,
      "loss": 5.3637,
      "mean_token_accuracy": 0.17530056238174438,
      "num_tokens": 36430068.0,
      "step": 19755
    },
    {
      "entropy": 5.718810319900513,
      "epoch": 1.6601134215500946,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000472526288005592,
      "loss": 5.4225,
      "mean_token_accuracy": 0.16737223863601686,
      "num_tokens": 36439808.0,
      "step": 19760
    },
    {
      "entropy": 5.656303453445434,
      "epoch": 1.6605335013652593,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000472511946234529,
      "loss": 5.3511,
      "mean_token_accuracy": 0.1715448319911957,
      "num_tokens": 36449609.0,
      "step": 19765
    },
    {
      "entropy": 5.6782220840454105,
      "epoch": 1.6609535811804244,
      "grad_norm": 1.375,
      "learning_rate": 0.0004724976009646435,
      "loss": 5.3123,
      "mean_token_accuracy": 0.17736656814813614,
      "num_tokens": 36457700.0,
      "step": 19770
    },
    {
      "entropy": 5.6240010261535645,
      "epoch": 1.6613736609955891,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004724832521961897,
      "loss": 5.3713,
      "mean_token_accuracy": 0.16938938945531845,
      "num_tokens": 36466881.0,
      "step": 19775
    },
    {
      "entropy": 5.6949227809906,
      "epoch": 1.661793740810754,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047246889992942187,
      "loss": 5.4742,
      "mean_token_accuracy": 0.1644657611846924,
      "num_tokens": 36475433.0,
      "step": 19780
    },
    {
      "entropy": 5.638772344589233,
      "epoch": 1.662213820625919,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004724545441645941,
      "loss": 5.3745,
      "mean_token_accuracy": 0.1678900107741356,
      "num_tokens": 36484232.0,
      "step": 19785
    },
    {
      "entropy": 5.688452816009521,
      "epoch": 1.662633900441084,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004724401849019608,
      "loss": 5.509,
      "mean_token_accuracy": 0.16093920022249222,
      "num_tokens": 36493588.0,
      "step": 19790
    },
    {
      "entropy": 5.600741863250732,
      "epoch": 1.6630539802562487,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047242582214177616,
      "loss": 5.269,
      "mean_token_accuracy": 0.16896196454763412,
      "num_tokens": 36502289.0,
      "step": 19795
    },
    {
      "entropy": 5.65166916847229,
      "epoch": 1.6634740600714135,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047241145588429483,
      "loss": 5.4177,
      "mean_token_accuracy": 0.1660093657672405,
      "num_tokens": 36511978.0,
      "step": 19800
    },
    {
      "entropy": 5.718343210220337,
      "epoch": 1.6638941398865783,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004723970861297712,
      "loss": 5.3984,
      "mean_token_accuracy": 0.17212554365396499,
      "num_tokens": 36520378.0,
      "step": 19805
    },
    {
      "entropy": 5.6724714756011965,
      "epoch": 1.6643142197017433,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004723827128784599,
      "loss": 5.3744,
      "mean_token_accuracy": 0.17019625902175903,
      "num_tokens": 36529965.0,
      "step": 19810
    },
    {
      "entropy": 5.7348864555358885,
      "epoch": 1.6647342995169083,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047236833613061534,
      "loss": 5.3928,
      "mean_token_accuracy": 0.17518503814935685,
      "num_tokens": 36539394.0,
      "step": 19815
    },
    {
      "entropy": 5.630881309509277,
      "epoch": 1.665154379332073,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004723539558864925,
      "loss": 5.4405,
      "mean_token_accuracy": 0.16701636016368865,
      "num_tokens": 36548608.0,
      "step": 19820
    },
    {
      "entropy": 5.652702283859253,
      "epoch": 1.665574459147238,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004723395721463459,
      "loss": 5.3157,
      "mean_token_accuracy": 0.1754217267036438,
      "num_tokens": 36557736.0,
      "step": 19825
    },
    {
      "entropy": 5.666959190368653,
      "epoch": 1.665994538962403,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004723251849104303,
      "loss": 5.3357,
      "mean_token_accuracy": 0.16968396306037903,
      "num_tokens": 36566745.0,
      "step": 19830
    },
    {
      "entropy": 5.573006582260132,
      "epoch": 1.6664146187775677,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047231079417900076,
      "loss": 5.2756,
      "mean_token_accuracy": 0.17210672199726104,
      "num_tokens": 36575956.0,
      "step": 19835
    },
    {
      "entropy": 5.6266522884368895,
      "epoch": 1.6668346985927327,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000472296399952312,
      "loss": 5.335,
      "mean_token_accuracy": 0.17685381472110748,
      "num_tokens": 36584673.0,
      "step": 19840
    },
    {
      "entropy": 5.6374927997589115,
      "epoch": 1.6672547784078975,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004722820022306192,
      "loss": 5.3894,
      "mean_token_accuracy": 0.17263043373823167,
      "num_tokens": 36593758.0,
      "step": 19845
    },
    {
      "entropy": 5.558726692199707,
      "epoch": 1.6676748582230623,
      "grad_norm": 1.625,
      "learning_rate": 0.0004722676010141773,
      "loss": 5.2582,
      "mean_token_accuracy": 0.16958121508359908,
      "num_tokens": 36603722.0,
      "step": 19850
    },
    {
      "entropy": 5.610972261428833,
      "epoch": 1.6680949380382273,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047225319630324136,
      "loss": 5.3043,
      "mean_token_accuracy": 0.17388844192028047,
      "num_tokens": 36612478.0,
      "step": 19855
    },
    {
      "entropy": 5.652765798568725,
      "epoch": 1.6685150178533923,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004722387880980667,
      "loss": 5.513,
      "mean_token_accuracy": 0.1610901191830635,
      "num_tokens": 36622399.0,
      "step": 19860
    },
    {
      "entropy": 5.713167476654053,
      "epoch": 1.668935097668557,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00047222437639890844,
      "loss": 5.3483,
      "mean_token_accuracy": 0.17080966681241988,
      "num_tokens": 36631798.0,
      "step": 19865
    },
    {
      "entropy": 5.566953468322754,
      "epoch": 1.6693551774837219,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047220996120602197,
      "loss": 5.3491,
      "mean_token_accuracy": 0.17151901721954346,
      "num_tokens": 36640405.0,
      "step": 19870
    },
    {
      "entropy": 5.711348295211792,
      "epoch": 1.6697752572988867,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047219554251966246,
      "loss": 5.499,
      "mean_token_accuracy": 0.16270583868026733,
      "num_tokens": 36650209.0,
      "step": 19875
    },
    {
      "entropy": 5.776115036010742,
      "epoch": 1.6701953371140517,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004721811203400855,
      "loss": 5.4447,
      "mean_token_accuracy": 0.1687805637717247,
      "num_tokens": 36660248.0,
      "step": 19880
    },
    {
      "entropy": 5.638616371154785,
      "epoch": 1.6706154169292167,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047216669466754657,
      "loss": 5.3301,
      "mean_token_accuracy": 0.17210840731859206,
      "num_tokens": 36669938.0,
      "step": 19885
    },
    {
      "entropy": 5.515147495269775,
      "epoch": 1.6710354967443815,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004721522655023012,
      "loss": 5.3521,
      "mean_token_accuracy": 0.17559095174074174,
      "num_tokens": 36679903.0,
      "step": 19890
    },
    {
      "entropy": 5.726369476318359,
      "epoch": 1.6714555765595462,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004721378328446049,
      "loss": 5.4151,
      "mean_token_accuracy": 0.168327933549881,
      "num_tokens": 36688424.0,
      "step": 19895
    },
    {
      "entropy": 5.7589884281158445,
      "epoch": 1.6718756563747112,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004721233966947134,
      "loss": 5.4881,
      "mean_token_accuracy": 0.16615458354353904,
      "num_tokens": 36698715.0,
      "step": 19900
    },
    {
      "entropy": 5.535595369338989,
      "epoch": 1.672295736189876,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047210895705288237,
      "loss": 5.3404,
      "mean_token_accuracy": 0.1847057580947876,
      "num_tokens": 36708456.0,
      "step": 19905
    },
    {
      "entropy": 5.6442008972167965,
      "epoch": 1.672715816005041,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004720945139193678,
      "loss": 5.3693,
      "mean_token_accuracy": 0.17023673355579377,
      "num_tokens": 36717596.0,
      "step": 19910
    },
    {
      "entropy": 5.7017570495605465,
      "epoch": 1.6731358958202058,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004720800672944253,
      "loss": 5.4672,
      "mean_token_accuracy": 0.1671689987182617,
      "num_tokens": 36727092.0,
      "step": 19915
    },
    {
      "entropy": 5.6182774066925045,
      "epoch": 1.6735559756353706,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004720656171783109,
      "loss": 5.1831,
      "mean_token_accuracy": 0.1812111958861351,
      "num_tokens": 36735910.0,
      "step": 19920
    },
    {
      "entropy": 5.534506034851074,
      "epoch": 1.6739760554505356,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004720511635712806,
      "loss": 5.3031,
      "mean_token_accuracy": 0.1774997293949127,
      "num_tokens": 36745237.0,
      "step": 19925
    },
    {
      "entropy": 5.663472890853882,
      "epoch": 1.6743961352657006,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047203670647359035,
      "loss": 5.4333,
      "mean_token_accuracy": 0.17369587272405623,
      "num_tokens": 36753603.0,
      "step": 19930
    },
    {
      "entropy": 5.742884635925293,
      "epoch": 1.6748162150808654,
      "grad_norm": 1.25,
      "learning_rate": 0.0004720222458854964,
      "loss": 5.4302,
      "mean_token_accuracy": 0.16438519358634948,
      "num_tokens": 36763010.0,
      "step": 19935
    },
    {
      "entropy": 5.700931024551392,
      "epoch": 1.6752362948960302,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047200778180725477,
      "loss": 5.3445,
      "mean_token_accuracy": 0.17566305994987488,
      "num_tokens": 36772156.0,
      "step": 19940
    },
    {
      "entropy": 5.604983901977539,
      "epoch": 1.675656374711195,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047199331423912174,
      "loss": 5.263,
      "mean_token_accuracy": 0.17966575622558595,
      "num_tokens": 36781386.0,
      "step": 19945
    },
    {
      "entropy": 5.6522300243377686,
      "epoch": 1.67607645452636,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004719788431813536,
      "loss": 5.4082,
      "mean_token_accuracy": 0.1677954003214836,
      "num_tokens": 36790754.0,
      "step": 19950
    },
    {
      "entropy": 5.715110874176025,
      "epoch": 1.676496534341525,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004719643686342066,
      "loss": 5.3842,
      "mean_token_accuracy": 0.1711716651916504,
      "num_tokens": 36799623.0,
      "step": 19955
    },
    {
      "entropy": 5.50349645614624,
      "epoch": 1.6769166141566898,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004719498905979373,
      "loss": 5.2022,
      "mean_token_accuracy": 0.18366880118846893,
      "num_tokens": 36808662.0,
      "step": 19960
    },
    {
      "entropy": 5.606126165390014,
      "epoch": 1.6773366939718546,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004719354090728021,
      "loss": 5.333,
      "mean_token_accuracy": 0.16664473712444305,
      "num_tokens": 36817730.0,
      "step": 19965
    },
    {
      "entropy": 5.675943326950073,
      "epoch": 1.6777567737870194,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047192092405905743,
      "loss": 5.299,
      "mean_token_accuracy": 0.1727191686630249,
      "num_tokens": 36827203.0,
      "step": 19970
    },
    {
      "entropy": 5.69139609336853,
      "epoch": 1.6781768536021844,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004719064355569601,
      "loss": 5.4747,
      "mean_token_accuracy": 0.16684583872556685,
      "num_tokens": 36836145.0,
      "step": 19975
    },
    {
      "entropy": 5.675600862503051,
      "epoch": 1.6785969334173494,
      "grad_norm": 1.5,
      "learning_rate": 0.00047189194356676666,
      "loss": 5.4615,
      "mean_token_accuracy": 0.17299748361110687,
      "num_tokens": 36845609.0,
      "step": 19980
    },
    {
      "entropy": 5.671309185028076,
      "epoch": 1.6790170132325142,
      "grad_norm": 1.765625,
      "learning_rate": 0.00047187744808873386,
      "loss": 5.4776,
      "mean_token_accuracy": 0.1707234039902687,
      "num_tokens": 36855367.0,
      "step": 19985
    },
    {
      "entropy": 5.660668659210205,
      "epoch": 1.679437093047679,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047186294912311835,
      "loss": 5.4361,
      "mean_token_accuracy": 0.16343557834625244,
      "num_tokens": 36864808.0,
      "step": 19990
    },
    {
      "entropy": 5.650803327560425,
      "epoch": 1.679857172862844,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047184844667017705,
      "loss": 5.2957,
      "mean_token_accuracy": 0.18071594536304475,
      "num_tokens": 36873651.0,
      "step": 19995
    },
    {
      "entropy": 5.63038010597229,
      "epoch": 1.680277252678009,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047183394073016695,
      "loss": 5.4275,
      "mean_token_accuracy": 0.16843625009059907,
      "num_tokens": 36883227.0,
      "step": 20000
    },
    {
      "entropy": 5.542231464385987,
      "epoch": 1.6806973324931738,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047181943130334493,
      "loss": 5.2085,
      "mean_token_accuracy": 0.1798401266336441,
      "num_tokens": 36891628.0,
      "step": 20005
    },
    {
      "entropy": 5.587174034118652,
      "epoch": 1.6811174123083386,
      "grad_norm": 1.328125,
      "learning_rate": 0.000471804918389968,
      "loss": 5.3465,
      "mean_token_accuracy": 0.16603483706712724,
      "num_tokens": 36901819.0,
      "step": 20010
    },
    {
      "entropy": 5.620696210861206,
      "epoch": 1.6815374921235033,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004717904019902933,
      "loss": 5.3775,
      "mean_token_accuracy": 0.16788558214902877,
      "num_tokens": 36911206.0,
      "step": 20015
    },
    {
      "entropy": 5.59770016670227,
      "epoch": 1.6819575719386683,
      "grad_norm": 1.390625,
      "learning_rate": 0.000471775882104578,
      "loss": 5.3248,
      "mean_token_accuracy": 0.1724284067749977,
      "num_tokens": 36920830.0,
      "step": 20020
    },
    {
      "entropy": 5.59469518661499,
      "epoch": 1.6823776517538334,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047176135873307917,
      "loss": 5.2371,
      "mean_token_accuracy": 0.17292287498712539,
      "num_tokens": 36929702.0,
      "step": 20025
    },
    {
      "entropy": 5.72202959060669,
      "epoch": 1.6827977315689981,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004717468318760543,
      "loss": 5.4422,
      "mean_token_accuracy": 0.17322811037302016,
      "num_tokens": 36938423.0,
      "step": 20030
    },
    {
      "entropy": 5.637274074554443,
      "epoch": 1.683217811384163,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047173230153376057,
      "loss": 5.3657,
      "mean_token_accuracy": 0.17268821895122527,
      "num_tokens": 36947198.0,
      "step": 20035
    },
    {
      "entropy": 5.681332159042358,
      "epoch": 1.6836378911993277,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004717177677064554,
      "loss": 5.3475,
      "mean_token_accuracy": 0.17645548433065414,
      "num_tokens": 36955636.0,
      "step": 20040
    },
    {
      "entropy": 5.5717028141021725,
      "epoch": 1.6840579710144927,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047170323039439634,
      "loss": 5.2942,
      "mean_token_accuracy": 0.17198298275470733,
      "num_tokens": 36964463.0,
      "step": 20045
    },
    {
      "entropy": 5.653405570983887,
      "epoch": 1.6844780508296577,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004716886895978408,
      "loss": 5.4239,
      "mean_token_accuracy": 0.17059071063995362,
      "num_tokens": 36974043.0,
      "step": 20050
    },
    {
      "entropy": 5.650971984863281,
      "epoch": 1.6848981306448225,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047167414531704637,
      "loss": 5.3113,
      "mean_token_accuracy": 0.17826482355594636,
      "num_tokens": 36983856.0,
      "step": 20055
    },
    {
      "entropy": 5.586727428436279,
      "epoch": 1.6853182104599873,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047165959755227077,
      "loss": 5.3404,
      "mean_token_accuracy": 0.1794610232114792,
      "num_tokens": 36992664.0,
      "step": 20060
    },
    {
      "entropy": 5.637934684753418,
      "epoch": 1.6857382902751523,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047164504630377166,
      "loss": 5.3883,
      "mean_token_accuracy": 0.17629259526729585,
      "num_tokens": 37001826.0,
      "step": 20065
    },
    {
      "entropy": 5.751244449615479,
      "epoch": 1.6861583700903173,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047163049157180676,
      "loss": 5.4197,
      "mean_token_accuracy": 0.1678980588912964,
      "num_tokens": 37010821.0,
      "step": 20070
    },
    {
      "entropy": 5.66287088394165,
      "epoch": 1.6865784499054821,
      "grad_norm": 1.390625,
      "learning_rate": 0.000471615933356634,
      "loss": 5.511,
      "mean_token_accuracy": 0.16093569695949556,
      "num_tokens": 37021293.0,
      "step": 20075
    },
    {
      "entropy": 5.621215581893921,
      "epoch": 1.686998529720647,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004716013716585112,
      "loss": 5.2801,
      "mean_token_accuracy": 0.1768314480781555,
      "num_tokens": 37031063.0,
      "step": 20080
    },
    {
      "entropy": 5.6183741092681885,
      "epoch": 1.6874186095358117,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004715868064776964,
      "loss": 5.3467,
      "mean_token_accuracy": 0.1790781870484352,
      "num_tokens": 37040879.0,
      "step": 20085
    },
    {
      "entropy": 5.582546186447144,
      "epoch": 1.6878386893509767,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004715722378144474,
      "loss": 5.2369,
      "mean_token_accuracy": 0.17983479499816896,
      "num_tokens": 37049452.0,
      "step": 20090
    },
    {
      "entropy": 5.526564359664917,
      "epoch": 1.6882587691661417,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004715576656690225,
      "loss": 5.2072,
      "mean_token_accuracy": 0.1780251145362854,
      "num_tokens": 37058010.0,
      "step": 20095
    },
    {
      "entropy": 5.623370695114136,
      "epoch": 1.6886788489813065,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047154309004167984,
      "loss": 5.4392,
      "mean_token_accuracy": 0.16597311049699784,
      "num_tokens": 37067580.0,
      "step": 20100
    },
    {
      "entropy": 5.5723961353302,
      "epoch": 1.6890989287964713,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047152851093267744,
      "loss": 5.2972,
      "mean_token_accuracy": 0.17383439987897872,
      "num_tokens": 37076584.0,
      "step": 20105
    },
    {
      "entropy": 5.624492597579956,
      "epoch": 1.689519008611636,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004715139283422737,
      "loss": 5.3287,
      "mean_token_accuracy": 0.1717214450240135,
      "num_tokens": 37086330.0,
      "step": 20110
    },
    {
      "entropy": 5.718088245391845,
      "epoch": 1.689939088426801,
      "grad_norm": 1.34375,
      "learning_rate": 0.000471499342270727,
      "loss": 5.3795,
      "mean_token_accuracy": 0.16804536879062654,
      "num_tokens": 37096323.0,
      "step": 20115
    },
    {
      "entropy": 5.566448926925659,
      "epoch": 1.690359168241966,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047148475271829556,
      "loss": 5.3369,
      "mean_token_accuracy": 0.17277998626232147,
      "num_tokens": 37106281.0,
      "step": 20120
    },
    {
      "entropy": 5.553963279724121,
      "epoch": 1.6907792480571309,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004714701596852379,
      "loss": 5.3093,
      "mean_token_accuracy": 0.18138035386800766,
      "num_tokens": 37116002.0,
      "step": 20125
    },
    {
      "entropy": 5.671520948410034,
      "epoch": 1.6911993278722957,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004714555631718125,
      "loss": 5.3866,
      "mean_token_accuracy": 0.17355457693338394,
      "num_tokens": 37125125.0,
      "step": 20130
    },
    {
      "entropy": 5.602753067016602,
      "epoch": 1.6916194076874607,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000471440963178278,
      "loss": 5.2238,
      "mean_token_accuracy": 0.17859983891248704,
      "num_tokens": 37134358.0,
      "step": 20135
    },
    {
      "entropy": 5.697351264953613,
      "epoch": 1.6920394875026254,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047142635970489293,
      "loss": 5.391,
      "mean_token_accuracy": 0.17513249516487123,
      "num_tokens": 37143732.0,
      "step": 20140
    },
    {
      "entropy": 5.603771734237671,
      "epoch": 1.6924595673177905,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004714117527519161,
      "loss": 5.3101,
      "mean_token_accuracy": 0.17167341709136963,
      "num_tokens": 37153809.0,
      "step": 20145
    },
    {
      "entropy": 5.6048431396484375,
      "epoch": 1.6928796471329552,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047139714231960616,
      "loss": 5.3323,
      "mean_token_accuracy": 0.16559370905160903,
      "num_tokens": 37163272.0,
      "step": 20150
    },
    {
      "entropy": 5.580324983596801,
      "epoch": 1.69329972694812,
      "grad_norm": 1.703125,
      "learning_rate": 0.000471382528408222,
      "loss": 5.2788,
      "mean_token_accuracy": 0.17852064073085785,
      "num_tokens": 37172323.0,
      "step": 20155
    },
    {
      "entropy": 5.675992965698242,
      "epoch": 1.693719806763285,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004713679110180225,
      "loss": 5.4522,
      "mean_token_accuracy": 0.16905311048030852,
      "num_tokens": 37181262.0,
      "step": 20160
    },
    {
      "entropy": 5.680888271331787,
      "epoch": 1.69413988657845,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004713532901492666,
      "loss": 5.4176,
      "mean_token_accuracy": 0.17566002309322357,
      "num_tokens": 37189576.0,
      "step": 20165
    },
    {
      "entropy": 5.71924204826355,
      "epoch": 1.6945599663936148,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004713386658022132,
      "loss": 5.3988,
      "mean_token_accuracy": 0.16911477744579315,
      "num_tokens": 37199502.0,
      "step": 20170
    },
    {
      "entropy": 5.64546275138855,
      "epoch": 1.6949800462087796,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004713240379771214,
      "loss": 5.3386,
      "mean_token_accuracy": 0.16714672446250917,
      "num_tokens": 37209028.0,
      "step": 20175
    },
    {
      "entropy": 5.681789112091065,
      "epoch": 1.6954001260239444,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004713094066742505,
      "loss": 5.4671,
      "mean_token_accuracy": 0.16919659227132797,
      "num_tokens": 37218087.0,
      "step": 20180
    },
    {
      "entropy": 5.655849552154541,
      "epoch": 1.6958202058391094,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047129477189385946,
      "loss": 5.4214,
      "mean_token_accuracy": 0.1684677392244339,
      "num_tokens": 37227345.0,
      "step": 20185
    },
    {
      "entropy": 5.712176752090454,
      "epoch": 1.6962402856542744,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004712801336362076,
      "loss": 5.3433,
      "mean_token_accuracy": 0.17029784768819808,
      "num_tokens": 37236011.0,
      "step": 20190
    },
    {
      "entropy": 5.624071741104126,
      "epoch": 1.6966603654694392,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004712654919015543,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1707897588610649,
      "num_tokens": 37244613.0,
      "step": 20195
    },
    {
      "entropy": 5.632029867172241,
      "epoch": 1.697080445284604,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004712508466901588,
      "loss": 5.3588,
      "mean_token_accuracy": 0.17021910697221757,
      "num_tokens": 37253768.0,
      "step": 20200
    },
    {
      "entropy": 5.714126300811768,
      "epoch": 1.697500525099769,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047123619800228057,
      "loss": 5.4254,
      "mean_token_accuracy": 0.16638519167900084,
      "num_tokens": 37263230.0,
      "step": 20205
    },
    {
      "entropy": 5.6618905544281,
      "epoch": 1.6979206049149338,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004712215458381792,
      "loss": 5.3196,
      "mean_token_accuracy": 0.16965548694133759,
      "num_tokens": 37272752.0,
      "step": 20210
    },
    {
      "entropy": 5.6685765266418455,
      "epoch": 1.6983406847300988,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004712068901981142,
      "loss": 5.3641,
      "mean_token_accuracy": 0.17733877152204514,
      "num_tokens": 37281465.0,
      "step": 20215
    },
    {
      "entropy": 5.619257259368896,
      "epoch": 1.6987607645452636,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004711922310823452,
      "loss": 5.3527,
      "mean_token_accuracy": 0.17117764949798583,
      "num_tokens": 37290408.0,
      "step": 20220
    },
    {
      "entropy": 5.6264947891235355,
      "epoch": 1.6991808443604284,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004711775684911318,
      "loss": 5.3098,
      "mean_token_accuracy": 0.18032520711421968,
      "num_tokens": 37298890.0,
      "step": 20225
    },
    {
      "entropy": 5.587586355209351,
      "epoch": 1.6996009241755934,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047116290242473375,
      "loss": 5.3268,
      "mean_token_accuracy": 0.16748166382312774,
      "num_tokens": 37307720.0,
      "step": 20230
    },
    {
      "entropy": 5.589243173599243,
      "epoch": 1.7000210039907584,
      "grad_norm": 2.78125,
      "learning_rate": 0.000471148232883411,
      "loss": 5.3612,
      "mean_token_accuracy": 0.17151859849691392,
      "num_tokens": 37317145.0,
      "step": 20235
    },
    {
      "entropy": 5.6368701457977295,
      "epoch": 1.7004410838059232,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047113355986742325,
      "loss": 5.3024,
      "mean_token_accuracy": 0.17907510250806807,
      "num_tokens": 37326579.0,
      "step": 20240
    },
    {
      "entropy": 5.653764629364014,
      "epoch": 1.700861163621088,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047111888337703046,
      "loss": 5.3823,
      "mean_token_accuracy": 0.16684922128915786,
      "num_tokens": 37336065.0,
      "step": 20245
    },
    {
      "entropy": 5.594270467758179,
      "epoch": 1.7012812434362528,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004711042034124926,
      "loss": 5.254,
      "mean_token_accuracy": 0.17758515924215318,
      "num_tokens": 37345297.0,
      "step": 20250
    },
    {
      "entropy": 5.677919578552246,
      "epoch": 1.7017013232514178,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004710895199740698,
      "loss": 5.409,
      "mean_token_accuracy": 0.17235231548547744,
      "num_tokens": 37354942.0,
      "step": 20255
    },
    {
      "entropy": 5.728347396850586,
      "epoch": 1.7021214030665828,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004710748330620222,
      "loss": 5.2902,
      "mean_token_accuracy": 0.17637401223182678,
      "num_tokens": 37364068.0,
      "step": 20260
    },
    {
      "entropy": 5.621635293960571,
      "epoch": 1.7025414828817476,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004710601426766098,
      "loss": 5.4024,
      "mean_token_accuracy": 0.17034409195184708,
      "num_tokens": 37373256.0,
      "step": 20265
    },
    {
      "entropy": 5.549058198928833,
      "epoch": 1.7029615626969123,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047104544881809295,
      "loss": 5.2589,
      "mean_token_accuracy": 0.18286144286394118,
      "num_tokens": 37382098.0,
      "step": 20270
    },
    {
      "entropy": 5.5857306003570555,
      "epoch": 1.7033816425120771,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004710307514867319,
      "loss": 5.263,
      "mean_token_accuracy": 0.1771240010857582,
      "num_tokens": 37390844.0,
      "step": 20275
    },
    {
      "entropy": 5.653777027130127,
      "epoch": 1.7038017223272421,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004710160506827871,
      "loss": 5.325,
      "mean_token_accuracy": 0.1691286474466324,
      "num_tokens": 37399617.0,
      "step": 20280
    },
    {
      "entropy": 5.630895280838013,
      "epoch": 1.7042218021424071,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004710013464065189,
      "loss": 5.4574,
      "mean_token_accuracy": 0.17007494047284127,
      "num_tokens": 37409368.0,
      "step": 20285
    },
    {
      "entropy": 5.594287967681884,
      "epoch": 1.704641881957572,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004709866386581877,
      "loss": 5.2478,
      "mean_token_accuracy": 0.18386808186769485,
      "num_tokens": 37418026.0,
      "step": 20290
    },
    {
      "entropy": 5.607952642440796,
      "epoch": 1.7050619617727367,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047097192743805413,
      "loss": 5.2828,
      "mean_token_accuracy": 0.1762753903865814,
      "num_tokens": 37426850.0,
      "step": 20295
    },
    {
      "entropy": 5.607682514190674,
      "epoch": 1.7054820415879017,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004709572127463788,
      "loss": 5.3177,
      "mean_token_accuracy": 0.1777769386768341,
      "num_tokens": 37436631.0,
      "step": 20300
    },
    {
      "entropy": 5.664573335647583,
      "epoch": 1.7059021214030667,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004709424945834223,
      "loss": 5.3399,
      "mean_token_accuracy": 0.17262666672468185,
      "num_tokens": 37445619.0,
      "step": 20305
    },
    {
      "entropy": 5.6053167343139645,
      "epoch": 1.7063222012182315,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047092777294944544,
      "loss": 5.3033,
      "mean_token_accuracy": 0.1732635587453842,
      "num_tokens": 37454205.0,
      "step": 20310
    },
    {
      "entropy": 5.5912878036499025,
      "epoch": 1.7067422810333963,
      "grad_norm": 1.234375,
      "learning_rate": 0.000470913047844709,
      "loss": 5.3962,
      "mean_token_accuracy": 0.16958887726068497,
      "num_tokens": 37463301.0,
      "step": 20315
    },
    {
      "entropy": 5.6892424583435055,
      "epoch": 1.707162360848561,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047089831926947374,
      "loss": 5.3877,
      "mean_token_accuracy": 0.1768128365278244,
      "num_tokens": 37471937.0,
      "step": 20320
    },
    {
      "entropy": 5.712940406799317,
      "epoch": 1.707582440663726,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004708835872240007,
      "loss": 5.3514,
      "mean_token_accuracy": 0.18018676340579987,
      "num_tokens": 37480779.0,
      "step": 20325
    },
    {
      "entropy": 5.672598600387573,
      "epoch": 1.7080025204788911,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00047086885170855074,
      "loss": 5.4039,
      "mean_token_accuracy": 0.16074793189764022,
      "num_tokens": 37491053.0,
      "step": 20330
    },
    {
      "entropy": 5.683733749389648,
      "epoch": 1.708422600294056,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000470854112723385,
      "loss": 5.3303,
      "mean_token_accuracy": 0.17568230628967285,
      "num_tokens": 37499091.0,
      "step": 20335
    },
    {
      "entropy": 5.6149462223052975,
      "epoch": 1.7088426801092207,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004708393702687644,
      "loss": 5.4012,
      "mean_token_accuracy": 0.16906957775354386,
      "num_tokens": 37507882.0,
      "step": 20340
    },
    {
      "entropy": 5.639767408370972,
      "epoch": 1.7092627599243855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047082462434495015,
      "loss": 5.3617,
      "mean_token_accuracy": 0.17350460588932037,
      "num_tokens": 37517048.0,
      "step": 20345
    },
    {
      "entropy": 5.690087032318115,
      "epoch": 1.7096828397395505,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004708098749522036,
      "loss": 5.4043,
      "mean_token_accuracy": 0.16676706671714783,
      "num_tokens": 37526355.0,
      "step": 20350
    },
    {
      "entropy": 5.672850656509399,
      "epoch": 1.7101029195547155,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004707951220907859,
      "loss": 5.427,
      "mean_token_accuracy": 0.1674654021859169,
      "num_tokens": 37535746.0,
      "step": 20355
    },
    {
      "entropy": 5.677463054656982,
      "epoch": 1.7105229993698803,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004707803657609585,
      "loss": 5.4048,
      "mean_token_accuracy": 0.16452567875385285,
      "num_tokens": 37546479.0,
      "step": 20360
    },
    {
      "entropy": 5.795746326446533,
      "epoch": 1.710943079185045,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047076560596298275,
      "loss": 5.4484,
      "mean_token_accuracy": 0.16925242766737938,
      "num_tokens": 37556805.0,
      "step": 20365
    },
    {
      "entropy": 5.735682010650635,
      "epoch": 1.71136315900021,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047075084269712,
      "loss": 5.4212,
      "mean_token_accuracy": 0.16944229751825332,
      "num_tokens": 37564748.0,
      "step": 20370
    },
    {
      "entropy": 5.592085313796997,
      "epoch": 1.711783238815375,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004707360759636319,
      "loss": 5.2322,
      "mean_token_accuracy": 0.1801615223288536,
      "num_tokens": 37574674.0,
      "step": 20375
    },
    {
      "entropy": 5.61457405090332,
      "epoch": 1.7122033186305399,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047072130576278,
      "loss": 5.3522,
      "mean_token_accuracy": 0.17090697586536407,
      "num_tokens": 37584459.0,
      "step": 20380
    },
    {
      "entropy": 5.69390230178833,
      "epoch": 1.7126233984457047,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004707065320948259,
      "loss": 5.3787,
      "mean_token_accuracy": 0.1731954336166382,
      "num_tokens": 37593570.0,
      "step": 20385
    },
    {
      "entropy": 5.645740938186646,
      "epoch": 1.7130434782608694,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047069175496003147,
      "loss": 5.3931,
      "mean_token_accuracy": 0.16824432015419005,
      "num_tokens": 37603032.0,
      "step": 20390
    },
    {
      "entropy": 5.580032157897949,
      "epoch": 1.7134635580760345,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004706769743586583,
      "loss": 5.3163,
      "mean_token_accuracy": 0.17552683204412461,
      "num_tokens": 37612404.0,
      "step": 20395
    },
    {
      "entropy": 5.654459714889526,
      "epoch": 1.7138836378911995,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047066219029096837,
      "loss": 5.337,
      "mean_token_accuracy": 0.17208801954984665,
      "num_tokens": 37621933.0,
      "step": 20400
    },
    {
      "entropy": 5.767523860931396,
      "epoch": 1.7143037177063642,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004706474027572234,
      "loss": 5.3707,
      "mean_token_accuracy": 0.17056208103895187,
      "num_tokens": 37632078.0,
      "step": 20405
    },
    {
      "entropy": 5.548586750030518,
      "epoch": 1.714723797521529,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047063261175768543,
      "loss": 5.2788,
      "mean_token_accuracy": 0.17406804412603377,
      "num_tokens": 37641665.0,
      "step": 20410
    },
    {
      "entropy": 5.6630007266998295,
      "epoch": 1.7151438773366938,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047061781729261656,
      "loss": 5.3519,
      "mean_token_accuracy": 0.1686239406466484,
      "num_tokens": 37650751.0,
      "step": 20415
    },
    {
      "entropy": 5.609832811355591,
      "epoch": 1.7155639571518588,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047060301936227865,
      "loss": 5.3353,
      "mean_token_accuracy": 0.17873973846435548,
      "num_tokens": 37659165.0,
      "step": 20420
    },
    {
      "entropy": 5.6234392642974855,
      "epoch": 1.7159840369670238,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004705882179669341,
      "loss": 5.3482,
      "mean_token_accuracy": 0.17245182543992996,
      "num_tokens": 37668057.0,
      "step": 20425
    },
    {
      "entropy": 5.678987598419189,
      "epoch": 1.7164041167821886,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004705734131068449,
      "loss": 5.3144,
      "mean_token_accuracy": 0.17199259847402573,
      "num_tokens": 37677674.0,
      "step": 20430
    },
    {
      "entropy": 5.5998069763183596,
      "epoch": 1.7168241965973534,
      "grad_norm": 1.5,
      "learning_rate": 0.0004705586047822734,
      "loss": 5.3624,
      "mean_token_accuracy": 0.17169613242149354,
      "num_tokens": 37687009.0,
      "step": 20435
    },
    {
      "entropy": 5.544752883911133,
      "epoch": 1.7172442764125184,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047054379299348194,
      "loss": 5.2178,
      "mean_token_accuracy": 0.17698446959257125,
      "num_tokens": 37696723.0,
      "step": 20440
    },
    {
      "entropy": 5.564749336242675,
      "epoch": 1.7176643562276832,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047052897774073295,
      "loss": 5.3366,
      "mean_token_accuracy": 0.17162106037139893,
      "num_tokens": 37706560.0,
      "step": 20445
    },
    {
      "entropy": 5.641170263290405,
      "epoch": 1.7180844360428482,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047051415902428875,
      "loss": 5.3508,
      "mean_token_accuracy": 0.1743730679154396,
      "num_tokens": 37715176.0,
      "step": 20450
    },
    {
      "entropy": 5.6161217212677,
      "epoch": 1.718504515858013,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004704993368444119,
      "loss": 5.3426,
      "mean_token_accuracy": 0.17417691200971602,
      "num_tokens": 37723956.0,
      "step": 20455
    },
    {
      "entropy": 5.639037036895752,
      "epoch": 1.7189245956731778,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004704845112013649,
      "loss": 5.3594,
      "mean_token_accuracy": 0.1704057276248932,
      "num_tokens": 37733236.0,
      "step": 20460
    },
    {
      "entropy": 5.6859582424163815,
      "epoch": 1.7193446754883428,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004704696820954105,
      "loss": 5.4025,
      "mean_token_accuracy": 0.1734488546848297,
      "num_tokens": 37742626.0,
      "step": 20465
    },
    {
      "entropy": 5.622436666488648,
      "epoch": 1.7197647553035078,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004704548495268113,
      "loss": 5.2902,
      "mean_token_accuracy": 0.18298593759536744,
      "num_tokens": 37751854.0,
      "step": 20470
    },
    {
      "entropy": 5.613380479812622,
      "epoch": 1.7201848351186726,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047044001349583,
      "loss": 5.3213,
      "mean_token_accuracy": 0.17561782002449036,
      "num_tokens": 37760993.0,
      "step": 20475
    },
    {
      "entropy": 5.6216028213500975,
      "epoch": 1.7206049149338374,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047042517400272966,
      "loss": 5.4165,
      "mean_token_accuracy": 0.1757997006177902,
      "num_tokens": 37771714.0,
      "step": 20480
    },
    {
      "entropy": 5.634616899490356,
      "epoch": 1.7210249947490022,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004704103310477729,
      "loss": 5.3293,
      "mean_token_accuracy": 0.1765611633658409,
      "num_tokens": 37780653.0,
      "step": 20485
    },
    {
      "entropy": 5.643177795410156,
      "epoch": 1.7214450745641672,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004703954846312228,
      "loss": 5.4034,
      "mean_token_accuracy": 0.17122139483690263,
      "num_tokens": 37790450.0,
      "step": 20490
    },
    {
      "entropy": 5.647590970993042,
      "epoch": 1.7218651543793322,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004703806347533423,
      "loss": 5.3909,
      "mean_token_accuracy": 0.1686463326215744,
      "num_tokens": 37800450.0,
      "step": 20495
    },
    {
      "entropy": 5.6734757900238035,
      "epoch": 1.722285234194497,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004703657814143945,
      "loss": 5.4133,
      "mean_token_accuracy": 0.16684700548648834,
      "num_tokens": 37809261.0,
      "step": 20500
    },
    {
      "entropy": 5.624744176864624,
      "epoch": 1.7227053140096618,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004703509246146424,
      "loss": 5.1896,
      "mean_token_accuracy": 0.17729185074567794,
      "num_tokens": 37818244.0,
      "step": 20505
    },
    {
      "entropy": 5.63604965209961,
      "epoch": 1.7231253938248268,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004703360643543493,
      "loss": 5.3173,
      "mean_token_accuracy": 0.17956820279359817,
      "num_tokens": 37828555.0,
      "step": 20510
    },
    {
      "entropy": 5.557352924346924,
      "epoch": 1.7235454736399916,
      "grad_norm": 1.734375,
      "learning_rate": 0.00047032120063377836,
      "loss": 5.2856,
      "mean_token_accuracy": 0.17612000703811645,
      "num_tokens": 37837840.0,
      "step": 20515
    },
    {
      "entropy": 5.625510835647583,
      "epoch": 1.7239655534551566,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047030633345319293,
      "loss": 5.3394,
      "mean_token_accuracy": 0.16890984773635864,
      "num_tokens": 37846910.0,
      "step": 20520
    },
    {
      "entropy": 5.499063491821289,
      "epoch": 1.7243856332703213,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047029146281285647,
      "loss": 5.167,
      "mean_token_accuracy": 0.1924958810210228,
      "num_tokens": 37855642.0,
      "step": 20525
    },
    {
      "entropy": 5.634795427322388,
      "epoch": 1.7248057130854861,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004702765887130322,
      "loss": 5.3514,
      "mean_token_accuracy": 0.17644862979650497,
      "num_tokens": 37864439.0,
      "step": 20530
    },
    {
      "entropy": 5.725802183151245,
      "epoch": 1.7252257929006511,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047026171115398377,
      "loss": 5.4247,
      "mean_token_accuracy": 0.1651463858783245,
      "num_tokens": 37873801.0,
      "step": 20535
    },
    {
      "entropy": 5.629425859451294,
      "epoch": 1.7256458727158162,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004702468301359746,
      "loss": 5.3034,
      "mean_token_accuracy": 0.18187215328216552,
      "num_tokens": 37883915.0,
      "step": 20540
    },
    {
      "entropy": 5.606316661834716,
      "epoch": 1.726065952530981,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004702319456592684,
      "loss": 5.4158,
      "mean_token_accuracy": 0.16988707780838014,
      "num_tokens": 37894083.0,
      "step": 20545
    },
    {
      "entropy": 5.671419095993042,
      "epoch": 1.7264860323461457,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047021705772412885,
      "loss": 5.4071,
      "mean_token_accuracy": 0.1727289840579033,
      "num_tokens": 37902264.0,
      "step": 20550
    },
    {
      "entropy": 5.630827617645264,
      "epoch": 1.7269061121613105,
      "grad_norm": 1.671875,
      "learning_rate": 0.00047020216633081964,
      "loss": 5.3187,
      "mean_token_accuracy": 0.17681977152824402,
      "num_tokens": 37911071.0,
      "step": 20555
    },
    {
      "entropy": 5.601305770874023,
      "epoch": 1.7273261919764755,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047018727147960453,
      "loss": 5.4065,
      "mean_token_accuracy": 0.1700693890452385,
      "num_tokens": 37920048.0,
      "step": 20560
    },
    {
      "entropy": 5.679267406463623,
      "epoch": 1.7277462717916405,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047017237317074743,
      "loss": 5.3555,
      "mean_token_accuracy": 0.1737640991806984,
      "num_tokens": 37928877.0,
      "step": 20565
    },
    {
      "entropy": 5.683482599258423,
      "epoch": 1.7281663516068053,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004701574714045123,
      "loss": 5.3789,
      "mean_token_accuracy": 0.16272292286157608,
      "num_tokens": 37937860.0,
      "step": 20570
    },
    {
      "entropy": 5.6335663318634035,
      "epoch": 1.72858643142197,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047014256618116304,
      "loss": 5.3882,
      "mean_token_accuracy": 0.16922035366296767,
      "num_tokens": 37947588.0,
      "step": 20575
    },
    {
      "entropy": 5.668964958190918,
      "epoch": 1.729006511237135,
      "grad_norm": 1.625,
      "learning_rate": 0.00047012765750096365,
      "loss": 5.3091,
      "mean_token_accuracy": 0.180650931596756,
      "num_tokens": 37957598.0,
      "step": 20580
    },
    {
      "entropy": 5.647602081298828,
      "epoch": 1.7294265910523,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047011274536417827,
      "loss": 5.287,
      "mean_token_accuracy": 0.18199337273836136,
      "num_tokens": 37965294.0,
      "step": 20585
    },
    {
      "entropy": 5.555703783035279,
      "epoch": 1.729846670867465,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047009782977107113,
      "loss": 5.3001,
      "mean_token_accuracy": 0.18158977925777436,
      "num_tokens": 37973977.0,
      "step": 20590
    },
    {
      "entropy": 5.75575270652771,
      "epoch": 1.7302667506826297,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047008291072190634,
      "loss": 5.4563,
      "mean_token_accuracy": 0.16498687267303466,
      "num_tokens": 37984492.0,
      "step": 20595
    },
    {
      "entropy": 5.735362529754639,
      "epoch": 1.7306868304977945,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004700679882169482,
      "loss": 5.3767,
      "mean_token_accuracy": 0.17322198301553726,
      "num_tokens": 37994404.0,
      "step": 20600
    },
    {
      "entropy": 5.527033042907715,
      "epoch": 1.7311069103129595,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004700530622564613,
      "loss": 5.2759,
      "mean_token_accuracy": 0.18220295011997223,
      "num_tokens": 38002659.0,
      "step": 20605
    },
    {
      "entropy": 5.568998527526856,
      "epoch": 1.7315269901281245,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004700381328407096,
      "loss": 5.2691,
      "mean_token_accuracy": 0.17584830820560454,
      "num_tokens": 38012290.0,
      "step": 20610
    },
    {
      "entropy": 5.694201183319092,
      "epoch": 1.7319470699432893,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004700231999699579,
      "loss": 5.3926,
      "mean_token_accuracy": 0.16920189261436464,
      "num_tokens": 38022163.0,
      "step": 20615
    },
    {
      "entropy": 5.633602952957153,
      "epoch": 1.732367149758454,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004700082636444706,
      "loss": 5.3287,
      "mean_token_accuracy": 0.16998203992843627,
      "num_tokens": 38031051.0,
      "step": 20620
    },
    {
      "entropy": 5.668176317214966,
      "epoch": 1.7327872295736189,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046999332386451245,
      "loss": 5.3945,
      "mean_token_accuracy": 0.17070767879486085,
      "num_tokens": 38040474.0,
      "step": 20625
    },
    {
      "entropy": 5.652907752990723,
      "epoch": 1.7332073093887839,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046997838063034784,
      "loss": 5.3608,
      "mean_token_accuracy": 0.1725230485200882,
      "num_tokens": 38049620.0,
      "step": 20630
    },
    {
      "entropy": 5.579212188720703,
      "epoch": 1.7336273892039489,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046996343394224173,
      "loss": 5.3199,
      "mean_token_accuracy": 0.17720960527658464,
      "num_tokens": 38059866.0,
      "step": 20635
    },
    {
      "entropy": 5.629855394363403,
      "epoch": 1.7340474690191137,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046994848380045866,
      "loss": 5.2989,
      "mean_token_accuracy": 0.17243426740169526,
      "num_tokens": 38068948.0,
      "step": 20640
    },
    {
      "entropy": 5.716435050964355,
      "epoch": 1.7344675488342784,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046993353020526366,
      "loss": 5.4902,
      "mean_token_accuracy": 0.16873880922794343,
      "num_tokens": 38079239.0,
      "step": 20645
    },
    {
      "entropy": 5.629824686050415,
      "epoch": 1.7348876286494432,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004699185731569215,
      "loss": 5.3585,
      "mean_token_accuracy": 0.17093686908483505,
      "num_tokens": 38087999.0,
      "step": 20650
    },
    {
      "entropy": 5.645922517776489,
      "epoch": 1.7353077084646082,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004699036126556972,
      "loss": 5.3398,
      "mean_token_accuracy": 0.16926045417785646,
      "num_tokens": 38096586.0,
      "step": 20655
    },
    {
      "entropy": 5.570487451553345,
      "epoch": 1.7357277882797733,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004698886487018558,
      "loss": 5.3078,
      "mean_token_accuracy": 0.1718907669186592,
      "num_tokens": 38104766.0,
      "step": 20660
    },
    {
      "entropy": 5.624381494522095,
      "epoch": 1.736147868094938,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004698736812956623,
      "loss": 5.3538,
      "mean_token_accuracy": 0.1738964796066284,
      "num_tokens": 38113574.0,
      "step": 20665
    },
    {
      "entropy": 5.653391695022583,
      "epoch": 1.7365679479101028,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004698587104373819,
      "loss": 5.2924,
      "mean_token_accuracy": 0.1672384649515152,
      "num_tokens": 38122513.0,
      "step": 20670
    },
    {
      "entropy": 5.525764417648316,
      "epoch": 1.7369880277252678,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046984373612727975,
      "loss": 5.2858,
      "mean_token_accuracy": 0.16313301622867585,
      "num_tokens": 38131105.0,
      "step": 20675
    },
    {
      "entropy": 5.602736520767212,
      "epoch": 1.7374081075404328,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046982875836562116,
      "loss": 5.3696,
      "mean_token_accuracy": 0.16137362718582154,
      "num_tokens": 38140106.0,
      "step": 20680
    },
    {
      "entropy": 5.689923334121704,
      "epoch": 1.7378281873555976,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046981377715267145,
      "loss": 5.3196,
      "mean_token_accuracy": 0.18165768086910247,
      "num_tokens": 38149215.0,
      "step": 20685
    },
    {
      "entropy": 5.619599199295044,
      "epoch": 1.7382482671707624,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000469798792488696,
      "loss": 5.2451,
      "mean_token_accuracy": 0.1791176289319992,
      "num_tokens": 38157591.0,
      "step": 20690
    },
    {
      "entropy": 5.5595197677612305,
      "epoch": 1.7386683469859272,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004697838043739602,
      "loss": 5.3683,
      "mean_token_accuracy": 0.16964025050401688,
      "num_tokens": 38167673.0,
      "step": 20695
    },
    {
      "entropy": 5.68686580657959,
      "epoch": 1.7390884268010922,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046976881280872974,
      "loss": 5.3417,
      "mean_token_accuracy": 0.17233023941516876,
      "num_tokens": 38177586.0,
      "step": 20700
    },
    {
      "entropy": 5.704075765609741,
      "epoch": 1.7395085066162572,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004697538177932699,
      "loss": 5.3421,
      "mean_token_accuracy": 0.1721431978046894,
      "num_tokens": 38187020.0,
      "step": 20705
    },
    {
      "entropy": 5.48824462890625,
      "epoch": 1.739928586431422,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004697388193278465,
      "loss": 5.1214,
      "mean_token_accuracy": 0.1863224521279335,
      "num_tokens": 38195705.0,
      "step": 20710
    },
    {
      "entropy": 5.517525148391724,
      "epoch": 1.7403486662465868,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004697238174127252,
      "loss": 5.236,
      "mean_token_accuracy": 0.17681074738502503,
      "num_tokens": 38204726.0,
      "step": 20715
    },
    {
      "entropy": 5.55948281288147,
      "epoch": 1.7407687460617516,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004697088120481717,
      "loss": 5.3709,
      "mean_token_accuracy": 0.1725475013256073,
      "num_tokens": 38214376.0,
      "step": 20720
    },
    {
      "entropy": 5.583048248291016,
      "epoch": 1.7411888258769166,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004696938032344519,
      "loss": 5.2713,
      "mean_token_accuracy": 0.17500107288360595,
      "num_tokens": 38223631.0,
      "step": 20725
    },
    {
      "entropy": 5.663219451904297,
      "epoch": 1.7416089056920816,
      "grad_norm": 1.25,
      "learning_rate": 0.0004696787909718317,
      "loss": 5.3046,
      "mean_token_accuracy": 0.18049535751342774,
      "num_tokens": 38233519.0,
      "step": 20730
    },
    {
      "entropy": 5.645912742614746,
      "epoch": 1.7420289855072464,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046966377526057686,
      "loss": 5.2781,
      "mean_token_accuracy": 0.1814132586121559,
      "num_tokens": 38242340.0,
      "step": 20735
    },
    {
      "entropy": 5.509473371505737,
      "epoch": 1.7424490653224112,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004696487561009535,
      "loss": 5.2639,
      "mean_token_accuracy": 0.17726294696331024,
      "num_tokens": 38251194.0,
      "step": 20740
    },
    {
      "entropy": 5.616220140457154,
      "epoch": 1.7428691451375762,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004696337334932277,
      "loss": 5.3227,
      "mean_token_accuracy": 0.1701137274503708,
      "num_tokens": 38259938.0,
      "step": 20745
    },
    {
      "entropy": 5.644875335693359,
      "epoch": 1.743289224952741,
      "grad_norm": 1.609375,
      "learning_rate": 0.00046961870743766546,
      "loss": 5.3258,
      "mean_token_accuracy": 0.17122088372707367,
      "num_tokens": 38268073.0,
      "step": 20750
    },
    {
      "entropy": 5.643024730682373,
      "epoch": 1.743709304767906,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046960367793453313,
      "loss": 5.4335,
      "mean_token_accuracy": 0.16416030675172805,
      "num_tokens": 38277667.0,
      "step": 20755
    },
    {
      "entropy": 5.734397554397583,
      "epoch": 1.7441293845830708,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046958864498409673,
      "loss": 5.3732,
      "mean_token_accuracy": 0.17459494173526763,
      "num_tokens": 38287142.0,
      "step": 20760
    },
    {
      "entropy": 5.67715163230896,
      "epoch": 1.7445494643982355,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046957360858662276,
      "loss": 5.3551,
      "mean_token_accuracy": 0.1794022336602211,
      "num_tokens": 38296199.0,
      "step": 20765
    },
    {
      "entropy": 5.6129755020141605,
      "epoch": 1.7449695442134006,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004695585687423775,
      "loss": 5.3584,
      "mean_token_accuracy": 0.17329048216342927,
      "num_tokens": 38305412.0,
      "step": 20770
    },
    {
      "entropy": 5.579747486114502,
      "epoch": 1.7453896240285656,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004695435254516273,
      "loss": 5.2848,
      "mean_token_accuracy": 0.1836205914616585,
      "num_tokens": 38313898.0,
      "step": 20775
    },
    {
      "entropy": 5.706262493133545,
      "epoch": 1.7458097038437304,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004695284787146388,
      "loss": 5.4057,
      "mean_token_accuracy": 0.16887173503637315,
      "num_tokens": 38322835.0,
      "step": 20780
    },
    {
      "entropy": 5.560011625289917,
      "epoch": 1.7462297836588951,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004695134285316784,
      "loss": 5.2277,
      "mean_token_accuracy": 0.18544360846281052,
      "num_tokens": 38331448.0,
      "step": 20785
    },
    {
      "entropy": 5.576202297210694,
      "epoch": 1.74664986347406,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046949837490301293,
      "loss": 5.3702,
      "mean_token_accuracy": 0.16890094429254532,
      "num_tokens": 38340837.0,
      "step": 20790
    },
    {
      "entropy": 5.622522449493408,
      "epoch": 1.747069943289225,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004694833178289088,
      "loss": 5.3047,
      "mean_token_accuracy": 0.17854925841093064,
      "num_tokens": 38349363.0,
      "step": 20795
    },
    {
      "entropy": 5.615919017791748,
      "epoch": 1.74749002310439,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004694682573096328,
      "loss": 5.3475,
      "mean_token_accuracy": 0.17742337733507157,
      "num_tokens": 38358017.0,
      "step": 20800
    },
    {
      "entropy": 5.634621286392212,
      "epoch": 1.7479101029195547,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046945319334545184,
      "loss": 5.3448,
      "mean_token_accuracy": 0.17155629098415376,
      "num_tokens": 38367256.0,
      "step": 20805
    },
    {
      "entropy": 5.615961217880249,
      "epoch": 1.7483301827347195,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004694381259366327,
      "loss": 5.3323,
      "mean_token_accuracy": 0.17451196014881135,
      "num_tokens": 38376169.0,
      "step": 20810
    },
    {
      "entropy": 5.65543155670166,
      "epoch": 1.7487502625498845,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046942305508344216,
      "loss": 5.3162,
      "mean_token_accuracy": 0.1746650904417038,
      "num_tokens": 38385379.0,
      "step": 20815
    },
    {
      "entropy": 5.698629426956177,
      "epoch": 1.7491703423650493,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004694079807861473,
      "loss": 5.3944,
      "mean_token_accuracy": 0.17143232375383377,
      "num_tokens": 38395217.0,
      "step": 20820
    },
    {
      "entropy": 5.601200771331787,
      "epoch": 1.7495904221802143,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004693929030450153,
      "loss": 5.3033,
      "mean_token_accuracy": 0.17674324959516524,
      "num_tokens": 38404347.0,
      "step": 20825
    },
    {
      "entropy": 5.650736331939697,
      "epoch": 1.750010501995379,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046937782186031303,
      "loss": 5.2698,
      "mean_token_accuracy": 0.17850795686244963,
      "num_tokens": 38413394.0,
      "step": 20830
    },
    {
      "entropy": 5.603127813339233,
      "epoch": 1.750430581810544,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004693627372323078,
      "loss": 5.3252,
      "mean_token_accuracy": 0.1711818739771843,
      "num_tokens": 38422043.0,
      "step": 20835
    },
    {
      "entropy": 5.684387111663819,
      "epoch": 1.750850661625709,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004693476491612667,
      "loss": 5.4741,
      "mean_token_accuracy": 0.16774453818798066,
      "num_tokens": 38430792.0,
      "step": 20840
    },
    {
      "entropy": 5.571090030670166,
      "epoch": 1.751270741440874,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004693325576474571,
      "loss": 5.2774,
      "mean_token_accuracy": 0.18051262348890304,
      "num_tokens": 38439105.0,
      "step": 20845
    },
    {
      "entropy": 5.661008882522583,
      "epoch": 1.7516908212560387,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004693174626911463,
      "loss": 5.3053,
      "mean_token_accuracy": 0.1730935499072075,
      "num_tokens": 38447944.0,
      "step": 20850
    },
    {
      "entropy": 5.609008836746216,
      "epoch": 1.7521109010712035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046930236429260173,
      "loss": 5.3487,
      "mean_token_accuracy": 0.1690741553902626,
      "num_tokens": 38457206.0,
      "step": 20855
    },
    {
      "entropy": 5.664777374267578,
      "epoch": 1.7525309808863683,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004692872624520908,
      "loss": 5.4134,
      "mean_token_accuracy": 0.1652311399579048,
      "num_tokens": 38467085.0,
      "step": 20860
    },
    {
      "entropy": 5.659315347671509,
      "epoch": 1.7529510607015333,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000469272157169881,
      "loss": 5.2578,
      "mean_token_accuracy": 0.17767437547445297,
      "num_tokens": 38475970.0,
      "step": 20865
    },
    {
      "entropy": 5.623927021026612,
      "epoch": 1.7533711405166983,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004692570484462401,
      "loss": 5.4005,
      "mean_token_accuracy": 0.17276509553194047,
      "num_tokens": 38484579.0,
      "step": 20870
    },
    {
      "entropy": 5.64286847114563,
      "epoch": 1.753791220331863,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046924193628143554,
      "loss": 5.4244,
      "mean_token_accuracy": 0.16616735756397247,
      "num_tokens": 38495107.0,
      "step": 20875
    },
    {
      "entropy": 5.7381328582763675,
      "epoch": 1.7542113001470279,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046922682067573516,
      "loss": 5.4399,
      "mean_token_accuracy": 0.17004961520433426,
      "num_tokens": 38505731.0,
      "step": 20880
    },
    {
      "entropy": 5.587978792190552,
      "epoch": 1.7546313799621929,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046921170162940657,
      "loss": 5.3282,
      "mean_token_accuracy": 0.17631073594093322,
      "num_tokens": 38514483.0,
      "step": 20885
    },
    {
      "entropy": 5.567329597473145,
      "epoch": 1.7550514597773577,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046919657914271774,
      "loss": 5.2284,
      "mean_token_accuracy": 0.18373051434755325,
      "num_tokens": 38522953.0,
      "step": 20890
    },
    {
      "entropy": 5.567907810211182,
      "epoch": 1.7554715395925227,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004691814532159365,
      "loss": 5.2285,
      "mean_token_accuracy": 0.1867031767964363,
      "num_tokens": 38531891.0,
      "step": 20895
    },
    {
      "entropy": 5.643687200546265,
      "epoch": 1.7558916194076875,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004691663238493308,
      "loss": 5.4085,
      "mean_token_accuracy": 0.17143741101026536,
      "num_tokens": 38541609.0,
      "step": 20900
    },
    {
      "entropy": 5.684701776504516,
      "epoch": 1.7563116992228522,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004691511910431686,
      "loss": 5.4022,
      "mean_token_accuracy": 0.1700277827680111,
      "num_tokens": 38550348.0,
      "step": 20905
    },
    {
      "entropy": 5.5740283012390135,
      "epoch": 1.7567317790380172,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004691360547977181,
      "loss": 5.2499,
      "mean_token_accuracy": 0.1790262922644615,
      "num_tokens": 38559493.0,
      "step": 20910
    },
    {
      "entropy": 5.547926139831543,
      "epoch": 1.7571518588531823,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004691209151132474,
      "loss": 5.2946,
      "mean_token_accuracy": 0.1681540161371231,
      "num_tokens": 38567888.0,
      "step": 20915
    },
    {
      "entropy": 5.647246170043945,
      "epoch": 1.757571938668347,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004691057719900246,
      "loss": 5.3456,
      "mean_token_accuracy": 0.17491035610437394,
      "num_tokens": 38577216.0,
      "step": 20920
    },
    {
      "entropy": 5.589968824386597,
      "epoch": 1.7579920184835118,
      "grad_norm": 1.375,
      "learning_rate": 0.00046909062542831794,
      "loss": 5.3244,
      "mean_token_accuracy": 0.17533470690250397,
      "num_tokens": 38586258.0,
      "step": 20925
    },
    {
      "entropy": 5.602022790908814,
      "epoch": 1.7584120982986766,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004690754754283959,
      "loss": 5.2534,
      "mean_token_accuracy": 0.17878633588552476,
      "num_tokens": 38594900.0,
      "step": 20930
    },
    {
      "entropy": 5.513794040679931,
      "epoch": 1.7588321781138416,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004690603219905266,
      "loss": 5.3423,
      "mean_token_accuracy": 0.17404213398694993,
      "num_tokens": 38603980.0,
      "step": 20935
    },
    {
      "entropy": 5.661549615859985,
      "epoch": 1.7592522579290066,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046904516511497873,
      "loss": 5.4394,
      "mean_token_accuracy": 0.16892625391483307,
      "num_tokens": 38613888.0,
      "step": 20940
    },
    {
      "entropy": 5.743622303009033,
      "epoch": 1.7596723377441714,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046903000480202065,
      "loss": 5.3715,
      "mean_token_accuracy": 0.17235313206911088,
      "num_tokens": 38623969.0,
      "step": 20945
    },
    {
      "entropy": 5.595367574691773,
      "epoch": 1.7600924175593362,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046901484105192094,
      "loss": 5.3169,
      "mean_token_accuracy": 0.17539885640144348,
      "num_tokens": 38633387.0,
      "step": 20950
    },
    {
      "entropy": 5.6161657810211185,
      "epoch": 1.760512497374501,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046899967386494816,
      "loss": 5.3888,
      "mean_token_accuracy": 0.168683260679245,
      "num_tokens": 38642481.0,
      "step": 20955
    },
    {
      "entropy": 5.679697847366333,
      "epoch": 1.760932577189666,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004689845032413712,
      "loss": 5.3756,
      "mean_token_accuracy": 0.16590355485677719,
      "num_tokens": 38652345.0,
      "step": 20960
    },
    {
      "entropy": 5.738378238677979,
      "epoch": 1.761352657004831,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004689693291814586,
      "loss": 5.4029,
      "mean_token_accuracy": 0.16636169701814651,
      "num_tokens": 38661529.0,
      "step": 20965
    },
    {
      "entropy": 5.614858245849609,
      "epoch": 1.7617727368199958,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004689541516854791,
      "loss": 5.2921,
      "mean_token_accuracy": 0.17880262881517411,
      "num_tokens": 38670191.0,
      "step": 20970
    },
    {
      "entropy": 5.576412343978882,
      "epoch": 1.7621928166351606,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004689389707537018,
      "loss": 5.3805,
      "mean_token_accuracy": 0.17319887578487397,
      "num_tokens": 38679089.0,
      "step": 20975
    },
    {
      "entropy": 5.645511341094971,
      "epoch": 1.7626128964503256,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046892378638639545,
      "loss": 5.3239,
      "mean_token_accuracy": 0.17503227889537812,
      "num_tokens": 38688821.0,
      "step": 20980
    },
    {
      "entropy": 5.686195182800293,
      "epoch": 1.7630329762654906,
      "grad_norm": 1.796875,
      "learning_rate": 0.00046890859858382913,
      "loss": 5.399,
      "mean_token_accuracy": 0.16521373838186265,
      "num_tokens": 38698232.0,
      "step": 20985
    },
    {
      "entropy": 5.711312294006348,
      "epoch": 1.7634530560806554,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004688934073462718,
      "loss": 5.4625,
      "mean_token_accuracy": 0.16444431319832803,
      "num_tokens": 38708090.0,
      "step": 20990
    },
    {
      "entropy": 5.64181432723999,
      "epoch": 1.7638731358958202,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046887821267399256,
      "loss": 5.3718,
      "mean_token_accuracy": 0.17506936937570572,
      "num_tokens": 38717370.0,
      "step": 20995
    },
    {
      "entropy": 5.650666809082031,
      "epoch": 1.764293215710985,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004688630145672607,
      "loss": 5.3444,
      "mean_token_accuracy": 0.17406408339738846,
      "num_tokens": 38726758.0,
      "step": 21000
    },
    {
      "epoch": 1.764293215710985,
      "eval_entropy": 5.46679323537782,
      "eval_loss": 5.404355049133301,
      "eval_mean_token_accuracy": 0.17909046986346275,
      "eval_num_tokens": 38726758.0,
      "eval_runtime": 27.4323,
      "eval_samples_per_second": 1362.119,
      "eval_steps_per_second": 170.274,
      "step": 21000
    },
    {
      "entropy": 5.620329236984253,
      "epoch": 1.76471329552615,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004688478130263453,
      "loss": 5.3306,
      "mean_token_accuracy": 0.1670632243156433,
      "num_tokens": 38736180.0,
      "step": 21005
    },
    {
      "entropy": 5.636257266998291,
      "epoch": 1.765133375341315,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004688326080515157,
      "loss": 5.2837,
      "mean_token_accuracy": 0.17704454064369202,
      "num_tokens": 38744529.0,
      "step": 21010
    },
    {
      "entropy": 5.511424922943116,
      "epoch": 1.7655534551564798,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046881739964304127,
      "loss": 5.1947,
      "mean_token_accuracy": 0.18059405237436293,
      "num_tokens": 38753434.0,
      "step": 21015
    },
    {
      "entropy": 5.588047027587891,
      "epoch": 1.7659735349716446,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046880218780119136,
      "loss": 5.3143,
      "mean_token_accuracy": 0.18549560904502868,
      "num_tokens": 38762021.0,
      "step": 21020
    },
    {
      "entropy": 5.67595100402832,
      "epoch": 1.7663936147868093,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004687869725262356,
      "loss": 5.4537,
      "mean_token_accuracy": 0.16397292017936707,
      "num_tokens": 38771373.0,
      "step": 21025
    },
    {
      "entropy": 5.694741201400757,
      "epoch": 1.7668136946019743,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004687717538184433,
      "loss": 5.4012,
      "mean_token_accuracy": 0.17259282767772674,
      "num_tokens": 38780388.0,
      "step": 21030
    },
    {
      "entropy": 5.591110801696777,
      "epoch": 1.7672337744171394,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046875653167808423,
      "loss": 5.2387,
      "mean_token_accuracy": 0.17751833349466323,
      "num_tokens": 38789285.0,
      "step": 21035
    },
    {
      "entropy": 5.52340259552002,
      "epoch": 1.7676538542323041,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046874130610542796,
      "loss": 5.338,
      "mean_token_accuracy": 0.17271726876497268,
      "num_tokens": 38799321.0,
      "step": 21040
    },
    {
      "entropy": 5.698584365844726,
      "epoch": 1.768073934047469,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004687260771007442,
      "loss": 5.3113,
      "mean_token_accuracy": 0.17328591495752335,
      "num_tokens": 38808515.0,
      "step": 21045
    },
    {
      "entropy": 5.662454223632812,
      "epoch": 1.768494013862634,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004687108446643027,
      "loss": 5.2783,
      "mean_token_accuracy": 0.176117005944252,
      "num_tokens": 38817634.0,
      "step": 21050
    },
    {
      "entropy": 5.759761238098145,
      "epoch": 1.7689140936777987,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004686956087963734,
      "loss": 5.4896,
      "mean_token_accuracy": 0.17106397151947023,
      "num_tokens": 38826766.0,
      "step": 21055
    },
    {
      "entropy": 5.564716720581055,
      "epoch": 1.7693341734929637,
      "grad_norm": 1.375,
      "learning_rate": 0.0004686803694972261,
      "loss": 5.2558,
      "mean_token_accuracy": 0.17207407355308532,
      "num_tokens": 38835942.0,
      "step": 21060
    },
    {
      "entropy": 5.64042592048645,
      "epoch": 1.7697542533081285,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046866512676713075,
      "loss": 5.3307,
      "mean_token_accuracy": 0.1687471315264702,
      "num_tokens": 38845691.0,
      "step": 21065
    },
    {
      "entropy": 5.624405384063721,
      "epoch": 1.7701743331232933,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046864988060635744,
      "loss": 5.3448,
      "mean_token_accuracy": 0.1696521073579788,
      "num_tokens": 38855737.0,
      "step": 21070
    },
    {
      "entropy": 5.681068420410156,
      "epoch": 1.7705944129384583,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004686346310151762,
      "loss": 5.3743,
      "mean_token_accuracy": 0.17566830962896346,
      "num_tokens": 38864887.0,
      "step": 21075
    },
    {
      "entropy": 5.708447790145874,
      "epoch": 1.7710144927536233,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046861937799385717,
      "loss": 5.3296,
      "mean_token_accuracy": 0.1722590684890747,
      "num_tokens": 38873924.0,
      "step": 21080
    },
    {
      "entropy": 5.605142498016358,
      "epoch": 1.7714345725687881,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004686041215426706,
      "loss": 5.3856,
      "mean_token_accuracy": 0.17279938757419586,
      "num_tokens": 38883447.0,
      "step": 21085
    },
    {
      "entropy": 5.620349788665772,
      "epoch": 1.771854652383953,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004685888616618867,
      "loss": 5.3558,
      "mean_token_accuracy": 0.17566331326961518,
      "num_tokens": 38892389.0,
      "step": 21090
    },
    {
      "entropy": 5.693254995346069,
      "epoch": 1.7722747321991177,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046857359835177575,
      "loss": 5.405,
      "mean_token_accuracy": 0.16723650842905044,
      "num_tokens": 38901574.0,
      "step": 21095
    },
    {
      "entropy": 5.680951833724976,
      "epoch": 1.7726948120142827,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046855833161260825,
      "loss": 5.3911,
      "mean_token_accuracy": 0.17658820152282714,
      "num_tokens": 38910070.0,
      "step": 21100
    },
    {
      "entropy": 5.605348539352417,
      "epoch": 1.7731148918294477,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004685430614446545,
      "loss": 5.2979,
      "mean_token_accuracy": 0.17724595218896866,
      "num_tokens": 38919868.0,
      "step": 21105
    },
    {
      "entropy": 5.6717053890228275,
      "epoch": 1.7735349716446125,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004685277878481852,
      "loss": 5.3495,
      "mean_token_accuracy": 0.16549432873725892,
      "num_tokens": 38928840.0,
      "step": 21110
    },
    {
      "entropy": 5.654265022277832,
      "epoch": 1.7739550514597773,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046851251082347063,
      "loss": 5.4316,
      "mean_token_accuracy": 0.172701333463192,
      "num_tokens": 38938112.0,
      "step": 21115
    },
    {
      "entropy": 5.676231622695923,
      "epoch": 1.7743751312749423,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004684972303707816,
      "loss": 5.34,
      "mean_token_accuracy": 0.17409232258796692,
      "num_tokens": 38947463.0,
      "step": 21120
    },
    {
      "entropy": 5.7194267272949215,
      "epoch": 1.774795211090107,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004684819464903888,
      "loss": 5.5293,
      "mean_token_accuracy": 0.15862380117177963,
      "num_tokens": 38957221.0,
      "step": 21125
    },
    {
      "entropy": 5.575458955764771,
      "epoch": 1.775215290905272,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000468466659182563,
      "loss": 5.2488,
      "mean_token_accuracy": 0.1777647152543068,
      "num_tokens": 38966656.0,
      "step": 21130
    },
    {
      "entropy": 5.615982246398926,
      "epoch": 1.7756353707204369,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004684513684475749,
      "loss": 5.263,
      "mean_token_accuracy": 0.18193438798189163,
      "num_tokens": 38975281.0,
      "step": 21135
    },
    {
      "entropy": 5.682458353042603,
      "epoch": 1.7760554505356017,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046843607428569546,
      "loss": 5.4001,
      "mean_token_accuracy": 0.17306498885154725,
      "num_tokens": 38985147.0,
      "step": 21140
    },
    {
      "entropy": 5.596554708480835,
      "epoch": 1.7764755303507667,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046842077669719554,
      "loss": 5.1905,
      "mean_token_accuracy": 0.18193431347608566,
      "num_tokens": 38994104.0,
      "step": 21145
    },
    {
      "entropy": 5.5931824207305905,
      "epoch": 1.7768956101659317,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046840547568234613,
      "loss": 5.3703,
      "mean_token_accuracy": 0.1740384057164192,
      "num_tokens": 39003983.0,
      "step": 21150
    },
    {
      "entropy": 5.586139678955078,
      "epoch": 1.7773156899810965,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046839017124141835,
      "loss": 5.2824,
      "mean_token_accuracy": 0.17901071906089783,
      "num_tokens": 39012636.0,
      "step": 21155
    },
    {
      "entropy": 5.648578691482544,
      "epoch": 1.7777357697962612,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046837486337468335,
      "loss": 5.4201,
      "mean_token_accuracy": 0.1673689067363739,
      "num_tokens": 39022173.0,
      "step": 21160
    },
    {
      "entropy": 5.743691349029541,
      "epoch": 1.778155849611426,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000468359552082412,
      "loss": 5.433,
      "mean_token_accuracy": 0.16453042402863502,
      "num_tokens": 39032651.0,
      "step": 21165
    },
    {
      "entropy": 5.660783433914185,
      "epoch": 1.778575929426591,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004683442373648759,
      "loss": 5.329,
      "mean_token_accuracy": 0.17049587965011598,
      "num_tokens": 39041543.0,
      "step": 21170
    },
    {
      "entropy": 5.585572719573975,
      "epoch": 1.778996009241756,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004683289192223462,
      "loss": 5.3066,
      "mean_token_accuracy": 0.17456333637237548,
      "num_tokens": 39050467.0,
      "step": 21175
    },
    {
      "entropy": 5.695899295806885,
      "epoch": 1.7794160890569208,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046831359765509424,
      "loss": 5.3699,
      "mean_token_accuracy": 0.1715001121163368,
      "num_tokens": 39059224.0,
      "step": 21180
    },
    {
      "entropy": 5.680329132080078,
      "epoch": 1.7798361688720856,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046829827266339134,
      "loss": 5.3889,
      "mean_token_accuracy": 0.1777914956212044,
      "num_tokens": 39067884.0,
      "step": 21185
    },
    {
      "entropy": 5.664518404006958,
      "epoch": 1.7802562486872506,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046828294424750916,
      "loss": 5.3442,
      "mean_token_accuracy": 0.16877941638231278,
      "num_tokens": 39076774.0,
      "step": 21190
    },
    {
      "entropy": 5.635379076004028,
      "epoch": 1.7806763285024154,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004682676124077192,
      "loss": 5.2705,
      "mean_token_accuracy": 0.1737694263458252,
      "num_tokens": 39086021.0,
      "step": 21195
    },
    {
      "entropy": 5.608938407897949,
      "epoch": 1.7810964083175804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046825227714429287,
      "loss": 5.2795,
      "mean_token_accuracy": 0.17561875730752946,
      "num_tokens": 39095682.0,
      "step": 21200
    },
    {
      "entropy": 5.57379412651062,
      "epoch": 1.7815164881327452,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046823693845750205,
      "loss": 5.301,
      "mean_token_accuracy": 0.17501964569091796,
      "num_tokens": 39104904.0,
      "step": 21205
    },
    {
      "entropy": 5.737131071090698,
      "epoch": 1.78193656794791,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046822159634761837,
      "loss": 5.4572,
      "mean_token_accuracy": 0.16661659479141236,
      "num_tokens": 39113128.0,
      "step": 21210
    },
    {
      "entropy": 5.58410234451294,
      "epoch": 1.782356647763075,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004682062508149136,
      "loss": 5.2936,
      "mean_token_accuracy": 0.17243808656930923,
      "num_tokens": 39122503.0,
      "step": 21215
    },
    {
      "entropy": 5.601550912857055,
      "epoch": 1.78277672757824,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004681909018596595,
      "loss": 5.2995,
      "mean_token_accuracy": 0.17471782863140106,
      "num_tokens": 39132020.0,
      "step": 21220
    },
    {
      "entropy": 5.624093914031983,
      "epoch": 1.7831968073934048,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046817554948212813,
      "loss": 5.3466,
      "mean_token_accuracy": 0.1698778361082077,
      "num_tokens": 39141542.0,
      "step": 21225
    },
    {
      "entropy": 5.668030691146851,
      "epoch": 1.7836168872085696,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046816019368259136,
      "loss": 5.3631,
      "mean_token_accuracy": 0.1788243904709816,
      "num_tokens": 39151573.0,
      "step": 21230
    },
    {
      "entropy": 5.66106071472168,
      "epoch": 1.7840369670237344,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004681448344613212,
      "loss": 5.3572,
      "mean_token_accuracy": 0.188814677298069,
      "num_tokens": 39160023.0,
      "step": 21235
    },
    {
      "entropy": 5.594762516021729,
      "epoch": 1.7844570468388994,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046812947181858986,
      "loss": 5.3205,
      "mean_token_accuracy": 0.17250119149684906,
      "num_tokens": 39169335.0,
      "step": 21240
    },
    {
      "entropy": 5.655810070037842,
      "epoch": 1.7848771266540644,
      "grad_norm": 1.625,
      "learning_rate": 0.0004681141057546693,
      "loss": 5.422,
      "mean_token_accuracy": 0.1657840609550476,
      "num_tokens": 39177953.0,
      "step": 21245
    },
    {
      "entropy": 5.660223913192749,
      "epoch": 1.7852972064692292,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046809873626983174,
      "loss": 5.3483,
      "mean_token_accuracy": 0.1715530127286911,
      "num_tokens": 39188984.0,
      "step": 21250
    },
    {
      "entropy": 5.646588373184204,
      "epoch": 1.785717286284394,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046808336336434946,
      "loss": 5.3307,
      "mean_token_accuracy": 0.17568795680999755,
      "num_tokens": 39198033.0,
      "step": 21255
    },
    {
      "entropy": 5.620237874984741,
      "epoch": 1.7861373660995588,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00046806798703849495,
      "loss": 5.2929,
      "mean_token_accuracy": 0.1784186542034149,
      "num_tokens": 39207429.0,
      "step": 21260
    },
    {
      "entropy": 5.590852308273315,
      "epoch": 1.7865574459147238,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004680526072925404,
      "loss": 5.328,
      "mean_token_accuracy": 0.1750107541680336,
      "num_tokens": 39216484.0,
      "step": 21265
    },
    {
      "entropy": 5.712055730819702,
      "epoch": 1.7869775257298888,
      "grad_norm": 3.453125,
      "learning_rate": 0.00046803722412675836,
      "loss": 5.4249,
      "mean_token_accuracy": 0.16705971956253052,
      "num_tokens": 39226385.0,
      "step": 21270
    },
    {
      "entropy": 5.657779169082642,
      "epoch": 1.7873976055450536,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046802183754142125,
      "loss": 5.3416,
      "mean_token_accuracy": 0.1762329265475273,
      "num_tokens": 39235424.0,
      "step": 21275
    },
    {
      "entropy": 5.573896265029907,
      "epoch": 1.7878176853602183,
      "grad_norm": 1.375,
      "learning_rate": 0.0004680064475368017,
      "loss": 5.3038,
      "mean_token_accuracy": 0.17640138417482376,
      "num_tokens": 39244109.0,
      "step": 21280
    },
    {
      "entropy": 5.630048990249634,
      "epoch": 1.7882377651753834,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046799105411317234,
      "loss": 5.3318,
      "mean_token_accuracy": 0.18149891048669814,
      "num_tokens": 39253685.0,
      "step": 21285
    },
    {
      "entropy": 5.616495609283447,
      "epoch": 1.7886578449905484,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046797565727080585,
      "loss": 5.2977,
      "mean_token_accuracy": 0.1702648401260376,
      "num_tokens": 39262743.0,
      "step": 21290
    },
    {
      "entropy": 5.537542009353638,
      "epoch": 1.7890779248057131,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046796025700997484,
      "loss": 5.2278,
      "mean_token_accuracy": 0.18377592116594316,
      "num_tokens": 39270962.0,
      "step": 21295
    },
    {
      "entropy": 5.609395599365234,
      "epoch": 1.789498004620878,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004679448533309523,
      "loss": 5.3207,
      "mean_token_accuracy": 0.1767561912536621,
      "num_tokens": 39279994.0,
      "step": 21300
    },
    {
      "entropy": 5.633199977874756,
      "epoch": 1.7899180844360427,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046792944623401107,
      "loss": 5.3639,
      "mean_token_accuracy": 0.1718871846795082,
      "num_tokens": 39289481.0,
      "step": 21305
    },
    {
      "entropy": 5.686638355255127,
      "epoch": 1.7903381642512077,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046791403571942405,
      "loss": 5.4496,
      "mean_token_accuracy": 0.16385510563850403,
      "num_tokens": 39298383.0,
      "step": 21310
    },
    {
      "entropy": 5.579773664474487,
      "epoch": 1.7907582440663727,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004678986217874642,
      "loss": 5.3325,
      "mean_token_accuracy": 0.1719669818878174,
      "num_tokens": 39307809.0,
      "step": 21315
    },
    {
      "entropy": 5.573810052871704,
      "epoch": 1.7911783238815375,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046788320443840457,
      "loss": 5.2154,
      "mean_token_accuracy": 0.18494438827037812,
      "num_tokens": 39316332.0,
      "step": 21320
    },
    {
      "entropy": 5.572280788421631,
      "epoch": 1.7915984036967023,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046786778367251833,
      "loss": 5.2599,
      "mean_token_accuracy": 0.1805357053875923,
      "num_tokens": 39325672.0,
      "step": 21325
    },
    {
      "entropy": 5.58879075050354,
      "epoch": 1.792018483511867,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046785235949007854,
      "loss": 5.3437,
      "mean_token_accuracy": 0.1765696808695793,
      "num_tokens": 39334478.0,
      "step": 21330
    },
    {
      "entropy": 5.553846168518066,
      "epoch": 1.792438563327032,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046783693189135863,
      "loss": 5.2329,
      "mean_token_accuracy": 0.17667731642723083,
      "num_tokens": 39343573.0,
      "step": 21335
    },
    {
      "entropy": 5.642227840423584,
      "epoch": 1.7928586431421971,
      "grad_norm": 1.375,
      "learning_rate": 0.00046782150087663167,
      "loss": 5.2879,
      "mean_token_accuracy": 0.1806900218129158,
      "num_tokens": 39351956.0,
      "step": 21340
    },
    {
      "entropy": 5.679204702377319,
      "epoch": 1.793278722957362,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004678060664461711,
      "loss": 5.4365,
      "mean_token_accuracy": 0.1642685651779175,
      "num_tokens": 39361911.0,
      "step": 21345
    },
    {
      "entropy": 5.736851692199707,
      "epoch": 1.7936988027725267,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004677906286002504,
      "loss": 5.3767,
      "mean_token_accuracy": 0.17011243253946304,
      "num_tokens": 39370916.0,
      "step": 21350
    },
    {
      "entropy": 5.638383674621582,
      "epoch": 1.7941188825876917,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004677751873391429,
      "loss": 5.3821,
      "mean_token_accuracy": 0.1671166494488716,
      "num_tokens": 39380662.0,
      "step": 21355
    },
    {
      "entropy": 5.618313932418824,
      "epoch": 1.7945389624028567,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046775974266312234,
      "loss": 5.2904,
      "mean_token_accuracy": 0.18158586770296098,
      "num_tokens": 39389644.0,
      "step": 21360
    },
    {
      "entropy": 5.629542827606201,
      "epoch": 1.7949590422180215,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046774429457246215,
      "loss": 5.2961,
      "mean_token_accuracy": 0.1706691563129425,
      "num_tokens": 39398662.0,
      "step": 21365
    },
    {
      "entropy": 5.655536222457886,
      "epoch": 1.7953791220331863,
      "grad_norm": 1.328125,
      "learning_rate": 0.000467728843067436,
      "loss": 5.3729,
      "mean_token_accuracy": 0.17466739267110826,
      "num_tokens": 39408064.0,
      "step": 21370
    },
    {
      "entropy": 5.647149515151978,
      "epoch": 1.795799201848351,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004677133881483177,
      "loss": 5.4039,
      "mean_token_accuracy": 0.16708059906959533,
      "num_tokens": 39418991.0,
      "step": 21375
    },
    {
      "entropy": 5.599534320831299,
      "epoch": 1.796219281663516,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004676979298153809,
      "loss": 5.2612,
      "mean_token_accuracy": 0.17878493815660476,
      "num_tokens": 39428707.0,
      "step": 21380
    },
    {
      "entropy": 5.71638765335083,
      "epoch": 1.796639361478681,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004676824680688996,
      "loss": 5.4112,
      "mean_token_accuracy": 0.17608396708965302,
      "num_tokens": 39437173.0,
      "step": 21385
    },
    {
      "entropy": 5.639132690429688,
      "epoch": 1.7970594412938459,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046766700290914743,
      "loss": 5.3222,
      "mean_token_accuracy": 0.1704282358288765,
      "num_tokens": 39446336.0,
      "step": 21390
    },
    {
      "entropy": 5.699531507492066,
      "epoch": 1.7974795211090107,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046765153433639856,
      "loss": 5.5179,
      "mean_token_accuracy": 0.1648102656006813,
      "num_tokens": 39456129.0,
      "step": 21395
    },
    {
      "entropy": 5.661737394332886,
      "epoch": 1.7978996009241754,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046763606235092705,
      "loss": 5.3657,
      "mean_token_accuracy": 0.17427606582641603,
      "num_tokens": 39465386.0,
      "step": 21400
    },
    {
      "entropy": 5.685305070877075,
      "epoch": 1.7983196807393405,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004676205869530068,
      "loss": 5.4142,
      "mean_token_accuracy": 0.17481597512960434,
      "num_tokens": 39475085.0,
      "step": 21405
    },
    {
      "entropy": 5.669212579727173,
      "epoch": 1.7987397605545055,
      "grad_norm": 1.53125,
      "learning_rate": 0.00046760510814291206,
      "loss": 5.4325,
      "mean_token_accuracy": 0.16868470460176468,
      "num_tokens": 39484500.0,
      "step": 21410
    },
    {
      "entropy": 5.600634145736694,
      "epoch": 1.7991598403696702,
      "grad_norm": 1.28125,
      "learning_rate": 0.000467589625920917,
      "loss": 5.3124,
      "mean_token_accuracy": 0.1692383334040642,
      "num_tokens": 39494049.0,
      "step": 21415
    },
    {
      "entropy": 5.677135896682739,
      "epoch": 1.799579920184835,
      "grad_norm": 1.3125,
      "learning_rate": 0.000467574140287296,
      "loss": 5.322,
      "mean_token_accuracy": 0.17405507266521453,
      "num_tokens": 39502874.0,
      "step": 21420
    },
    {
      "entropy": 5.637249565124511,
      "epoch": 1.8,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004675586512423231,
      "loss": 5.3598,
      "mean_token_accuracy": 0.1670572504401207,
      "num_tokens": 39512371.0,
      "step": 21425
    },
    {
      "entropy": 5.631359481811524,
      "epoch": 1.8004200798151648,
      "grad_norm": 1.640625,
      "learning_rate": 0.000467543158786273,
      "loss": 5.3898,
      "mean_token_accuracy": 0.17028600424528123,
      "num_tokens": 39521477.0,
      "step": 21430
    },
    {
      "entropy": 5.667789363861084,
      "epoch": 1.8008401596303298,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046752766291941985,
      "loss": 5.3868,
      "mean_token_accuracy": 0.16468580663204194,
      "num_tokens": 39530072.0,
      "step": 21435
    },
    {
      "entropy": 5.680898189544678,
      "epoch": 1.8012602394454946,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004675121636420383,
      "loss": 5.3553,
      "mean_token_accuracy": 0.16746123433113097,
      "num_tokens": 39540762.0,
      "step": 21440
    },
    {
      "entropy": 5.664238691329956,
      "epoch": 1.8016803192606594,
      "grad_norm": 1.5625,
      "learning_rate": 0.000467496660954403,
      "loss": 5.3928,
      "mean_token_accuracy": 0.16439993530511857,
      "num_tokens": 39549699.0,
      "step": 21445
    },
    {
      "entropy": 5.619314193725586,
      "epoch": 1.8021003990758244,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046748115485678837,
      "loss": 5.4166,
      "mean_token_accuracy": 0.17035546004772187,
      "num_tokens": 39558725.0,
      "step": 21450
    },
    {
      "entropy": 5.6013425350189205,
      "epoch": 1.8025204788909894,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046746564534946926,
      "loss": 5.259,
      "mean_token_accuracy": 0.18013878613710405,
      "num_tokens": 39567357.0,
      "step": 21455
    },
    {
      "entropy": 5.616558694839478,
      "epoch": 1.8029405587061542,
      "grad_norm": 1.5,
      "learning_rate": 0.0004674501324327203,
      "loss": 5.2653,
      "mean_token_accuracy": 0.18013423681259155,
      "num_tokens": 39576147.0,
      "step": 21460
    },
    {
      "entropy": 5.6275801181793215,
      "epoch": 1.803360638521319,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046743461610681636,
      "loss": 5.4,
      "mean_token_accuracy": 0.18480840176343918,
      "num_tokens": 39584963.0,
      "step": 21465
    },
    {
      "entropy": 5.579710531234741,
      "epoch": 1.8037807183364838,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004674190963720323,
      "loss": 5.2823,
      "mean_token_accuracy": 0.17663349509239196,
      "num_tokens": 39594420.0,
      "step": 21470
    },
    {
      "entropy": 5.5862171173095705,
      "epoch": 1.8042007981516488,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000467403573228643,
      "loss": 5.368,
      "mean_token_accuracy": 0.1672628715634346,
      "num_tokens": 39603276.0,
      "step": 21475
    },
    {
      "entropy": 5.6616593360900875,
      "epoch": 1.8046208779668138,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004673880466769235,
      "loss": 5.4329,
      "mean_token_accuracy": 0.16768532544374465,
      "num_tokens": 39613161.0,
      "step": 21480
    },
    {
      "entropy": 5.646716594696045,
      "epoch": 1.8050409577819786,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046737251671714886,
      "loss": 5.276,
      "mean_token_accuracy": 0.17793395817279817,
      "num_tokens": 39621889.0,
      "step": 21485
    },
    {
      "entropy": 5.679522609710693,
      "epoch": 1.8054610375971434,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046735698334959407,
      "loss": 5.4571,
      "mean_token_accuracy": 0.17540767341852187,
      "num_tokens": 39632009.0,
      "step": 21490
    },
    {
      "entropy": 5.742201519012451,
      "epoch": 1.8058811174123084,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046734144657453443,
      "loss": 5.3542,
      "mean_token_accuracy": 0.17290813773870467,
      "num_tokens": 39640639.0,
      "step": 21495
    },
    {
      "entropy": 5.620148944854736,
      "epoch": 1.8063011972274732,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046732590639224505,
      "loss": 5.3707,
      "mean_token_accuracy": 0.17690147161483766,
      "num_tokens": 39649837.0,
      "step": 21500
    },
    {
      "entropy": 5.62182240486145,
      "epoch": 1.8067212770426382,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046731036280300126,
      "loss": 5.393,
      "mean_token_accuracy": 0.17408841252326965,
      "num_tokens": 39659890.0,
      "step": 21505
    },
    {
      "entropy": 5.638282728195191,
      "epoch": 1.807141356857803,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046729481580707846,
      "loss": 5.3025,
      "mean_token_accuracy": 0.1740954950451851,
      "num_tokens": 39669550.0,
      "step": 21510
    },
    {
      "entropy": 5.603858900070191,
      "epoch": 1.8075614366729678,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046727926540475207,
      "loss": 5.297,
      "mean_token_accuracy": 0.17114973217248916,
      "num_tokens": 39678471.0,
      "step": 21515
    },
    {
      "entropy": 5.549762296676636,
      "epoch": 1.8079815164881328,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004672637115962974,
      "loss": 5.2272,
      "mean_token_accuracy": 0.18205825835466385,
      "num_tokens": 39686600.0,
      "step": 21520
    },
    {
      "entropy": 5.619561386108399,
      "epoch": 1.8084015963032978,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046724815438199007,
      "loss": 5.3808,
      "mean_token_accuracy": 0.16773713529109954,
      "num_tokens": 39696848.0,
      "step": 21525
    },
    {
      "entropy": 5.586813831329346,
      "epoch": 1.8088216761184626,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046723259376210577,
      "loss": 5.307,
      "mean_token_accuracy": 0.17681580185890197,
      "num_tokens": 39706051.0,
      "step": 21530
    },
    {
      "entropy": 5.691588592529297,
      "epoch": 1.8092417559336273,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046721702973692,
      "loss": 5.3826,
      "mean_token_accuracy": 0.16805659085512162,
      "num_tokens": 39716035.0,
      "step": 21535
    },
    {
      "entropy": 5.633076953887939,
      "epoch": 1.8096618357487921,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046720146230670853,
      "loss": 5.3489,
      "mean_token_accuracy": 0.1706215813755989,
      "num_tokens": 39725717.0,
      "step": 21540
    },
    {
      "entropy": 5.657817125320435,
      "epoch": 1.8100819155639571,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004671858914717471,
      "loss": 5.3741,
      "mean_token_accuracy": 0.17011357694864274,
      "num_tokens": 39734543.0,
      "step": 21545
    },
    {
      "entropy": 5.6470294952392575,
      "epoch": 1.8105019953791222,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046717031723231164,
      "loss": 5.3757,
      "mean_token_accuracy": 0.1764549121260643,
      "num_tokens": 39744503.0,
      "step": 21550
    },
    {
      "entropy": 5.604462337493897,
      "epoch": 1.810922075194287,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004671547395886779,
      "loss": 5.36,
      "mean_token_accuracy": 0.16842081546783447,
      "num_tokens": 39753484.0,
      "step": 21555
    },
    {
      "entropy": 5.60592474937439,
      "epoch": 1.8113421550094517,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004671391585411219,
      "loss": 5.2779,
      "mean_token_accuracy": 0.1773709163069725,
      "num_tokens": 39762673.0,
      "step": 21560
    },
    {
      "entropy": 5.708010625839234,
      "epoch": 1.8117622348246165,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046712357408991965,
      "loss": 5.4279,
      "mean_token_accuracy": 0.16579594165086747,
      "num_tokens": 39773138.0,
      "step": 21565
    },
    {
      "entropy": 5.712737894058227,
      "epoch": 1.8121823146397815,
      "grad_norm": 1.875,
      "learning_rate": 0.0004671079862353472,
      "loss": 5.4252,
      "mean_token_accuracy": 0.17003078311681746,
      "num_tokens": 39782282.0,
      "step": 21570
    },
    {
      "entropy": 5.618183326721192,
      "epoch": 1.8126023944549465,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046709239497768067,
      "loss": 5.3235,
      "mean_token_accuracy": 0.17582568526268005,
      "num_tokens": 39792035.0,
      "step": 21575
    },
    {
      "entropy": 5.715729093551635,
      "epoch": 1.8130224742701113,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046707680031719633,
      "loss": 5.4163,
      "mean_token_accuracy": 0.1705210655927658,
      "num_tokens": 39801696.0,
      "step": 21580
    },
    {
      "entropy": 5.72704849243164,
      "epoch": 1.813442554085276,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004670612022541705,
      "loss": 5.4477,
      "mean_token_accuracy": 0.16532195806503297,
      "num_tokens": 39811449.0,
      "step": 21585
    },
    {
      "entropy": 5.668391561508178,
      "epoch": 1.8138626339004411,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004670456007888792,
      "loss": 5.3956,
      "mean_token_accuracy": 0.1733089879155159,
      "num_tokens": 39820339.0,
      "step": 21590
    },
    {
      "entropy": 5.606714677810669,
      "epoch": 1.8142827137156061,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004670299959215989,
      "loss": 5.3149,
      "mean_token_accuracy": 0.17719878107309342,
      "num_tokens": 39829861.0,
      "step": 21595
    },
    {
      "entropy": 5.632393932342529,
      "epoch": 1.814702793530771,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004670143876526062,
      "loss": 5.2809,
      "mean_token_accuracy": 0.18143440186977386,
      "num_tokens": 39838568.0,
      "step": 21600
    },
    {
      "entropy": 5.641485643386841,
      "epoch": 1.8151228733459357,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046699877598217754,
      "loss": 5.3003,
      "mean_token_accuracy": 0.17472619712352752,
      "num_tokens": 39847705.0,
      "step": 21605
    },
    {
      "entropy": 5.638858842849731,
      "epoch": 1.8155429531611005,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046698316091058946,
      "loss": 5.4095,
      "mean_token_accuracy": 0.16513228118419648,
      "num_tokens": 39856700.0,
      "step": 21610
    },
    {
      "entropy": 5.667533874511719,
      "epoch": 1.8159630329762655,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046696754243811845,
      "loss": 5.2872,
      "mean_token_accuracy": 0.17423474192619323,
      "num_tokens": 39865647.0,
      "step": 21615
    },
    {
      "entropy": 5.676429414749146,
      "epoch": 1.8163831127914305,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004669519205650413,
      "loss": 5.4044,
      "mean_token_accuracy": 0.17259556651115418,
      "num_tokens": 39874705.0,
      "step": 21620
    },
    {
      "entropy": 5.6331953525543215,
      "epoch": 1.8168031926065953,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046693629529163467,
      "loss": 5.236,
      "mean_token_accuracy": 0.17874569296836854,
      "num_tokens": 39883795.0,
      "step": 21625
    },
    {
      "entropy": 5.627444171905518,
      "epoch": 1.81722327242176,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004669206666181755,
      "loss": 5.327,
      "mean_token_accuracy": 0.18165523260831834,
      "num_tokens": 39893165.0,
      "step": 21630
    },
    {
      "entropy": 5.575399112701416,
      "epoch": 1.8176433522369249,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004669050345449404,
      "loss": 5.3686,
      "mean_token_accuracy": 0.16783955916762353,
      "num_tokens": 39902241.0,
      "step": 21635
    },
    {
      "entropy": 5.608601427078247,
      "epoch": 1.8180634320520899,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004668893990722066,
      "loss": 5.3213,
      "mean_token_accuracy": 0.16776576787233352,
      "num_tokens": 39911211.0,
      "step": 21640
    },
    {
      "entropy": 5.644370222091675,
      "epoch": 1.8184835118672549,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004668737602002508,
      "loss": 5.3176,
      "mean_token_accuracy": 0.17453420162200928,
      "num_tokens": 39920192.0,
      "step": 21645
    },
    {
      "entropy": 5.6630859375,
      "epoch": 1.8189035916824197,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046685811792935016,
      "loss": 5.3435,
      "mean_token_accuracy": 0.17098472267389297,
      "num_tokens": 39929169.0,
      "step": 21650
    },
    {
      "entropy": 5.634385013580323,
      "epoch": 1.8193236714975844,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046684247225978176,
      "loss": 5.3576,
      "mean_token_accuracy": 0.1713916167616844,
      "num_tokens": 39939333.0,
      "step": 21655
    },
    {
      "entropy": 5.635967779159546,
      "epoch": 1.8197437513127495,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00046682682319182275,
      "loss": 5.3459,
      "mean_token_accuracy": 0.17324915826320647,
      "num_tokens": 39948042.0,
      "step": 21660
    },
    {
      "entropy": 5.64743332862854,
      "epoch": 1.8201638311279145,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046681117072575035,
      "loss": 5.2929,
      "mean_token_accuracy": 0.17710241228342055,
      "num_tokens": 39956847.0,
      "step": 21665
    },
    {
      "entropy": 5.795207166671753,
      "epoch": 1.8205839109430793,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004667955148618418,
      "loss": 5.54,
      "mean_token_accuracy": 0.1618648052215576,
      "num_tokens": 39966598.0,
      "step": 21670
    },
    {
      "entropy": 5.571113538742066,
      "epoch": 1.821003990758244,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004667798556003745,
      "loss": 5.2033,
      "mean_token_accuracy": 0.17449332773685455,
      "num_tokens": 39975236.0,
      "step": 21675
    },
    {
      "entropy": 5.586743497848511,
      "epoch": 1.8214240705734088,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004667641929416258,
      "loss": 5.353,
      "mean_token_accuracy": 0.17485176473855973,
      "num_tokens": 39984582.0,
      "step": 21680
    },
    {
      "entropy": 5.642992782592773,
      "epoch": 1.8218441503885738,
      "grad_norm": 1.25,
      "learning_rate": 0.0004667485268858731,
      "loss": 5.3664,
      "mean_token_accuracy": 0.1721978545188904,
      "num_tokens": 39993122.0,
      "step": 21685
    },
    {
      "entropy": 5.653185701370239,
      "epoch": 1.8222642302037388,
      "grad_norm": 1.625,
      "learning_rate": 0.00046673285743339406,
      "loss": 5.323,
      "mean_token_accuracy": 0.1807267799973488,
      "num_tokens": 40002974.0,
      "step": 21690
    },
    {
      "entropy": 5.6511951923370365,
      "epoch": 1.8226843100189036,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046671718458446616,
      "loss": 5.3597,
      "mean_token_accuracy": 0.17250411510467528,
      "num_tokens": 40011790.0,
      "step": 21695
    },
    {
      "entropy": 5.6587913036346436,
      "epoch": 1.8231043898340684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004667015083393671,
      "loss": 5.3655,
      "mean_token_accuracy": 0.16823602020740508,
      "num_tokens": 40021327.0,
      "step": 21700
    },
    {
      "entropy": 5.63710470199585,
      "epoch": 1.8235244696492332,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004666858286983744,
      "loss": 5.3586,
      "mean_token_accuracy": 0.16855680644512178,
      "num_tokens": 40030471.0,
      "step": 21705
    },
    {
      "entropy": 5.681593751907348,
      "epoch": 1.8239445494643982,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004666701456617661,
      "loss": 5.3664,
      "mean_token_accuracy": 0.17184626013040544,
      "num_tokens": 40039305.0,
      "step": 21710
    },
    {
      "entropy": 5.609969806671143,
      "epoch": 1.8243646292795632,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046665445922981975,
      "loss": 5.2699,
      "mean_token_accuracy": 0.17720975875854492,
      "num_tokens": 40047389.0,
      "step": 21715
    },
    {
      "entropy": 5.694111299514771,
      "epoch": 1.824784709094728,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004666387694028134,
      "loss": 5.3638,
      "mean_token_accuracy": 0.17174751609563826,
      "num_tokens": 40057640.0,
      "step": 21720
    },
    {
      "entropy": 5.548367547988891,
      "epoch": 1.8252047889098928,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004666230761810249,
      "loss": 5.3243,
      "mean_token_accuracy": 0.17332008928060533,
      "num_tokens": 40066770.0,
      "step": 21725
    },
    {
      "entropy": 5.549862003326416,
      "epoch": 1.8256248687250578,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004666073795647323,
      "loss": 5.3015,
      "mean_token_accuracy": 0.17694138437509538,
      "num_tokens": 40075902.0,
      "step": 21730
    },
    {
      "entropy": 5.6276435375213625,
      "epoch": 1.8260449485402226,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046659167955421366,
      "loss": 5.3486,
      "mean_token_accuracy": 0.1689707115292549,
      "num_tokens": 40084945.0,
      "step": 21735
    },
    {
      "entropy": 5.576631355285644,
      "epoch": 1.8264650283553876,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000466575976149747,
      "loss": 5.2522,
      "mean_token_accuracy": 0.17926783561706544,
      "num_tokens": 40095104.0,
      "step": 21740
    },
    {
      "entropy": 5.670596361160278,
      "epoch": 1.8268851081705524,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004665602693516106,
      "loss": 5.4077,
      "mean_token_accuracy": 0.1704809010028839,
      "num_tokens": 40105329.0,
      "step": 21745
    },
    {
      "entropy": 5.599937200546265,
      "epoch": 1.8273051879857172,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004665445591600827,
      "loss": 5.2048,
      "mean_token_accuracy": 0.18144048005342484,
      "num_tokens": 40114555.0,
      "step": 21750
    },
    {
      "entropy": 5.643712377548217,
      "epoch": 1.8277252678008822,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004665288455754415,
      "loss": 5.2642,
      "mean_token_accuracy": 0.18136225491762162,
      "num_tokens": 40123314.0,
      "step": 21755
    },
    {
      "entropy": 5.597834587097168,
      "epoch": 1.8281453476160472,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004665131285979655,
      "loss": 5.2887,
      "mean_token_accuracy": 0.17629961371421815,
      "num_tokens": 40132483.0,
      "step": 21760
    },
    {
      "entropy": 5.597902536392212,
      "epoch": 1.828565427431212,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00046649740822793303,
      "loss": 5.3287,
      "mean_token_accuracy": 0.1733765110373497,
      "num_tokens": 40141800.0,
      "step": 21765
    },
    {
      "entropy": 5.679841613769531,
      "epoch": 1.8289855072463768,
      "grad_norm": 3.75,
      "learning_rate": 0.0004664816844656225,
      "loss": 5.3299,
      "mean_token_accuracy": 0.17633609622716903,
      "num_tokens": 40149892.0,
      "step": 21770
    },
    {
      "entropy": 5.576864194869995,
      "epoch": 1.8294055870615415,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046646595731131263,
      "loss": 5.2928,
      "mean_token_accuracy": 0.17471731156110765,
      "num_tokens": 40159376.0,
      "step": 21775
    },
    {
      "entropy": 5.567289018630982,
      "epoch": 1.8298256668767066,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004664502267652819,
      "loss": 5.2672,
      "mean_token_accuracy": 0.1796356737613678,
      "num_tokens": 40168497.0,
      "step": 21780
    },
    {
      "entropy": 5.655853986740112,
      "epoch": 1.8302457466918716,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046643449282780894,
      "loss": 5.3601,
      "mean_token_accuracy": 0.16903401017189026,
      "num_tokens": 40177432.0,
      "step": 21785
    },
    {
      "entropy": 5.628530502319336,
      "epoch": 1.8306658265070364,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004664187554991725,
      "loss": 5.2298,
      "mean_token_accuracy": 0.17844704687595367,
      "num_tokens": 40186582.0,
      "step": 21790
    },
    {
      "entropy": 5.574361181259155,
      "epoch": 1.8310859063222011,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004664030147796514,
      "loss": 5.2978,
      "mean_token_accuracy": 0.17666405737400054,
      "num_tokens": 40196094.0,
      "step": 21795
    },
    {
      "entropy": 5.566331148147583,
      "epoch": 1.8315059861373661,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004663872706695244,
      "loss": 5.3698,
      "mean_token_accuracy": 0.1692861020565033,
      "num_tokens": 40205239.0,
      "step": 21800
    },
    {
      "entropy": 5.62432599067688,
      "epoch": 1.831926065952531,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004663715231690706,
      "loss": 5.4233,
      "mean_token_accuracy": 0.173600272834301,
      "num_tokens": 40213908.0,
      "step": 21805
    },
    {
      "entropy": 5.681668758392334,
      "epoch": 1.832346145767696,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046635577227856873,
      "loss": 5.3788,
      "mean_token_accuracy": 0.17203570753335953,
      "num_tokens": 40223370.0,
      "step": 21810
    },
    {
      "entropy": 5.7056420803070065,
      "epoch": 1.8327662255828607,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004663400179982978,
      "loss": 5.4602,
      "mean_token_accuracy": 0.16442577838897704,
      "num_tokens": 40233934.0,
      "step": 21815
    },
    {
      "entropy": 5.666644048690796,
      "epoch": 1.8331863053980255,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046632426032853705,
      "loss": 5.3352,
      "mean_token_accuracy": 0.17081063985824585,
      "num_tokens": 40244335.0,
      "step": 21820
    },
    {
      "entropy": 5.617442274093628,
      "epoch": 1.8336063852131905,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046630849926956555,
      "loss": 5.2878,
      "mean_token_accuracy": 0.17147090584039687,
      "num_tokens": 40254354.0,
      "step": 21825
    },
    {
      "entropy": 5.640371322631836,
      "epoch": 1.8340264650283555,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046629273482166244,
      "loss": 5.2935,
      "mean_token_accuracy": 0.17569748163223267,
      "num_tokens": 40262748.0,
      "step": 21830
    },
    {
      "entropy": 5.641110467910766,
      "epoch": 1.8344465448435203,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046627696698510706,
      "loss": 5.3549,
      "mean_token_accuracy": 0.1760078102350235,
      "num_tokens": 40271818.0,
      "step": 21835
    },
    {
      "entropy": 5.610189628601074,
      "epoch": 1.834866624658685,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004662611957601788,
      "loss": 5.392,
      "mean_token_accuracy": 0.17088382542133332,
      "num_tokens": 40280552.0,
      "step": 21840
    },
    {
      "entropy": 5.656275892257691,
      "epoch": 1.83528670447385,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00046624542114715687,
      "loss": 5.2808,
      "mean_token_accuracy": 0.18224855810403823,
      "num_tokens": 40289368.0,
      "step": 21845
    },
    {
      "entropy": 5.792998313903809,
      "epoch": 1.835706784289015,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004662296431463208,
      "loss": 5.4917,
      "mean_token_accuracy": 0.1611533358693123,
      "num_tokens": 40298884.0,
      "step": 21850
    },
    {
      "entropy": 5.6724272727966305,
      "epoch": 1.83612686410418,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046621386175795,
      "loss": 5.3982,
      "mean_token_accuracy": 0.1661871299147606,
      "num_tokens": 40307886.0,
      "step": 21855
    },
    {
      "entropy": 5.563909292221069,
      "epoch": 1.8365469439193447,
      "grad_norm": 1.671875,
      "learning_rate": 0.00046619807698232413,
      "loss": 5.313,
      "mean_token_accuracy": 0.17353412508964539,
      "num_tokens": 40317688.0,
      "step": 21860
    },
    {
      "entropy": 5.649786376953125,
      "epoch": 1.8369670237345095,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004661822888197228,
      "loss": 5.3547,
      "mean_token_accuracy": 0.17302406430244446,
      "num_tokens": 40327630.0,
      "step": 21865
    },
    {
      "entropy": 5.619706535339356,
      "epoch": 1.8373871035496743,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046616649727042564,
      "loss": 5.337,
      "mean_token_accuracy": 0.17453055679798127,
      "num_tokens": 40336613.0,
      "step": 21870
    },
    {
      "entropy": 5.627038192749024,
      "epoch": 1.8378071833648393,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046615070233471244,
      "loss": 5.4291,
      "mean_token_accuracy": 0.17172746509313583,
      "num_tokens": 40346582.0,
      "step": 21875
    },
    {
      "entropy": 5.72258825302124,
      "epoch": 1.8382272631800043,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046613490401286304,
      "loss": 5.4594,
      "mean_token_accuracy": 0.16815469712018966,
      "num_tokens": 40355960.0,
      "step": 21880
    },
    {
      "entropy": 5.724008083343506,
      "epoch": 1.838647342995169,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046611910230515716,
      "loss": 5.2927,
      "mean_token_accuracy": 0.17314579337835312,
      "num_tokens": 40366043.0,
      "step": 21885
    },
    {
      "entropy": 5.6509898662567135,
      "epoch": 1.8390674228103339,
      "grad_norm": 1.25,
      "learning_rate": 0.0004661032972118748,
      "loss": 5.3526,
      "mean_token_accuracy": 0.17593541890382766,
      "num_tokens": 40374919.0,
      "step": 21890
    },
    {
      "entropy": 5.578866338729858,
      "epoch": 1.8394875026254989,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046608748873329587,
      "loss": 5.3349,
      "mean_token_accuracy": 0.1761649116873741,
      "num_tokens": 40383415.0,
      "step": 21895
    },
    {
      "entropy": 5.706315612792968,
      "epoch": 1.8399075824406639,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004660716768697005,
      "loss": 5.372,
      "mean_token_accuracy": 0.16670249551534652,
      "num_tokens": 40392252.0,
      "step": 21900
    },
    {
      "entropy": 5.6273149967193605,
      "epoch": 1.8403276622558287,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004660558616213689,
      "loss": 5.2247,
      "mean_token_accuracy": 0.1862599477171898,
      "num_tokens": 40400717.0,
      "step": 21905
    },
    {
      "entropy": 5.560521984100342,
      "epoch": 1.8407477420709935,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046604004298858093,
      "loss": 5.2559,
      "mean_token_accuracy": 0.183683280646801,
      "num_tokens": 40409236.0,
      "step": 21910
    },
    {
      "entropy": 5.530434703826904,
      "epoch": 1.8411678218861582,
      "grad_norm": 1.25,
      "learning_rate": 0.0004660242209716171,
      "loss": 5.2713,
      "mean_token_accuracy": 0.18041382730007172,
      "num_tokens": 40419073.0,
      "step": 21915
    },
    {
      "entropy": 5.677064657211304,
      "epoch": 1.8415879017013232,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004660083955707575,
      "loss": 5.3909,
      "mean_token_accuracy": 0.17495720088481903,
      "num_tokens": 40428427.0,
      "step": 21920
    },
    {
      "entropy": 5.696070051193237,
      "epoch": 1.8420079815164883,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004659925667862825,
      "loss": 5.3503,
      "mean_token_accuracy": 0.17721674591302872,
      "num_tokens": 40437350.0,
      "step": 21925
    },
    {
      "entropy": 5.614517307281494,
      "epoch": 1.842428061331653,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004659767346184725,
      "loss": 5.3643,
      "mean_token_accuracy": 0.16970486044883729,
      "num_tokens": 40446059.0,
      "step": 21930
    },
    {
      "entropy": 5.656706428527832,
      "epoch": 1.8428481411468178,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00046596089906760803,
      "loss": 5.356,
      "mean_token_accuracy": 0.17482821494340897,
      "num_tokens": 40454959.0,
      "step": 21935
    },
    {
      "entropy": 5.662093782424927,
      "epoch": 1.8432682209619826,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004659450601339696,
      "loss": 5.3565,
      "mean_token_accuracy": 0.1761407434940338,
      "num_tokens": 40464202.0,
      "step": 21940
    },
    {
      "entropy": 5.593807172775269,
      "epoch": 1.8436883007771476,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004659292178178377,
      "loss": 5.3023,
      "mean_token_accuracy": 0.18095338940620423,
      "num_tokens": 40473331.0,
      "step": 21945
    },
    {
      "entropy": 5.583661985397339,
      "epoch": 1.8441083805923126,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000465913372119493,
      "loss": 5.2869,
      "mean_token_accuracy": 0.16963033825159074,
      "num_tokens": 40482098.0,
      "step": 21950
    },
    {
      "entropy": 5.641996908187866,
      "epoch": 1.8445284604074774,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004658975230392162,
      "loss": 5.3233,
      "mean_token_accuracy": 0.18288161158561705,
      "num_tokens": 40491134.0,
      "step": 21955
    },
    {
      "entropy": 5.718777418136597,
      "epoch": 1.8449485402226422,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004658816705772882,
      "loss": 5.4616,
      "mean_token_accuracy": 0.16886609494686128,
      "num_tokens": 40501488.0,
      "step": 21960
    },
    {
      "entropy": 5.566116714477539,
      "epoch": 1.8453686200378072,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004658658147339896,
      "loss": 5.2047,
      "mean_token_accuracy": 0.1825152888894081,
      "num_tokens": 40510506.0,
      "step": 21965
    },
    {
      "entropy": 5.652201843261719,
      "epoch": 1.8457886998529722,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00046584995550960146,
      "loss": 5.3393,
      "mean_token_accuracy": 0.18174153417348862,
      "num_tokens": 40520222.0,
      "step": 21970
    },
    {
      "entropy": 5.555640316009521,
      "epoch": 1.846208779668137,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046583409290440453,
      "loss": 5.2525,
      "mean_token_accuracy": 0.18225459456443788,
      "num_tokens": 40528824.0,
      "step": 21975
    },
    {
      "entropy": 5.549523544311524,
      "epoch": 1.8466288594833018,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004658182269186799,
      "loss": 5.3449,
      "mean_token_accuracy": 0.16833662688732148,
      "num_tokens": 40538144.0,
      "step": 21980
    },
    {
      "entropy": 5.620516061782837,
      "epoch": 1.8470489392984666,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004658023575527087,
      "loss": 5.3834,
      "mean_token_accuracy": 0.16834884285926818,
      "num_tokens": 40547457.0,
      "step": 21985
    },
    {
      "entropy": 5.6714612483978275,
      "epoch": 1.8474690191136316,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000465786484806772,
      "loss": 5.266,
      "mean_token_accuracy": 0.18217454850673676,
      "num_tokens": 40556005.0,
      "step": 21990
    },
    {
      "entropy": 5.475649690628051,
      "epoch": 1.8478890989287966,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046577060868115095,
      "loss": 5.2442,
      "mean_token_accuracy": 0.18496721237897873,
      "num_tokens": 40565018.0,
      "step": 21995
    },
    {
      "entropy": 5.596051168441773,
      "epoch": 1.8483091787439614,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004657547291761268,
      "loss": 5.3036,
      "mean_token_accuracy": 0.17018678337335585,
      "num_tokens": 40574931.0,
      "step": 22000
    },
    {
      "entropy": 5.675346565246582,
      "epoch": 1.8487292585591262,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046573884629198077,
      "loss": 5.2839,
      "mean_token_accuracy": 0.17571477442979813,
      "num_tokens": 40584496.0,
      "step": 22005
    },
    {
      "entropy": 5.692410802841186,
      "epoch": 1.849149338374291,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004657229600289944,
      "loss": 5.376,
      "mean_token_accuracy": 0.1714554563164711,
      "num_tokens": 40594363.0,
      "step": 22010
    },
    {
      "entropy": 5.631646060943604,
      "epoch": 1.849569418189456,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004657070703874489,
      "loss": 5.3958,
      "mean_token_accuracy": 0.1703170582652092,
      "num_tokens": 40603001.0,
      "step": 22015
    },
    {
      "entropy": 5.617373275756836,
      "epoch": 1.849989498004621,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046569117736762597,
      "loss": 5.3394,
      "mean_token_accuracy": 0.1786193698644638,
      "num_tokens": 40612660.0,
      "step": 22020
    },
    {
      "entropy": 5.577786636352539,
      "epoch": 1.8504095778197858,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00046567528096980686,
      "loss": 5.2644,
      "mean_token_accuracy": 0.17572253346443176,
      "num_tokens": 40622209.0,
      "step": 22025
    },
    {
      "entropy": 5.646335411071777,
      "epoch": 1.8508296576349506,
      "grad_norm": 1.921875,
      "learning_rate": 0.00046565938119427346,
      "loss": 5.3556,
      "mean_token_accuracy": 0.1685689240694046,
      "num_tokens": 40632011.0,
      "step": 22030
    },
    {
      "entropy": 5.5993945598602295,
      "epoch": 1.8512497374501156,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004656434780413073,
      "loss": 5.2398,
      "mean_token_accuracy": 0.1800299823284149,
      "num_tokens": 40641201.0,
      "step": 22035
    },
    {
      "entropy": 5.552846813201905,
      "epoch": 1.8516698172652803,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046562757151119,
      "loss": 5.2858,
      "mean_token_accuracy": 0.17844798266887665,
      "num_tokens": 40650752.0,
      "step": 22040
    },
    {
      "entropy": 5.581203413009644,
      "epoch": 1.8520898970804454,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004656116616042035,
      "loss": 5.2909,
      "mean_token_accuracy": 0.1712008073925972,
      "num_tokens": 40659975.0,
      "step": 22045
    },
    {
      "entropy": 5.586377143859863,
      "epoch": 1.8525099768956101,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046559574832062955,
      "loss": 5.3331,
      "mean_token_accuracy": 0.1807105213403702,
      "num_tokens": 40668944.0,
      "step": 22050
    },
    {
      "entropy": 5.698160552978516,
      "epoch": 1.852930056710775,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046557983166075,
      "loss": 5.4465,
      "mean_token_accuracy": 0.172979998588562,
      "num_tokens": 40678333.0,
      "step": 22055
    },
    {
      "entropy": 5.54446816444397,
      "epoch": 1.85335013652594,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046556391162484696,
      "loss": 5.1986,
      "mean_token_accuracy": 0.18016323447227478,
      "num_tokens": 40687781.0,
      "step": 22060
    },
    {
      "entropy": 5.604084444046021,
      "epoch": 1.853770216341105,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004655479882132023,
      "loss": 5.3654,
      "mean_token_accuracy": 0.17439820766448974,
      "num_tokens": 40697637.0,
      "step": 22065
    },
    {
      "entropy": 5.54901351928711,
      "epoch": 1.8541902961562697,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004655320614260982,
      "loss": 5.2468,
      "mean_token_accuracy": 0.18014926761388778,
      "num_tokens": 40707097.0,
      "step": 22070
    },
    {
      "entropy": 5.72498574256897,
      "epoch": 1.8546103759714345,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046551613126381673,
      "loss": 5.4193,
      "mean_token_accuracy": 0.17032432705163955,
      "num_tokens": 40716821.0,
      "step": 22075
    },
    {
      "entropy": 5.617212343215942,
      "epoch": 1.8550304557865993,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004655001977266401,
      "loss": 5.3188,
      "mean_token_accuracy": 0.17428034245967866,
      "num_tokens": 40726731.0,
      "step": 22080
    },
    {
      "entropy": 5.686445569992065,
      "epoch": 1.8554505356017643,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046548426081485046,
      "loss": 5.277,
      "mean_token_accuracy": 0.17660748213529587,
      "num_tokens": 40736935.0,
      "step": 22085
    },
    {
      "entropy": 5.648962354660034,
      "epoch": 1.8558706154169293,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046546832052873026,
      "loss": 5.4276,
      "mean_token_accuracy": 0.16823360919952393,
      "num_tokens": 40746643.0,
      "step": 22090
    },
    {
      "entropy": 5.748947095870972,
      "epoch": 1.8562906952320941,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046545237686856195,
      "loss": 5.4659,
      "mean_token_accuracy": 0.16358623802661895,
      "num_tokens": 40755713.0,
      "step": 22095
    },
    {
      "entropy": 5.683595037460327,
      "epoch": 1.856710775047259,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046543642983462775,
      "loss": 5.4533,
      "mean_token_accuracy": 0.17363260835409164,
      "num_tokens": 40764878.0,
      "step": 22100
    },
    {
      "entropy": 5.671045780181885,
      "epoch": 1.857130854862424,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046542047942721025,
      "loss": 5.3395,
      "mean_token_accuracy": 0.1731833338737488,
      "num_tokens": 40774101.0,
      "step": 22105
    },
    {
      "entropy": 5.6399682521820065,
      "epoch": 1.8575509346775887,
      "grad_norm": 1.640625,
      "learning_rate": 0.000465404525646592,
      "loss": 5.3238,
      "mean_token_accuracy": 0.17589335143566132,
      "num_tokens": 40783126.0,
      "step": 22110
    },
    {
      "entropy": 5.552609825134278,
      "epoch": 1.8579710144927537,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004653885684930557,
      "loss": 5.3089,
      "mean_token_accuracy": 0.17975108027458192,
      "num_tokens": 40792508.0,
      "step": 22115
    },
    {
      "entropy": 5.620856618881225,
      "epoch": 1.8583910943079185,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004653726079668839,
      "loss": 5.3545,
      "mean_token_accuracy": 0.17612426429986955,
      "num_tokens": 40802252.0,
      "step": 22120
    },
    {
      "entropy": 5.566342496871949,
      "epoch": 1.8588111741230833,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004653566440683594,
      "loss": 5.1846,
      "mean_token_accuracy": 0.1835705369710922,
      "num_tokens": 40811041.0,
      "step": 22125
    },
    {
      "entropy": 5.602682447433471,
      "epoch": 1.8592312539382483,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000465340676797765,
      "loss": 5.2327,
      "mean_token_accuracy": 0.1757661372423172,
      "num_tokens": 40819976.0,
      "step": 22130
    },
    {
      "entropy": 5.583587217330932,
      "epoch": 1.8596513337534133,
      "grad_norm": 1.625,
      "learning_rate": 0.00046532470615538344,
      "loss": 5.2729,
      "mean_token_accuracy": 0.17985140085220336,
      "num_tokens": 40828544.0,
      "step": 22135
    },
    {
      "entropy": 5.615645503997802,
      "epoch": 1.860071413568578,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046530873214149776,
      "loss": 5.363,
      "mean_token_accuracy": 0.18175422698259353,
      "num_tokens": 40838386.0,
      "step": 22140
    },
    {
      "entropy": 5.704379415512085,
      "epoch": 1.8604914933837429,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004652927547563908,
      "loss": 5.3436,
      "mean_token_accuracy": 0.172553414106369,
      "num_tokens": 40847047.0,
      "step": 22145
    },
    {
      "entropy": 5.567825746536255,
      "epoch": 1.8609115731989077,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004652767740003458,
      "loss": 5.389,
      "mean_token_accuracy": 0.17185918167233466,
      "num_tokens": 40856653.0,
      "step": 22150
    },
    {
      "entropy": 5.68084487915039,
      "epoch": 1.8613316530140727,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046526078987364566,
      "loss": 5.4367,
      "mean_token_accuracy": 0.1669323831796646,
      "num_tokens": 40865176.0,
      "step": 22155
    },
    {
      "entropy": 5.728604030609131,
      "epoch": 1.8617517328292377,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004652448023765736,
      "loss": 5.4836,
      "mean_token_accuracy": 0.17508383989334106,
      "num_tokens": 40874084.0,
      "step": 22160
    },
    {
      "entropy": 5.710894060134888,
      "epoch": 1.8621718126444025,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004652288115094129,
      "loss": 5.3881,
      "mean_token_accuracy": 0.17289437353610992,
      "num_tokens": 40883704.0,
      "step": 22165
    },
    {
      "entropy": 5.6572003841400145,
      "epoch": 1.8625918924595672,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004652128172724466,
      "loss": 5.3997,
      "mean_token_accuracy": 0.16919400095939635,
      "num_tokens": 40893232.0,
      "step": 22170
    },
    {
      "entropy": 5.635032892227173,
      "epoch": 1.8630119722747323,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00046519681966595834,
      "loss": 5.3041,
      "mean_token_accuracy": 0.17997007369995116,
      "num_tokens": 40902242.0,
      "step": 22175
    },
    {
      "entropy": 5.5873143672943115,
      "epoch": 1.863432052089897,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004651808186902313,
      "loss": 5.2982,
      "mean_token_accuracy": 0.17320200502872468,
      "num_tokens": 40912349.0,
      "step": 22180
    },
    {
      "entropy": 5.5867170810699465,
      "epoch": 1.863852131905062,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000465164814345549,
      "loss": 5.3169,
      "mean_token_accuracy": 0.1751672551035881,
      "num_tokens": 40922206.0,
      "step": 22185
    },
    {
      "entropy": 5.657592630386352,
      "epoch": 1.8642722117202268,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046514880663219493,
      "loss": 5.272,
      "mean_token_accuracy": 0.1812136799097061,
      "num_tokens": 40931145.0,
      "step": 22190
    },
    {
      "entropy": 5.5490403175354,
      "epoch": 1.8646922915353916,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004651327955504526,
      "loss": 5.2162,
      "mean_token_accuracy": 0.184327432513237,
      "num_tokens": 40939917.0,
      "step": 22195
    },
    {
      "entropy": 5.579754066467285,
      "epoch": 1.8651123713505566,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004651167811006058,
      "loss": 5.259,
      "mean_token_accuracy": 0.18346578031778335,
      "num_tokens": 40947972.0,
      "step": 22200
    },
    {
      "entropy": 5.529523134231567,
      "epoch": 1.8655324511657216,
      "grad_norm": 1.40625,
      "learning_rate": 0.000465100763282938,
      "loss": 5.2243,
      "mean_token_accuracy": 0.1837876707315445,
      "num_tokens": 40956999.0,
      "step": 22205
    },
    {
      "entropy": 5.557683277130127,
      "epoch": 1.8659525309808864,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004650847420977332,
      "loss": 5.2137,
      "mean_token_accuracy": 0.1851109817624092,
      "num_tokens": 40965917.0,
      "step": 22210
    },
    {
      "entropy": 5.564731502532959,
      "epoch": 1.8663726107960512,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046506871754527495,
      "loss": 5.309,
      "mean_token_accuracy": 0.172707436978817,
      "num_tokens": 40976545.0,
      "step": 22215
    },
    {
      "entropy": 5.646349143981934,
      "epoch": 1.866792690611216,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046505268962584735,
      "loss": 5.3244,
      "mean_token_accuracy": 0.18334989249706268,
      "num_tokens": 40985890.0,
      "step": 22220
    },
    {
      "entropy": 5.69091272354126,
      "epoch": 1.867212770426381,
      "grad_norm": 1.5,
      "learning_rate": 0.0004650366583397342,
      "loss": 5.3375,
      "mean_token_accuracy": 0.17421140223741532,
      "num_tokens": 40995255.0,
      "step": 22225
    },
    {
      "entropy": 5.666626453399658,
      "epoch": 1.867632850241546,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004650206236872194,
      "loss": 5.3847,
      "mean_token_accuracy": 0.16714793890714646,
      "num_tokens": 41004419.0,
      "step": 22230
    },
    {
      "entropy": 5.511905288696289,
      "epoch": 1.8680529300567108,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004650045856685872,
      "loss": 5.1406,
      "mean_token_accuracy": 0.19823487997055053,
      "num_tokens": 41013179.0,
      "step": 22235
    },
    {
      "entropy": 5.555891275405884,
      "epoch": 1.8684730098718756,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00046498854428412157,
      "loss": 5.2309,
      "mean_token_accuracy": 0.17854980230331421,
      "num_tokens": 41022307.0,
      "step": 22240
    },
    {
      "entropy": 5.614151954650879,
      "epoch": 1.8688930896870404,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046497249953410675,
      "loss": 5.3878,
      "mean_token_accuracy": 0.17011225521564483,
      "num_tokens": 41032331.0,
      "step": 22245
    },
    {
      "entropy": 5.649067735671997,
      "epoch": 1.8693131695022054,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004649564514188269,
      "loss": 5.4098,
      "mean_token_accuracy": 0.17005167603492738,
      "num_tokens": 41041895.0,
      "step": 22250
    },
    {
      "entropy": 5.5712181568145756,
      "epoch": 1.8697332493173704,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004649403999385662,
      "loss": 5.2283,
      "mean_token_accuracy": 0.1801239386200905,
      "num_tokens": 41051643.0,
      "step": 22255
    },
    {
      "entropy": 5.592392826080323,
      "epoch": 1.8701533291325352,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004649243450936092,
      "loss": 5.2447,
      "mean_token_accuracy": 0.1818147912621498,
      "num_tokens": 41060478.0,
      "step": 22260
    },
    {
      "entropy": 5.599646282196045,
      "epoch": 1.8705734089477,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004649082868842403,
      "loss": 5.306,
      "mean_token_accuracy": 0.17611197531223297,
      "num_tokens": 41069389.0,
      "step": 22265
    },
    {
      "entropy": 5.496925687789917,
      "epoch": 1.870993488762865,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00046489222531074376,
      "loss": 5.263,
      "mean_token_accuracy": 0.18368336856365203,
      "num_tokens": 41078529.0,
      "step": 22270
    },
    {
      "entropy": 5.637415742874145,
      "epoch": 1.87141356857803,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046487616037340436,
      "loss": 5.367,
      "mean_token_accuracy": 0.17148163318634033,
      "num_tokens": 41087593.0,
      "step": 22275
    },
    {
      "entropy": 5.719729089736939,
      "epoch": 1.8718336483931948,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004648600920725065,
      "loss": 5.3344,
      "mean_token_accuracy": 0.1739830419421196,
      "num_tokens": 41098317.0,
      "step": 22280
    },
    {
      "entropy": 5.6388925075531,
      "epoch": 1.8722537282083596,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046484402040833486,
      "loss": 5.3309,
      "mean_token_accuracy": 0.1756902500987053,
      "num_tokens": 41108659.0,
      "step": 22285
    },
    {
      "entropy": 5.66907901763916,
      "epoch": 1.8726738080235243,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046482794538117413,
      "loss": 5.4085,
      "mean_token_accuracy": 0.1705798998475075,
      "num_tokens": 41117504.0,
      "step": 22290
    },
    {
      "entropy": 5.649161672592163,
      "epoch": 1.8730938878386894,
      "grad_norm": 1.375,
      "learning_rate": 0.00046481186699130913,
      "loss": 5.2902,
      "mean_token_accuracy": 0.17799595296382903,
      "num_tokens": 41126249.0,
      "step": 22295
    },
    {
      "entropy": 5.509735250473023,
      "epoch": 1.8735139676538544,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004647957852390247,
      "loss": 5.1802,
      "mean_token_accuracy": 0.17718643099069595,
      "num_tokens": 41134956.0,
      "step": 22300
    },
    {
      "entropy": 5.55515456199646,
      "epoch": 1.8739340474690191,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00046477970012460555,
      "loss": 5.2847,
      "mean_token_accuracy": 0.17597563564777374,
      "num_tokens": 41144340.0,
      "step": 22305
    },
    {
      "entropy": 5.544983386993408,
      "epoch": 1.874354127284184,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004647636116483367,
      "loss": 5.284,
      "mean_token_accuracy": 0.1800857350230217,
      "num_tokens": 41152937.0,
      "step": 22310
    },
    {
      "entropy": 5.692800140380859,
      "epoch": 1.8747742070993487,
      "grad_norm": 1.8125,
      "learning_rate": 0.00046474751981050334,
      "loss": 5.4637,
      "mean_token_accuracy": 0.16750447154045106,
      "num_tokens": 41162361.0,
      "step": 22315
    },
    {
      "entropy": 5.759031677246094,
      "epoch": 1.8751942869145137,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046473142461139034,
      "loss": 5.4421,
      "mean_token_accuracy": 0.16442356407642364,
      "num_tokens": 41171979.0,
      "step": 22320
    },
    {
      "entropy": 5.566287183761597,
      "epoch": 1.8756143667296787,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004647153260512828,
      "loss": 5.2963,
      "mean_token_accuracy": 0.18174279481172562,
      "num_tokens": 41182145.0,
      "step": 22325
    },
    {
      "entropy": 5.588994455337525,
      "epoch": 1.8760344465448435,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004646992241304659,
      "loss": 5.3106,
      "mean_token_accuracy": 0.17807296365499498,
      "num_tokens": 41191522.0,
      "step": 22330
    },
    {
      "entropy": 5.691400194168091,
      "epoch": 1.8764545263600083,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000464683118849225,
      "loss": 5.4092,
      "mean_token_accuracy": 0.17308613955974578,
      "num_tokens": 41201052.0,
      "step": 22335
    },
    {
      "entropy": 5.620829677581787,
      "epoch": 1.8768746061751733,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004646670102078453,
      "loss": 5.2867,
      "mean_token_accuracy": 0.17631472051143646,
      "num_tokens": 41210211.0,
      "step": 22340
    },
    {
      "entropy": 5.602048444747925,
      "epoch": 1.8772946859903383,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004646508982066122,
      "loss": 5.4207,
      "mean_token_accuracy": 0.17002287060022353,
      "num_tokens": 41219778.0,
      "step": 22345
    },
    {
      "entropy": 5.630302619934082,
      "epoch": 1.8777147658055031,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046463478284581114,
      "loss": 5.3967,
      "mean_token_accuracy": 0.17620732486248017,
      "num_tokens": 41229550.0,
      "step": 22350
    },
    {
      "entropy": 5.65893874168396,
      "epoch": 1.878134845620668,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004646186641257275,
      "loss": 5.2564,
      "mean_token_accuracy": 0.17994114756584167,
      "num_tokens": 41238130.0,
      "step": 22355
    },
    {
      "entropy": 5.588531064987182,
      "epoch": 1.8785549254358327,
      "grad_norm": 1.375,
      "learning_rate": 0.0004646025420466468,
      "loss": 5.2627,
      "mean_token_accuracy": 0.17994257509708406,
      "num_tokens": 41247324.0,
      "step": 22360
    },
    {
      "entropy": 5.548427629470825,
      "epoch": 1.8789750052509977,
      "grad_norm": 1.609375,
      "learning_rate": 0.00046458641660885474,
      "loss": 5.3512,
      "mean_token_accuracy": 0.17306893765926362,
      "num_tokens": 41256131.0,
      "step": 22365
    },
    {
      "entropy": 5.5881916046142575,
      "epoch": 1.8793950850661627,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046457028781263693,
      "loss": 5.376,
      "mean_token_accuracy": 0.169412861764431,
      "num_tokens": 41265225.0,
      "step": 22370
    },
    {
      "entropy": 5.69885573387146,
      "epoch": 1.8798151648813275,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046455415565827907,
      "loss": 5.3616,
      "mean_token_accuracy": 0.17275987863540648,
      "num_tokens": 41274023.0,
      "step": 22375
    },
    {
      "entropy": 5.68825888633728,
      "epoch": 1.8802352446964923,
      "grad_norm": 1.515625,
      "learning_rate": 0.000464538020146067,
      "loss": 5.3877,
      "mean_token_accuracy": 0.171722649037838,
      "num_tokens": 41283030.0,
      "step": 22380
    },
    {
      "entropy": 5.699346685409546,
      "epoch": 1.880655324511657,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004645218812762864,
      "loss": 5.4471,
      "mean_token_accuracy": 0.16928850412368773,
      "num_tokens": 41292654.0,
      "step": 22385
    },
    {
      "entropy": 5.599189758300781,
      "epoch": 1.881075404326822,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004645057390492234,
      "loss": 5.1952,
      "mean_token_accuracy": 0.1837023377418518,
      "num_tokens": 41301838.0,
      "step": 22390
    },
    {
      "entropy": 5.578370809555054,
      "epoch": 1.881495484141987,
      "grad_norm": 1.375,
      "learning_rate": 0.0004644895934651638,
      "loss": 5.3084,
      "mean_token_accuracy": 0.1745603010058403,
      "num_tokens": 41311104.0,
      "step": 22395
    },
    {
      "entropy": 5.682902097702026,
      "epoch": 1.8819155639571519,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046447344452439356,
      "loss": 5.3848,
      "mean_token_accuracy": 0.16891918182373047,
      "num_tokens": 41320213.0,
      "step": 22400
    },
    {
      "entropy": 5.570336532592774,
      "epoch": 1.8823356437723167,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004644572922271988,
      "loss": 5.2648,
      "mean_token_accuracy": 0.1755649209022522,
      "num_tokens": 41330027.0,
      "step": 22405
    },
    {
      "entropy": 5.609524965286255,
      "epoch": 1.8827557235874817,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046444113657386567,
      "loss": 5.3842,
      "mean_token_accuracy": 0.16911978274583817,
      "num_tokens": 41339481.0,
      "step": 22410
    },
    {
      "entropy": 5.68157730102539,
      "epoch": 1.8831758034026465,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046442497756468037,
      "loss": 5.4,
      "mean_token_accuracy": 0.16899559795856475,
      "num_tokens": 41348679.0,
      "step": 22415
    },
    {
      "entropy": 5.675786781311035,
      "epoch": 1.8835958832178115,
      "grad_norm": 1.8125,
      "learning_rate": 0.00046440881519992924,
      "loss": 5.2523,
      "mean_token_accuracy": 0.17877359986305236,
      "num_tokens": 41358736.0,
      "step": 22420
    },
    {
      "entropy": 5.675031614303589,
      "epoch": 1.8840159630329762,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004643926494798983,
      "loss": 5.3976,
      "mean_token_accuracy": 0.1677003726363182,
      "num_tokens": 41368284.0,
      "step": 22425
    },
    {
      "entropy": 5.579582262039184,
      "epoch": 1.884436042848141,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046437648040487426,
      "loss": 5.3127,
      "mean_token_accuracy": 0.1707193672657013,
      "num_tokens": 41377789.0,
      "step": 22430
    },
    {
      "entropy": 5.558114242553711,
      "epoch": 1.884856122663306,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046436030797514325,
      "loss": 5.3121,
      "mean_token_accuracy": 0.17467215210199355,
      "num_tokens": 41386909.0,
      "step": 22435
    },
    {
      "entropy": 5.652815866470337,
      "epoch": 1.885276202478471,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004643441321909919,
      "loss": 5.3307,
      "mean_token_accuracy": 0.17680877894163133,
      "num_tokens": 41396693.0,
      "step": 22440
    },
    {
      "entropy": 5.6828672885894775,
      "epoch": 1.8856962822936358,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046432795305270674,
      "loss": 5.4318,
      "mean_token_accuracy": 0.16189803034067154,
      "num_tokens": 41407193.0,
      "step": 22445
    },
    {
      "entropy": 5.6640712261199955,
      "epoch": 1.8861163621088006,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046431177056057446,
      "loss": 5.3643,
      "mean_token_accuracy": 0.17731093019247054,
      "num_tokens": 41416567.0,
      "step": 22450
    },
    {
      "entropy": 5.545360565185547,
      "epoch": 1.8865364419239654,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046429558471488164,
      "loss": 5.2488,
      "mean_token_accuracy": 0.17948656231164933,
      "num_tokens": 41425328.0,
      "step": 22455
    },
    {
      "entropy": 5.686221599578857,
      "epoch": 1.8869565217391304,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000464279395515915,
      "loss": 5.3906,
      "mean_token_accuracy": 0.17132183760404587,
      "num_tokens": 41435229.0,
      "step": 22460
    },
    {
      "entropy": 5.636958932876587,
      "epoch": 1.8873766015542954,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046426320296396136,
      "loss": 5.3208,
      "mean_token_accuracy": 0.1728391796350479,
      "num_tokens": 41445471.0,
      "step": 22465
    },
    {
      "entropy": 5.575035905838012,
      "epoch": 1.8877966813694602,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046424700705930745,
      "loss": 5.2426,
      "mean_token_accuracy": 0.18849639296531678,
      "num_tokens": 41454654.0,
      "step": 22470
    },
    {
      "entropy": 5.555283164978027,
      "epoch": 1.888216761184625,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004642308078022403,
      "loss": 5.2523,
      "mean_token_accuracy": 0.17713084071874619,
      "num_tokens": 41463341.0,
      "step": 22475
    },
    {
      "entropy": 5.581478595733643,
      "epoch": 1.88863684099979,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046421460519304684,
      "loss": 5.2937,
      "mean_token_accuracy": 0.17563444674015044,
      "num_tokens": 41472677.0,
      "step": 22480
    },
    {
      "entropy": 5.666422128677368,
      "epoch": 1.8890569208149548,
      "grad_norm": 1.328125,
      "learning_rate": 0.000464198399232014,
      "loss": 5.4289,
      "mean_token_accuracy": 0.16743830740451812,
      "num_tokens": 41482867.0,
      "step": 22485
    },
    {
      "entropy": 5.726667976379394,
      "epoch": 1.8894770006301198,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004641821899194291,
      "loss": 5.3803,
      "mean_token_accuracy": 0.1680409014225006,
      "num_tokens": 41493432.0,
      "step": 22490
    },
    {
      "entropy": 5.745121717453003,
      "epoch": 1.8898970804452846,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046416597725557903,
      "loss": 5.4058,
      "mean_token_accuracy": 0.16858120858669282,
      "num_tokens": 41503807.0,
      "step": 22495
    },
    {
      "entropy": 5.6267173290252686,
      "epoch": 1.8903171602604494,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000464149761240751,
      "loss": 5.3085,
      "mean_token_accuracy": 0.17716021090745926,
      "num_tokens": 41512524.0,
      "step": 22500
    },
    {
      "entropy": 5.633813524246216,
      "epoch": 1.8907372400756144,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046413354187523244,
      "loss": 5.4094,
      "mean_token_accuracy": 0.16981790065765381,
      "num_tokens": 41521915.0,
      "step": 22505
    },
    {
      "entropy": 5.596736764907837,
      "epoch": 1.8911573198907794,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004641173191593105,
      "loss": 5.336,
      "mean_token_accuracy": 0.17258377075195314,
      "num_tokens": 41530293.0,
      "step": 22510
    },
    {
      "entropy": 5.640922927856446,
      "epoch": 1.8915773997059442,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046410109309327275,
      "loss": 5.3898,
      "mean_token_accuracy": 0.171537646651268,
      "num_tokens": 41538660.0,
      "step": 22515
    },
    {
      "entropy": 5.6202301502227785,
      "epoch": 1.891997479521109,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046408486367740647,
      "loss": 5.3641,
      "mean_token_accuracy": 0.18050749897956847,
      "num_tokens": 41547952.0,
      "step": 22520
    },
    {
      "entropy": 5.632935285568237,
      "epoch": 1.8924175593362738,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004640686309119992,
      "loss": 5.286,
      "mean_token_accuracy": 0.18273382484912873,
      "num_tokens": 41557093.0,
      "step": 22525
    },
    {
      "entropy": 5.6084208488464355,
      "epoch": 1.8928376391514388,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046405239479733844,
      "loss": 5.3019,
      "mean_token_accuracy": 0.17673699259757997,
      "num_tokens": 41565836.0,
      "step": 22530
    },
    {
      "entropy": 5.5692564964294435,
      "epoch": 1.8932577189666038,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004640361553337119,
      "loss": 5.3532,
      "mean_token_accuracy": 0.1848425567150116,
      "num_tokens": 41575365.0,
      "step": 22535
    },
    {
      "entropy": 5.591070652008057,
      "epoch": 1.8936777987817686,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046401991252140715,
      "loss": 5.3199,
      "mean_token_accuracy": 0.17346022874116898,
      "num_tokens": 41583690.0,
      "step": 22540
    },
    {
      "entropy": 5.681325674057007,
      "epoch": 1.8940978785969333,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000464003666360712,
      "loss": 5.2973,
      "mean_token_accuracy": 0.17554327994585037,
      "num_tokens": 41593536.0,
      "step": 22545
    },
    {
      "entropy": 5.622394227981568,
      "epoch": 1.8945179584120981,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004639874168519143,
      "loss": 5.2714,
      "mean_token_accuracy": 0.17500520646572112,
      "num_tokens": 41602543.0,
      "step": 22550
    },
    {
      "entropy": 5.586856508255005,
      "epoch": 1.8949380382272631,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004639711639953017,
      "loss": 5.3708,
      "mean_token_accuracy": 0.17548366636037827,
      "num_tokens": 41611634.0,
      "step": 22555
    },
    {
      "entropy": 5.55074553489685,
      "epoch": 1.8953581180424282,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004639549077911623,
      "loss": 5.3129,
      "mean_token_accuracy": 0.1746702238917351,
      "num_tokens": 41621400.0,
      "step": 22560
    },
    {
      "entropy": 5.614761638641357,
      "epoch": 1.895778197857593,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046393864823978406,
      "loss": 5.3009,
      "mean_token_accuracy": 0.1818379744887352,
      "num_tokens": 41631070.0,
      "step": 22565
    },
    {
      "entropy": 5.714692258834839,
      "epoch": 1.8961982776727577,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004639223853414549,
      "loss": 5.378,
      "mean_token_accuracy": 0.17304340600967408,
      "num_tokens": 41641442.0,
      "step": 22570
    },
    {
      "entropy": 5.669845771789551,
      "epoch": 1.8966183574879227,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000463906119096463,
      "loss": 5.3981,
      "mean_token_accuracy": 0.17233923375606536,
      "num_tokens": 41651616.0,
      "step": 22575
    },
    {
      "entropy": 5.671889591217041,
      "epoch": 1.8970384373030877,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004638898495050963,
      "loss": 5.2985,
      "mean_token_accuracy": 0.1773570492863655,
      "num_tokens": 41660704.0,
      "step": 22580
    },
    {
      "entropy": 5.574475049972534,
      "epoch": 1.8974585171182525,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004638735765676434,
      "loss": 5.37,
      "mean_token_accuracy": 0.17592561095952988,
      "num_tokens": 41669824.0,
      "step": 22585
    },
    {
      "entropy": 5.585171318054199,
      "epoch": 1.8978785969334173,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004638573002843922,
      "loss": 5.3074,
      "mean_token_accuracy": 0.1802604079246521,
      "num_tokens": 41680082.0,
      "step": 22590
    },
    {
      "entropy": 5.607627630233765,
      "epoch": 1.898298676748582,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004638410206556312,
      "loss": 5.2507,
      "mean_token_accuracy": 0.1771162837743759,
      "num_tokens": 41689282.0,
      "step": 22595
    },
    {
      "entropy": 5.635611963272095,
      "epoch": 1.8987187565637471,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004638247376816489,
      "loss": 5.3783,
      "mean_token_accuracy": 0.17096421867609024,
      "num_tokens": 41699059.0,
      "step": 22600
    },
    {
      "entropy": 5.723477602005005,
      "epoch": 1.8991388363789121,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004638084513627335,
      "loss": 5.4861,
      "mean_token_accuracy": 0.1670008048415184,
      "num_tokens": 41708674.0,
      "step": 22605
    },
    {
      "entropy": 5.6795759201049805,
      "epoch": 1.899558916194077,
      "grad_norm": 1.5,
      "learning_rate": 0.00046379216169917356,
      "loss": 5.3728,
      "mean_token_accuracy": 0.1751355528831482,
      "num_tokens": 41718418.0,
      "step": 22610
    },
    {
      "entropy": 5.622887325286865,
      "epoch": 1.8999789960092417,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004637758686912577,
      "loss": 5.3792,
      "mean_token_accuracy": 0.16746392101049423,
      "num_tokens": 41728229.0,
      "step": 22615
    },
    {
      "entropy": 5.591051912307739,
      "epoch": 1.9003990758244065,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046375957233927456,
      "loss": 5.3332,
      "mean_token_accuracy": 0.1768563762307167,
      "num_tokens": 41737074.0,
      "step": 22620
    },
    {
      "entropy": 5.639246606826783,
      "epoch": 1.9008191556395715,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046374327264351277,
      "loss": 5.2258,
      "mean_token_accuracy": 0.18000006675720215,
      "num_tokens": 41745823.0,
      "step": 22625
    },
    {
      "entropy": 5.580969619750976,
      "epoch": 1.9012392354547365,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046372696960426116,
      "loss": 5.3116,
      "mean_token_accuracy": 0.18088625371456146,
      "num_tokens": 41754591.0,
      "step": 22630
    },
    {
      "entropy": 5.626723051071167,
      "epoch": 1.9016593152699013,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046371066322180846,
      "loss": 5.3394,
      "mean_token_accuracy": 0.1772809773683548,
      "num_tokens": 41763585.0,
      "step": 22635
    },
    {
      "entropy": 5.6320226192474365,
      "epoch": 1.902079395085066,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046369435349644344,
      "loss": 5.3569,
      "mean_token_accuracy": 0.1758773297071457,
      "num_tokens": 41772712.0,
      "step": 22640
    },
    {
      "entropy": 5.636237001419067,
      "epoch": 1.902499474900231,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046367804042845515,
      "loss": 5.2304,
      "mean_token_accuracy": 0.18543338179588317,
      "num_tokens": 41781516.0,
      "step": 22645
    },
    {
      "entropy": 5.6099756240844725,
      "epoch": 1.902919554715396,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046366172401813253,
      "loss": 5.3328,
      "mean_token_accuracy": 0.1702340930700302,
      "num_tokens": 41790731.0,
      "step": 22650
    },
    {
      "entropy": 5.643015718460083,
      "epoch": 1.9033396345305609,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004636454042657647,
      "loss": 5.3307,
      "mean_token_accuracy": 0.1741826504468918,
      "num_tokens": 41799654.0,
      "step": 22655
    },
    {
      "entropy": 5.541084480285645,
      "epoch": 1.9037597143457257,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046362908117164055,
      "loss": 5.2316,
      "mean_token_accuracy": 0.17913161367177963,
      "num_tokens": 41809408.0,
      "step": 22660
    },
    {
      "entropy": 5.602960395812988,
      "epoch": 1.9041797941608904,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004636127547360494,
      "loss": 5.3884,
      "mean_token_accuracy": 0.17082755416631698,
      "num_tokens": 41818868.0,
      "step": 22665
    },
    {
      "entropy": 5.605136156082153,
      "epoch": 1.9045998739760555,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004635964249592804,
      "loss": 5.3031,
      "mean_token_accuracy": 0.1818327710032463,
      "num_tokens": 41827156.0,
      "step": 22670
    },
    {
      "entropy": 5.699454736709595,
      "epoch": 1.9050199537912205,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004635800918416229,
      "loss": 5.401,
      "mean_token_accuracy": 0.16892854273319244,
      "num_tokens": 41837025.0,
      "step": 22675
    },
    {
      "entropy": 5.742538833618164,
      "epoch": 1.9054400336063853,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046356375538336616,
      "loss": 5.3211,
      "mean_token_accuracy": 0.1761800616979599,
      "num_tokens": 41846196.0,
      "step": 22680
    },
    {
      "entropy": 5.529394149780273,
      "epoch": 1.90586011342155,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046354741558479956,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17033856213092805,
      "num_tokens": 41855030.0,
      "step": 22685
    },
    {
      "entropy": 5.543329429626465,
      "epoch": 1.9062801932367148,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004635310724462126,
      "loss": 5.188,
      "mean_token_accuracy": 0.1789797842502594,
      "num_tokens": 41863740.0,
      "step": 22690
    },
    {
      "entropy": 5.535472774505616,
      "epoch": 1.9067002730518798,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004635147259678948,
      "loss": 5.3335,
      "mean_token_accuracy": 0.1750457227230072,
      "num_tokens": 41873376.0,
      "step": 22695
    },
    {
      "entropy": 5.677917432785034,
      "epoch": 1.9071203528670448,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046349837615013563,
      "loss": 5.4281,
      "mean_token_accuracy": 0.1685136914253235,
      "num_tokens": 41882491.0,
      "step": 22700
    },
    {
      "entropy": 5.6457596778869625,
      "epoch": 1.9075404326822096,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004634820229932248,
      "loss": 5.3009,
      "mean_token_accuracy": 0.17515089362859726,
      "num_tokens": 41891357.0,
      "step": 22705
    },
    {
      "entropy": 5.6087073802948,
      "epoch": 1.9079605124973744,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046346566649745205,
      "loss": 5.3492,
      "mean_token_accuracy": 0.17544787973165513,
      "num_tokens": 41899874.0,
      "step": 22710
    },
    {
      "entropy": 5.65878267288208,
      "epoch": 1.9083805923125394,
      "grad_norm": 1.53125,
      "learning_rate": 0.000463449306663107,
      "loss": 5.3578,
      "mean_token_accuracy": 0.17512507140636444,
      "num_tokens": 41909673.0,
      "step": 22715
    },
    {
      "entropy": 5.689415645599365,
      "epoch": 1.9088006721277042,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004634329434904796,
      "loss": 5.4043,
      "mean_token_accuracy": 0.17174190282821655,
      "num_tokens": 41919126.0,
      "step": 22720
    },
    {
      "entropy": 5.55366153717041,
      "epoch": 1.9092207519428692,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004634165769798596,
      "loss": 5.2367,
      "mean_token_accuracy": 0.17866196781396865,
      "num_tokens": 41927751.0,
      "step": 22725
    },
    {
      "entropy": 5.617975664138794,
      "epoch": 1.909640831758034,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004634002071315369,
      "loss": 5.3131,
      "mean_token_accuracy": 0.17952744513750077,
      "num_tokens": 41937290.0,
      "step": 22730
    },
    {
      "entropy": 5.611281299591065,
      "epoch": 1.9100609115731988,
      "grad_norm": 1.75,
      "learning_rate": 0.00046338383394580157,
      "loss": 5.2728,
      "mean_token_accuracy": 0.18018366992473603,
      "num_tokens": 41947186.0,
      "step": 22735
    },
    {
      "entropy": 5.579268455505371,
      "epoch": 1.9104809913883638,
      "grad_norm": 1.375,
      "learning_rate": 0.00046336745742294366,
      "loss": 5.2785,
      "mean_token_accuracy": 0.1728681966662407,
      "num_tokens": 41956197.0,
      "step": 22740
    },
    {
      "entropy": 5.6366523742675785,
      "epoch": 1.9109010712035288,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046335107756325316,
      "loss": 5.2734,
      "mean_token_accuracy": 0.17634716629981995,
      "num_tokens": 41965881.0,
      "step": 22745
    },
    {
      "entropy": 5.651572942733765,
      "epoch": 1.9113211510186936,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004633346943670204,
      "loss": 5.3545,
      "mean_token_accuracy": 0.1676716163754463,
      "num_tokens": 41975031.0,
      "step": 22750
    },
    {
      "entropy": 5.640260982513428,
      "epoch": 1.9117412308338584,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004633183078345355,
      "loss": 5.2904,
      "mean_token_accuracy": 0.1763753116130829,
      "num_tokens": 41984187.0,
      "step": 22755
    },
    {
      "entropy": 5.6649223327636715,
      "epoch": 1.9121613106490232,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046330191796608867,
      "loss": 5.4045,
      "mean_token_accuracy": 0.17316106408834459,
      "num_tokens": 41993185.0,
      "step": 22760
    },
    {
      "entropy": 5.6566869735717775,
      "epoch": 1.9125813904641882,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004632855247619704,
      "loss": 5.3462,
      "mean_token_accuracy": 0.18005757331848143,
      "num_tokens": 42002521.0,
      "step": 22765
    },
    {
      "entropy": 5.679284381866455,
      "epoch": 1.9130014702793532,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000463269128222471,
      "loss": 5.4417,
      "mean_token_accuracy": 0.17086408883333207,
      "num_tokens": 42011444.0,
      "step": 22770
    },
    {
      "entropy": 5.67745451927185,
      "epoch": 1.913421550094518,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004632527283478809,
      "loss": 5.3579,
      "mean_token_accuracy": 0.17271676063537597,
      "num_tokens": 42020916.0,
      "step": 22775
    },
    {
      "entropy": 5.72650179862976,
      "epoch": 1.9138416299096828,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046323632513849063,
      "loss": 5.3896,
      "mean_token_accuracy": 0.17431481182575226,
      "num_tokens": 42029467.0,
      "step": 22780
    },
    {
      "entropy": 5.487216663360596,
      "epoch": 1.9142617097248478,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004632199185945908,
      "loss": 5.1377,
      "mean_token_accuracy": 0.18585379868745805,
      "num_tokens": 42037435.0,
      "step": 22785
    },
    {
      "entropy": 5.657347822189331,
      "epoch": 1.9146817895400126,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004632035087164721,
      "loss": 5.3651,
      "mean_token_accuracy": 0.1772521071135998,
      "num_tokens": 42046943.0,
      "step": 22790
    },
    {
      "entropy": 5.611826705932617,
      "epoch": 1.9151018693551776,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004631870955044251,
      "loss": 5.2533,
      "mean_token_accuracy": 0.18375349789857864,
      "num_tokens": 42055804.0,
      "step": 22795
    },
    {
      "entropy": 5.612897062301636,
      "epoch": 1.9155219491703424,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046317067895874063,
      "loss": 5.2595,
      "mean_token_accuracy": 0.18950011879205703,
      "num_tokens": 42064655.0,
      "step": 22800
    },
    {
      "entropy": 5.662775278091431,
      "epoch": 1.9159420289855071,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046315425907970947,
      "loss": 5.3179,
      "mean_token_accuracy": 0.1727246731519699,
      "num_tokens": 42073663.0,
      "step": 22805
    },
    {
      "entropy": 5.655098152160645,
      "epoch": 1.9163621088006721,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004631378358676225,
      "loss": 5.3756,
      "mean_token_accuracy": 0.1802870512008667,
      "num_tokens": 42083931.0,
      "step": 22810
    },
    {
      "entropy": 5.6888645648956295,
      "epoch": 1.9167821886158372,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004631214093227706,
      "loss": 5.362,
      "mean_token_accuracy": 0.17064680755138398,
      "num_tokens": 42093782.0,
      "step": 22815
    },
    {
      "entropy": 5.663200759887696,
      "epoch": 1.917202268431002,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004631049794454448,
      "loss": 5.3207,
      "mean_token_accuracy": 0.17533529698848724,
      "num_tokens": 42103392.0,
      "step": 22820
    },
    {
      "entropy": 5.671980714797973,
      "epoch": 1.9176223482461667,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004630885462359362,
      "loss": 5.2579,
      "mean_token_accuracy": 0.18281664997339248,
      "num_tokens": 42112051.0,
      "step": 22825
    },
    {
      "entropy": 5.540486764907837,
      "epoch": 1.9180424280613315,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004630721096945358,
      "loss": 5.2966,
      "mean_token_accuracy": 0.17852336168289185,
      "num_tokens": 42120156.0,
      "step": 22830
    },
    {
      "entropy": 5.63999605178833,
      "epoch": 1.9184625078764965,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004630556698215349,
      "loss": 5.3906,
      "mean_token_accuracy": 0.18401091694831848,
      "num_tokens": 42129564.0,
      "step": 22835
    },
    {
      "entropy": 5.727736186981201,
      "epoch": 1.9188825876916615,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046303922661722466,
      "loss": 5.4313,
      "mean_token_accuracy": 0.16911460608243942,
      "num_tokens": 42138144.0,
      "step": 22840
    },
    {
      "entropy": 5.636684226989746,
      "epoch": 1.9193026675068263,
      "grad_norm": 1.375,
      "learning_rate": 0.00046302278008189627,
      "loss": 5.2797,
      "mean_token_accuracy": 0.1710377350449562,
      "num_tokens": 42147701.0,
      "step": 22845
    },
    {
      "entropy": 5.5380760669708256,
      "epoch": 1.919722747321991,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004630063302158412,
      "loss": 5.2335,
      "mean_token_accuracy": 0.18256814032793045,
      "num_tokens": 42156772.0,
      "step": 22850
    },
    {
      "entropy": 5.552340364456176,
      "epoch": 1.920142827137156,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046298987701935066,
      "loss": 5.1744,
      "mean_token_accuracy": 0.18554712682962418,
      "num_tokens": 42165227.0,
      "step": 22855
    },
    {
      "entropy": 5.538019847869873,
      "epoch": 1.920562906952321,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004629734204927164,
      "loss": 5.2352,
      "mean_token_accuracy": 0.18170234560966492,
      "num_tokens": 42174800.0,
      "step": 22860
    },
    {
      "entropy": 5.61850528717041,
      "epoch": 1.920982986767486,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004629569606362298,
      "loss": 5.2904,
      "mean_token_accuracy": 0.17445085048675538,
      "num_tokens": 42184301.0,
      "step": 22865
    },
    {
      "entropy": 5.669399261474609,
      "epoch": 1.9214030665826507,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004629404974501823,
      "loss": 5.3126,
      "mean_token_accuracy": 0.1747346490621567,
      "num_tokens": 42193266.0,
      "step": 22870
    },
    {
      "entropy": 5.583701515197754,
      "epoch": 1.9218231463978155,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004629240309348658,
      "loss": 5.2488,
      "mean_token_accuracy": 0.17563164234161377,
      "num_tokens": 42202051.0,
      "step": 22875
    },
    {
      "entropy": 5.5733880519866945,
      "epoch": 1.9222432262129805,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004629075610905717,
      "loss": 5.1597,
      "mean_token_accuracy": 0.18898734897375108,
      "num_tokens": 42210716.0,
      "step": 22880
    },
    {
      "entropy": 5.51941614151001,
      "epoch": 1.9226633060281455,
      "grad_norm": 1.1875,
      "learning_rate": 0.000462891087917592,
      "loss": 5.2403,
      "mean_token_accuracy": 0.17882364243268967,
      "num_tokens": 42219930.0,
      "step": 22885
    },
    {
      "entropy": 5.605342102050781,
      "epoch": 1.9230833858433103,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046287461141621844,
      "loss": 5.2816,
      "mean_token_accuracy": 0.18385231494903564,
      "num_tokens": 42228864.0,
      "step": 22890
    },
    {
      "entropy": 5.589842081069946,
      "epoch": 1.923503465658475,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004628581315867429,
      "loss": 5.3448,
      "mean_token_accuracy": 0.17380790710449218,
      "num_tokens": 42238030.0,
      "step": 22895
    },
    {
      "entropy": 5.660066270828247,
      "epoch": 1.9239235454736399,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00046284164842945723,
      "loss": 5.3266,
      "mean_token_accuracy": 0.17029749155044555,
      "num_tokens": 42247818.0,
      "step": 22900
    },
    {
      "entropy": 5.6344915390014645,
      "epoch": 1.9243436252888049,
      "grad_norm": 1.375,
      "learning_rate": 0.0004628251619446536,
      "loss": 5.2761,
      "mean_token_accuracy": 0.17787475734949112,
      "num_tokens": 42256772.0,
      "step": 22905
    },
    {
      "entropy": 5.545624780654907,
      "epoch": 1.9247637051039699,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046280867213262385,
      "loss": 5.3281,
      "mean_token_accuracy": 0.16812801659107207,
      "num_tokens": 42265620.0,
      "step": 22910
    },
    {
      "entropy": 5.668614768981934,
      "epoch": 1.9251837849191347,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004627921789936602,
      "loss": 5.4026,
      "mean_token_accuracy": 0.16873037666082383,
      "num_tokens": 42274998.0,
      "step": 22915
    },
    {
      "entropy": 5.711544752120972,
      "epoch": 1.9256038647342995,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046277568252805476,
      "loss": 5.3343,
      "mean_token_accuracy": 0.17508528381586075,
      "num_tokens": 42284849.0,
      "step": 22920
    },
    {
      "entropy": 5.571464967727661,
      "epoch": 1.9260239445494642,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004627591827360998,
      "loss": 5.3178,
      "mean_token_accuracy": 0.1741255983710289,
      "num_tokens": 42294133.0,
      "step": 22925
    },
    {
      "entropy": 5.595095205307007,
      "epoch": 1.9264440243646292,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004627426796180876,
      "loss": 5.2886,
      "mean_token_accuracy": 0.17940344363451005,
      "num_tokens": 42302765.0,
      "step": 22930
    },
    {
      "entropy": 5.625879144668579,
      "epoch": 1.9268641041797943,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046272617317431056,
      "loss": 5.2584,
      "mean_token_accuracy": 0.1767005905508995,
      "num_tokens": 42311829.0,
      "step": 22935
    },
    {
      "entropy": 5.687358283996582,
      "epoch": 1.927284183994959,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046270966340506087,
      "loss": 5.3961,
      "mean_token_accuracy": 0.181781767308712,
      "num_tokens": 42321294.0,
      "step": 22940
    },
    {
      "entropy": 5.610746717453003,
      "epoch": 1.9277042638101238,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046269315031063137,
      "loss": 5.2101,
      "mean_token_accuracy": 0.1810872346162796,
      "num_tokens": 42329379.0,
      "step": 22945
    },
    {
      "entropy": 5.655484533309936,
      "epoch": 1.9281243436252888,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046267663389131425,
      "loss": 5.3994,
      "mean_token_accuracy": 0.1621880903840065,
      "num_tokens": 42339867.0,
      "step": 22950
    },
    {
      "entropy": 5.680404472351074,
      "epoch": 1.9285444234404538,
      "grad_norm": 2.546875,
      "learning_rate": 0.00046266011414740213,
      "loss": 5.4098,
      "mean_token_accuracy": 0.1692059487104416,
      "num_tokens": 42350174.0,
      "step": 22955
    },
    {
      "entropy": 5.59824571609497,
      "epoch": 1.9289645032556186,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004626435910791878,
      "loss": 5.2684,
      "mean_token_accuracy": 0.17532526701688766,
      "num_tokens": 42359214.0,
      "step": 22960
    },
    {
      "entropy": 5.586638927459717,
      "epoch": 1.9293845830707834,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046262706468696386,
      "loss": 5.336,
      "mean_token_accuracy": 0.17176856845617294,
      "num_tokens": 42367965.0,
      "step": 22965
    },
    {
      "entropy": 5.5947588920593265,
      "epoch": 1.9298046628859482,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004626105349710231,
      "loss": 5.3466,
      "mean_token_accuracy": 0.1715581923723221,
      "num_tokens": 42377233.0,
      "step": 22970
    },
    {
      "entropy": 5.765683174133301,
      "epoch": 1.9302247427011132,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004625940019316584,
      "loss": 5.3619,
      "mean_token_accuracy": 0.17158213257789612,
      "num_tokens": 42386060.0,
      "step": 22975
    },
    {
      "entropy": 5.691909599304199,
      "epoch": 1.9306448225162782,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00046257746556916236,
      "loss": 5.354,
      "mean_token_accuracy": 0.17822704017162322,
      "num_tokens": 42395659.0,
      "step": 22980
    },
    {
      "entropy": 5.679384422302246,
      "epoch": 1.931064902331443,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046256092588382825,
      "loss": 5.3614,
      "mean_token_accuracy": 0.17406841814517976,
      "num_tokens": 42403531.0,
      "step": 22985
    },
    {
      "entropy": 5.629655504226685,
      "epoch": 1.9314849821466078,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046254438287594884,
      "loss": 5.3241,
      "mean_token_accuracy": 0.17861594408750534,
      "num_tokens": 42412364.0,
      "step": 22990
    },
    {
      "entropy": 5.561359930038452,
      "epoch": 1.9319050619617726,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046252783654581733,
      "loss": 5.3111,
      "mean_token_accuracy": 0.17233571112155915,
      "num_tokens": 42422276.0,
      "step": 22995
    },
    {
      "entropy": 5.69279670715332,
      "epoch": 1.9323251417769376,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004625112868937267,
      "loss": 5.3278,
      "mean_token_accuracy": 0.17849384993314743,
      "num_tokens": 42430853.0,
      "step": 23000
    },
    {
      "entropy": 5.556269121170044,
      "epoch": 1.9327452215921026,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004624947339199702,
      "loss": 5.2115,
      "mean_token_accuracy": 0.18026719242334366,
      "num_tokens": 42439034.0,
      "step": 23005
    },
    {
      "entropy": 5.640739822387696,
      "epoch": 1.9331653014072674,
      "grad_norm": 1.21875,
      "learning_rate": 0.000462478177624841,
      "loss": 5.3957,
      "mean_token_accuracy": 0.17255858480930328,
      "num_tokens": 42448494.0,
      "step": 23010
    },
    {
      "entropy": 5.715470886230468,
      "epoch": 1.9335853812224322,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046246161800863244,
      "loss": 5.3013,
      "mean_token_accuracy": 0.17863426506519317,
      "num_tokens": 42457188.0,
      "step": 23015
    },
    {
      "entropy": 5.610437822341919,
      "epoch": 1.9340054610375972,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004624450550716379,
      "loss": 5.3791,
      "mean_token_accuracy": 0.17333876043558122,
      "num_tokens": 42466321.0,
      "step": 23020
    },
    {
      "entropy": 5.626680183410644,
      "epoch": 1.934425540852762,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004624284888141507,
      "loss": 5.3274,
      "mean_token_accuracy": 0.17454350292682647,
      "num_tokens": 42475879.0,
      "step": 23025
    },
    {
      "entropy": 5.684919118881226,
      "epoch": 1.934845620667927,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004624119192364643,
      "loss": 5.4693,
      "mean_token_accuracy": 0.17058544009923934,
      "num_tokens": 42484988.0,
      "step": 23030
    },
    {
      "entropy": 5.639768934249878,
      "epoch": 1.9352657004830918,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046239534633887223,
      "loss": 5.2572,
      "mean_token_accuracy": 0.17340181916952133,
      "num_tokens": 42493764.0,
      "step": 23035
    },
    {
      "entropy": 5.75081877708435,
      "epoch": 1.9356857802982566,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004623787701216682,
      "loss": 5.4831,
      "mean_token_accuracy": 0.17213430851697922,
      "num_tokens": 42503312.0,
      "step": 23040
    },
    {
      "entropy": 5.624970293045044,
      "epoch": 1.9361058601134216,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046236219058514566,
      "loss": 5.327,
      "mean_token_accuracy": 0.17737353593111038,
      "num_tokens": 42512303.0,
      "step": 23045
    },
    {
      "entropy": 5.5386903285980225,
      "epoch": 1.9365259399285866,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004623456077295984,
      "loss": 5.2048,
      "mean_token_accuracy": 0.18753930926322937,
      "num_tokens": 42520928.0,
      "step": 23050
    },
    {
      "entropy": 5.584413862228393,
      "epoch": 1.9369460197437514,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004623290215553201,
      "loss": 5.2527,
      "mean_token_accuracy": 0.186776827275753,
      "num_tokens": 42529945.0,
      "step": 23055
    },
    {
      "entropy": 5.608163118362427,
      "epoch": 1.9373660995589161,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004623124320626048,
      "loss": 5.3163,
      "mean_token_accuracy": 0.1753854662179947,
      "num_tokens": 42539078.0,
      "step": 23060
    },
    {
      "entropy": 5.599880838394165,
      "epoch": 1.937786179374081,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004622958392517461,
      "loss": 5.2736,
      "mean_token_accuracy": 0.17854499071836472,
      "num_tokens": 42547842.0,
      "step": 23065
    },
    {
      "entropy": 5.576862001419068,
      "epoch": 1.938206259189246,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004622792431230381,
      "loss": 5.2194,
      "mean_token_accuracy": 0.18754540979862214,
      "num_tokens": 42556574.0,
      "step": 23070
    },
    {
      "entropy": 5.648509311676025,
      "epoch": 1.938626339004411,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046226264367677476,
      "loss": 5.3269,
      "mean_token_accuracy": 0.16598548144102096,
      "num_tokens": 42565906.0,
      "step": 23075
    },
    {
      "entropy": 5.610691165924072,
      "epoch": 1.9390464188195757,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004622460409132501,
      "loss": 5.2887,
      "mean_token_accuracy": 0.1836601510643959,
      "num_tokens": 42574929.0,
      "step": 23080
    },
    {
      "entropy": 5.639229393005371,
      "epoch": 1.9394664986347405,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004622294348327582,
      "loss": 5.3275,
      "mean_token_accuracy": 0.16833080351352692,
      "num_tokens": 42585185.0,
      "step": 23085
    },
    {
      "entropy": 5.597856187820435,
      "epoch": 1.9398865784499055,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00046221282543559334,
      "loss": 5.3498,
      "mean_token_accuracy": 0.17007184028625488,
      "num_tokens": 42594272.0,
      "step": 23090
    },
    {
      "entropy": 5.558249568939209,
      "epoch": 1.9403066582650703,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046219621272204967,
      "loss": 5.257,
      "mean_token_accuracy": 0.17922381311655045,
      "num_tokens": 42603410.0,
      "step": 23095
    },
    {
      "entropy": 5.566586065292358,
      "epoch": 1.9407267380802353,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046217959669242145,
      "loss": 5.4705,
      "mean_token_accuracy": 0.1623099833726883,
      "num_tokens": 42613879.0,
      "step": 23100
    },
    {
      "entropy": 5.663104724884033,
      "epoch": 1.9411468178954001,
      "grad_norm": 1.546875,
      "learning_rate": 0.000462162977347003,
      "loss": 5.2523,
      "mean_token_accuracy": 0.1779240071773529,
      "num_tokens": 42623323.0,
      "step": 23105
    },
    {
      "entropy": 5.68038911819458,
      "epoch": 1.941566897710565,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046214635468608885,
      "loss": 5.3075,
      "mean_token_accuracy": 0.1784870833158493,
      "num_tokens": 42632365.0,
      "step": 23110
    },
    {
      "entropy": 5.594471216201782,
      "epoch": 1.94198697752573,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046212972870997336,
      "loss": 5.276,
      "mean_token_accuracy": 0.1815234288573265,
      "num_tokens": 42641872.0,
      "step": 23115
    },
    {
      "entropy": 5.652111959457398,
      "epoch": 1.942407057340895,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004621130994189511,
      "loss": 5.351,
      "mean_token_accuracy": 0.17740897983312606,
      "num_tokens": 42652031.0,
      "step": 23120
    },
    {
      "entropy": 5.580147981643677,
      "epoch": 1.9428271371560597,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004620964668133166,
      "loss": 5.2775,
      "mean_token_accuracy": 0.1754400461912155,
      "num_tokens": 42661040.0,
      "step": 23125
    },
    {
      "entropy": 5.634384346008301,
      "epoch": 1.9432472169712245,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004620798308933646,
      "loss": 5.3154,
      "mean_token_accuracy": 0.17391279190778733,
      "num_tokens": 42670559.0,
      "step": 23130
    },
    {
      "entropy": 5.6503113269805905,
      "epoch": 1.9436672967863893,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004620631916593897,
      "loss": 5.33,
      "mean_token_accuracy": 0.17253556698560715,
      "num_tokens": 42679883.0,
      "step": 23135
    },
    {
      "entropy": 5.738234233856201,
      "epoch": 1.9440873766015543,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004620465491116867,
      "loss": 5.4438,
      "mean_token_accuracy": 0.16332100480794906,
      "num_tokens": 42689746.0,
      "step": 23140
    },
    {
      "entropy": 5.706368541717529,
      "epoch": 1.9445074564167193,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046202990325055034,
      "loss": 5.3664,
      "mean_token_accuracy": 0.1664703294634819,
      "num_tokens": 42699685.0,
      "step": 23145
    },
    {
      "entropy": 5.569720458984375,
      "epoch": 1.944927536231884,
      "grad_norm": 1.375,
      "learning_rate": 0.0004620132540762756,
      "loss": 5.2138,
      "mean_token_accuracy": 0.17890601605176926,
      "num_tokens": 42708873.0,
      "step": 23150
    },
    {
      "entropy": 5.562334966659546,
      "epoch": 1.9453476160470489,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046199660158915734,
      "loss": 5.2845,
      "mean_token_accuracy": 0.1692815437912941,
      "num_tokens": 42717807.0,
      "step": 23155
    },
    {
      "entropy": 5.59727029800415,
      "epoch": 1.9457676958622139,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046197994578949056,
      "loss": 5.3582,
      "mean_token_accuracy": 0.17602658569812774,
      "num_tokens": 42726674.0,
      "step": 23160
    },
    {
      "entropy": 5.657729721069336,
      "epoch": 1.9461877756773787,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004619632866775704,
      "loss": 5.4231,
      "mean_token_accuracy": 0.175442436337471,
      "num_tokens": 42735621.0,
      "step": 23165
    },
    {
      "entropy": 5.604097843170166,
      "epoch": 1.9466078554925437,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004619466242536918,
      "loss": 5.2998,
      "mean_token_accuracy": 0.17277355939149858,
      "num_tokens": 42744945.0,
      "step": 23170
    },
    {
      "entropy": 5.6855233669281,
      "epoch": 1.9470279353077085,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004619299585181501,
      "loss": 5.4262,
      "mean_token_accuracy": 0.17296514809131622,
      "num_tokens": 42754906.0,
      "step": 23175
    },
    {
      "entropy": 5.671021318435669,
      "epoch": 1.9474480151228732,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046191328947124027,
      "loss": 5.3148,
      "mean_token_accuracy": 0.17584728300571442,
      "num_tokens": 42764673.0,
      "step": 23180
    },
    {
      "entropy": 5.528969097137451,
      "epoch": 1.9478680949380383,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046189661711325784,
      "loss": 5.3105,
      "mean_token_accuracy": 0.18131239712238312,
      "num_tokens": 42774528.0,
      "step": 23185
    },
    {
      "entropy": 5.583880233764648,
      "epoch": 1.9482881747532033,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046187994144449815,
      "loss": 5.2106,
      "mean_token_accuracy": 0.1867326408624649,
      "num_tokens": 42783813.0,
      "step": 23190
    },
    {
      "entropy": 5.611995267868042,
      "epoch": 1.948708254568368,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004618632624652565,
      "loss": 5.294,
      "mean_token_accuracy": 0.17145321667194366,
      "num_tokens": 42793483.0,
      "step": 23195
    },
    {
      "entropy": 5.629390096664428,
      "epoch": 1.9491283343835328,
      "grad_norm": 1.375,
      "learning_rate": 0.0004618465801758283,
      "loss": 5.3742,
      "mean_token_accuracy": 0.16771534085273743,
      "num_tokens": 42803177.0,
      "step": 23200
    },
    {
      "entropy": 5.6585023403167725,
      "epoch": 1.9495484141986976,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046182989457650925,
      "loss": 5.3724,
      "mean_token_accuracy": 0.17262449264526367,
      "num_tokens": 42812395.0,
      "step": 23205
    },
    {
      "entropy": 5.621705532073975,
      "epoch": 1.9499684940138626,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046181320566759476,
      "loss": 5.338,
      "mean_token_accuracy": 0.17661595195531846,
      "num_tokens": 42821495.0,
      "step": 23210
    },
    {
      "entropy": 5.650861120223999,
      "epoch": 1.9503885738290276,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046179651344938055,
      "loss": 5.321,
      "mean_token_accuracy": 0.17407327592372895,
      "num_tokens": 42832219.0,
      "step": 23215
    },
    {
      "entropy": 5.610399389266968,
      "epoch": 1.9508086536441924,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046177981792216234,
      "loss": 5.2586,
      "mean_token_accuracy": 0.17683097422122956,
      "num_tokens": 42841368.0,
      "step": 23220
    },
    {
      "entropy": 5.56700644493103,
      "epoch": 1.9512287334593572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046176311908623574,
      "loss": 5.3005,
      "mean_token_accuracy": 0.18181250393390655,
      "num_tokens": 42850512.0,
      "step": 23225
    },
    {
      "entropy": 5.636547994613648,
      "epoch": 1.951648813274522,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004617464169418967,
      "loss": 5.3551,
      "mean_token_accuracy": 0.1731240376830101,
      "num_tokens": 42860749.0,
      "step": 23230
    },
    {
      "entropy": 5.62040343284607,
      "epoch": 1.952068893089687,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046172971148944106,
      "loss": 5.296,
      "mean_token_accuracy": 0.1800855353474617,
      "num_tokens": 42869880.0,
      "step": 23235
    },
    {
      "entropy": 5.624702453613281,
      "epoch": 1.952488972904852,
      "grad_norm": 1.375,
      "learning_rate": 0.00046171300272916465,
      "loss": 5.2848,
      "mean_token_accuracy": 0.1809431105852127,
      "num_tokens": 42879001.0,
      "step": 23240
    },
    {
      "entropy": 5.57149829864502,
      "epoch": 1.9529090527200168,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046169629066136357,
      "loss": 5.2171,
      "mean_token_accuracy": 0.17806676477193834,
      "num_tokens": 42888036.0,
      "step": 23245
    },
    {
      "entropy": 5.624969148635865,
      "epoch": 1.9533291325351816,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046167957528633387,
      "loss": 5.3135,
      "mean_token_accuracy": 0.18084749430418015,
      "num_tokens": 42897460.0,
      "step": 23250
    },
    {
      "entropy": 5.567489099502564,
      "epoch": 1.9537492123503466,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046166285660437164,
      "loss": 5.3459,
      "mean_token_accuracy": 0.1768065482378006,
      "num_tokens": 42907010.0,
      "step": 23255
    },
    {
      "entropy": 5.676025152206421,
      "epoch": 1.9541692921655116,
      "grad_norm": 1.34375,
      "learning_rate": 0.000461646134615773,
      "loss": 5.2882,
      "mean_token_accuracy": 0.17346674352884292,
      "num_tokens": 42915684.0,
      "step": 23260
    },
    {
      "entropy": 5.625718021392823,
      "epoch": 1.9545893719806764,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046162940932083414,
      "loss": 5.2803,
      "mean_token_accuracy": 0.1799171894788742,
      "num_tokens": 42924903.0,
      "step": 23265
    },
    {
      "entropy": 5.5670435428619385,
      "epoch": 1.9550094517958412,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046161268071985144,
      "loss": 5.3092,
      "mean_token_accuracy": 0.1729440674185753,
      "num_tokens": 42935234.0,
      "step": 23270
    },
    {
      "entropy": 5.519500255584717,
      "epoch": 1.955429531611006,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004615959488131212,
      "loss": 5.2267,
      "mean_token_accuracy": 0.18312966376543044,
      "num_tokens": 42944093.0,
      "step": 23275
    },
    {
      "entropy": 5.630027627944946,
      "epoch": 1.955849611426171,
      "grad_norm": 1.375,
      "learning_rate": 0.0004615792136009398,
      "loss": 5.2571,
      "mean_token_accuracy": 0.17761701047420503,
      "num_tokens": 42953504.0,
      "step": 23280
    },
    {
      "entropy": 5.626860952377319,
      "epoch": 1.956269691241336,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046156247508360375,
      "loss": 5.2992,
      "mean_token_accuracy": 0.1789937436580658,
      "num_tokens": 42962205.0,
      "step": 23285
    },
    {
      "entropy": 5.5647505760192875,
      "epoch": 1.9566897710565008,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004615457332614095,
      "loss": 5.2328,
      "mean_token_accuracy": 0.18050500899553298,
      "num_tokens": 42971240.0,
      "step": 23290
    },
    {
      "entropy": 5.653928756713867,
      "epoch": 1.9571098508716656,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046152898813465353,
      "loss": 5.3725,
      "mean_token_accuracy": 0.17226796448230744,
      "num_tokens": 42981573.0,
      "step": 23295
    },
    {
      "entropy": 5.636758136749267,
      "epoch": 1.9575299306868303,
      "grad_norm": 1.5,
      "learning_rate": 0.0004615122397036327,
      "loss": 5.3061,
      "mean_token_accuracy": 0.17363144904375077,
      "num_tokens": 42991383.0,
      "step": 23300
    },
    {
      "entropy": 5.623042821884155,
      "epoch": 1.9579500105019954,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046149548796864355,
      "loss": 5.2427,
      "mean_token_accuracy": 0.18039155006408691,
      "num_tokens": 43000029.0,
      "step": 23305
    },
    {
      "entropy": 5.626794862747192,
      "epoch": 1.9583700903171604,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046147873292998285,
      "loss": 5.3351,
      "mean_token_accuracy": 0.17696916162967682,
      "num_tokens": 43008880.0,
      "step": 23310
    },
    {
      "entropy": 5.567250394821167,
      "epoch": 1.9587901701323251,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004614619745879475,
      "loss": 5.3045,
      "mean_token_accuracy": 0.1828549921512604,
      "num_tokens": 43017417.0,
      "step": 23315
    },
    {
      "entropy": 5.557860469818115,
      "epoch": 1.95921024994749,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004614452129428342,
      "loss": 5.2382,
      "mean_token_accuracy": 0.18235221952199937,
      "num_tokens": 43025738.0,
      "step": 23320
    },
    {
      "entropy": 5.649650287628174,
      "epoch": 1.959630329762655,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004614284479949399,
      "loss": 5.3348,
      "mean_token_accuracy": 0.17602566182613372,
      "num_tokens": 43035485.0,
      "step": 23325
    },
    {
      "entropy": 5.716347026824951,
      "epoch": 1.96005040957782,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004614116797445617,
      "loss": 5.3567,
      "mean_token_accuracy": 0.17880905866622926,
      "num_tokens": 43044627.0,
      "step": 23330
    },
    {
      "entropy": 5.577663278579712,
      "epoch": 1.9604704893929847,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046139490819199666,
      "loss": 5.2768,
      "mean_token_accuracy": 0.17549317479133605,
      "num_tokens": 43053790.0,
      "step": 23335
    },
    {
      "entropy": 5.63191933631897,
      "epoch": 1.9608905692081495,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004613781333375417,
      "loss": 5.2767,
      "mean_token_accuracy": 0.18686962276697158,
      "num_tokens": 43063511.0,
      "step": 23340
    },
    {
      "entropy": 5.500326824188233,
      "epoch": 1.9613106490233143,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004613613551814941,
      "loss": 5.2156,
      "mean_token_accuracy": 0.18084116280078888,
      "num_tokens": 43072349.0,
      "step": 23345
    },
    {
      "entropy": 5.607284307479858,
      "epoch": 1.9617307288384793,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004613445737241511,
      "loss": 5.3267,
      "mean_token_accuracy": 0.17642860859632492,
      "num_tokens": 43081552.0,
      "step": 23350
    },
    {
      "entropy": 5.672708463668823,
      "epoch": 1.9621508086536443,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046132778896581,
      "loss": 5.3661,
      "mean_token_accuracy": 0.1795952647924423,
      "num_tokens": 43092321.0,
      "step": 23355
    },
    {
      "entropy": 5.662492322921753,
      "epoch": 1.9625708884688091,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004613110009067679,
      "loss": 5.314,
      "mean_token_accuracy": 0.17554208636283875,
      "num_tokens": 43102326.0,
      "step": 23360
    },
    {
      "entropy": 5.621742010116577,
      "epoch": 1.962990968283974,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046129420954732237,
      "loss": 5.3411,
      "mean_token_accuracy": 0.17261791080236435,
      "num_tokens": 43110895.0,
      "step": 23365
    },
    {
      "entropy": 5.575307989120484,
      "epoch": 1.9634110480991387,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004612774148877709,
      "loss": 5.1997,
      "mean_token_accuracy": 0.18181467056274414,
      "num_tokens": 43119948.0,
      "step": 23370
    },
    {
      "entropy": 5.658673667907715,
      "epoch": 1.9638311279143037,
      "grad_norm": 1.25,
      "learning_rate": 0.000461260616928411,
      "loss": 5.3908,
      "mean_token_accuracy": 0.1741318941116333,
      "num_tokens": 43129876.0,
      "step": 23375
    },
    {
      "entropy": 5.654330539703369,
      "epoch": 1.9642512077294687,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046124381566954006,
      "loss": 5.3429,
      "mean_token_accuracy": 0.1764851152896881,
      "num_tokens": 43138831.0,
      "step": 23380
    },
    {
      "entropy": 5.680088329315185,
      "epoch": 1.9646712875446335,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046122701111145587,
      "loss": 5.3206,
      "mean_token_accuracy": 0.17365952283143998,
      "num_tokens": 43147338.0,
      "step": 23385
    },
    {
      "entropy": 5.603476905822754,
      "epoch": 1.9650913673597983,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004612102032544561,
      "loss": 5.272,
      "mean_token_accuracy": 0.17501939684152604,
      "num_tokens": 43158587.0,
      "step": 23390
    },
    {
      "entropy": 5.596035623550415,
      "epoch": 1.9655114471749633,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046119339209883846,
      "loss": 5.2678,
      "mean_token_accuracy": 0.17789842933416367,
      "num_tokens": 43167610.0,
      "step": 23395
    },
    {
      "entropy": 5.516913652420044,
      "epoch": 1.965931526990128,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004611765776449007,
      "loss": 5.233,
      "mean_token_accuracy": 0.17464807480573655,
      "num_tokens": 43176374.0,
      "step": 23400
    },
    {
      "entropy": 5.638162851333618,
      "epoch": 1.966351606805293,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046115975989294083,
      "loss": 5.3907,
      "mean_token_accuracy": 0.17190734297037125,
      "num_tokens": 43187038.0,
      "step": 23405
    },
    {
      "entropy": 5.742623424530029,
      "epoch": 1.9667716866204579,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004611429388432566,
      "loss": 5.4008,
      "mean_token_accuracy": 0.17257195413112641,
      "num_tokens": 43197868.0,
      "step": 23410
    },
    {
      "entropy": 5.627084398269654,
      "epoch": 1.9671917664356227,
      "grad_norm": 2.671875,
      "learning_rate": 0.00046112611449614603,
      "loss": 5.3556,
      "mean_token_accuracy": 0.16652406454086305,
      "num_tokens": 43207675.0,
      "step": 23415
    },
    {
      "entropy": 5.622139120101929,
      "epoch": 1.9676118462507877,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004611092868519072,
      "loss": 5.3457,
      "mean_token_accuracy": 0.17196379005908966,
      "num_tokens": 43217154.0,
      "step": 23420
    },
    {
      "entropy": 5.620547437667847,
      "epoch": 1.9680319260659527,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004610924559108383,
      "loss": 5.3552,
      "mean_token_accuracy": 0.18111842423677443,
      "num_tokens": 43226912.0,
      "step": 23425
    },
    {
      "entropy": 5.657605123519898,
      "epoch": 1.9684520058811175,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004610756216732372,
      "loss": 5.3413,
      "mean_token_accuracy": 0.1757437616586685,
      "num_tokens": 43236711.0,
      "step": 23430
    },
    {
      "entropy": 5.635782384872437,
      "epoch": 1.9688720856962822,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046105878413940237,
      "loss": 5.326,
      "mean_token_accuracy": 0.1855035811662674,
      "num_tokens": 43247005.0,
      "step": 23435
    },
    {
      "entropy": 5.40121488571167,
      "epoch": 1.969292165511447,
      "grad_norm": 1.375,
      "learning_rate": 0.000461041943309632,
      "loss": 5.1641,
      "mean_token_accuracy": 0.1891426831483841,
      "num_tokens": 43255868.0,
      "step": 23440
    },
    {
      "entropy": 5.597007608413696,
      "epoch": 1.969712245326612,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004610250991842244,
      "loss": 5.2841,
      "mean_token_accuracy": 0.17631662786006927,
      "num_tokens": 43265727.0,
      "step": 23445
    },
    {
      "entropy": 5.665576028823852,
      "epoch": 1.970132325141777,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046100825176347796,
      "loss": 5.3301,
      "mean_token_accuracy": 0.18131863325834274,
      "num_tokens": 43274530.0,
      "step": 23450
    },
    {
      "entropy": 5.465054082870483,
      "epoch": 1.9705524049569418,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000460991401047691,
      "loss": 5.2339,
      "mean_token_accuracy": 0.1741779625415802,
      "num_tokens": 43285130.0,
      "step": 23455
    },
    {
      "entropy": 5.596411275863647,
      "epoch": 1.9709724847721066,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004609745470371622,
      "loss": 5.3064,
      "mean_token_accuracy": 0.1783148989081383,
      "num_tokens": 43293574.0,
      "step": 23460
    },
    {
      "entropy": 5.528018665313721,
      "epoch": 1.9713925645872716,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004609576897321902,
      "loss": 5.1384,
      "mean_token_accuracy": 0.1825125053524971,
      "num_tokens": 43301989.0,
      "step": 23465
    },
    {
      "entropy": 5.656487989425659,
      "epoch": 1.9718126444024364,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046094082913307336,
      "loss": 5.3474,
      "mean_token_accuracy": 0.1780552476644516,
      "num_tokens": 43310934.0,
      "step": 23470
    },
    {
      "entropy": 5.615532064437867,
      "epoch": 1.9722327242176014,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004609239652401104,
      "loss": 5.2518,
      "mean_token_accuracy": 0.1761912152171135,
      "num_tokens": 43320703.0,
      "step": 23475
    },
    {
      "entropy": 5.527682876586914,
      "epoch": 1.9726528040327662,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046090709805360027,
      "loss": 5.2351,
      "mean_token_accuracy": 0.18263690620660783,
      "num_tokens": 43329444.0,
      "step": 23480
    },
    {
      "entropy": 5.593132877349854,
      "epoch": 1.973072883847931,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004608902275738416,
      "loss": 5.3448,
      "mean_token_accuracy": 0.1838742733001709,
      "num_tokens": 43337853.0,
      "step": 23485
    },
    {
      "entropy": 5.673547887802124,
      "epoch": 1.973492963663096,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004608733538011333,
      "loss": 5.3891,
      "mean_token_accuracy": 0.17783911377191544,
      "num_tokens": 43347901.0,
      "step": 23490
    },
    {
      "entropy": 5.581468725204468,
      "epoch": 1.973913043478261,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004608564767357741,
      "loss": 5.2342,
      "mean_token_accuracy": 0.17998829782009124,
      "num_tokens": 43357358.0,
      "step": 23495
    },
    {
      "entropy": 5.573663568496704,
      "epoch": 1.9743331232934258,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004608395963780632,
      "loss": 5.3534,
      "mean_token_accuracy": 0.17246935814619063,
      "num_tokens": 43366749.0,
      "step": 23500
    },
    {
      "entropy": 5.558063793182373,
      "epoch": 1.9747532031085906,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004608227127282996,
      "loss": 5.3028,
      "mean_token_accuracy": 0.17924269586801528,
      "num_tokens": 43375243.0,
      "step": 23505
    },
    {
      "entropy": 5.641796207427978,
      "epoch": 1.9751732829237554,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004608058257867823,
      "loss": 5.285,
      "mean_token_accuracy": 0.18259629756212234,
      "num_tokens": 43383470.0,
      "step": 23510
    },
    {
      "entropy": 5.654600429534912,
      "epoch": 1.9755933627389204,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004607889355538105,
      "loss": 5.3901,
      "mean_token_accuracy": 0.17257382869720458,
      "num_tokens": 43393527.0,
      "step": 23515
    },
    {
      "entropy": 5.576977109909057,
      "epoch": 1.9760134425540854,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046077204202968325,
      "loss": 5.255,
      "mean_token_accuracy": 0.18039424121379852,
      "num_tokens": 43402390.0,
      "step": 23520
    },
    {
      "entropy": 5.569544172286987,
      "epoch": 1.9764335223692502,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046075514521470005,
      "loss": 5.2473,
      "mean_token_accuracy": 0.1748657539486885,
      "num_tokens": 43411479.0,
      "step": 23525
    },
    {
      "entropy": 5.51814866065979,
      "epoch": 1.976853602184415,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046073824510916005,
      "loss": 5.1941,
      "mean_token_accuracy": 0.18002250790596008,
      "num_tokens": 43420402.0,
      "step": 23530
    },
    {
      "entropy": 5.560939025878906,
      "epoch": 1.9772736819995798,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046072134171336267,
      "loss": 5.3284,
      "mean_token_accuracy": 0.17190781831741334,
      "num_tokens": 43429011.0,
      "step": 23535
    },
    {
      "entropy": 5.5692507266998295,
      "epoch": 1.9776937618147448,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004607044350276074,
      "loss": 5.2297,
      "mean_token_accuracy": 0.18149014711380004,
      "num_tokens": 43438548.0,
      "step": 23540
    },
    {
      "entropy": 5.669879388809204,
      "epoch": 1.9781138416299098,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046068752505219366,
      "loss": 5.3185,
      "mean_token_accuracy": 0.17451345026493073,
      "num_tokens": 43448332.0,
      "step": 23545
    },
    {
      "entropy": 5.651257038116455,
      "epoch": 1.9785339214450746,
      "grad_norm": 1.703125,
      "learning_rate": 0.000460670611787421,
      "loss": 5.3845,
      "mean_token_accuracy": 0.16970988661050795,
      "num_tokens": 43457726.0,
      "step": 23550
    },
    {
      "entropy": 5.562670087814331,
      "epoch": 1.9789540012602393,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004606536952335891,
      "loss": 5.2988,
      "mean_token_accuracy": 0.17578113079071045,
      "num_tokens": 43466617.0,
      "step": 23555
    },
    {
      "entropy": 5.514257860183716,
      "epoch": 1.9793740810754044,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046063677539099756,
      "loss": 5.2875,
      "mean_token_accuracy": 0.17199433147907256,
      "num_tokens": 43476044.0,
      "step": 23560
    },
    {
      "entropy": 5.598515272140503,
      "epoch": 1.9797941608905694,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046061985225994616,
      "loss": 5.2629,
      "mean_token_accuracy": 0.18017773777246476,
      "num_tokens": 43485488.0,
      "step": 23565
    },
    {
      "entropy": 5.629073429107666,
      "epoch": 1.9802142407057342,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046060292584073465,
      "loss": 5.2923,
      "mean_token_accuracy": 0.17964733392000198,
      "num_tokens": 43494423.0,
      "step": 23570
    },
    {
      "entropy": 5.573811626434326,
      "epoch": 1.980634320520899,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046058599613366287,
      "loss": 5.1699,
      "mean_token_accuracy": 0.1919006109237671,
      "num_tokens": 43502874.0,
      "step": 23575
    },
    {
      "entropy": 5.652579832077026,
      "epoch": 1.9810544003360637,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004605690631390308,
      "loss": 5.4173,
      "mean_token_accuracy": 0.16988031268119813,
      "num_tokens": 43512222.0,
      "step": 23580
    },
    {
      "entropy": 5.615391969680786,
      "epoch": 1.9814744801512287,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004605521268571382,
      "loss": 5.3316,
      "mean_token_accuracy": 0.1798727795481682,
      "num_tokens": 43521577.0,
      "step": 23585
    },
    {
      "entropy": 5.656839513778687,
      "epoch": 1.9818945599663937,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00046053518728828534,
      "loss": 5.2929,
      "mean_token_accuracy": 0.17951315343379975,
      "num_tokens": 43529763.0,
      "step": 23590
    },
    {
      "entropy": 5.651676654815674,
      "epoch": 1.9823146397815585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004605182444327721,
      "loss": 5.3453,
      "mean_token_accuracy": 0.17168731093406678,
      "num_tokens": 43538663.0,
      "step": 23595
    },
    {
      "entropy": 5.571462154388428,
      "epoch": 1.9827347195967233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004605012982908987,
      "loss": 5.1472,
      "mean_token_accuracy": 0.18435227423906325,
      "num_tokens": 43547302.0,
      "step": 23600
    },
    {
      "entropy": 5.575665998458862,
      "epoch": 1.983154799411888,
      "grad_norm": 1.25,
      "learning_rate": 0.00046048434886296536,
      "loss": 5.354,
      "mean_token_accuracy": 0.1710020199418068,
      "num_tokens": 43557222.0,
      "step": 23605
    },
    {
      "entropy": 5.5998461723327635,
      "epoch": 1.9835748792270531,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004604673961492722,
      "loss": 5.2461,
      "mean_token_accuracy": 0.17823938131332398,
      "num_tokens": 43566210.0,
      "step": 23610
    },
    {
      "entropy": 5.557861852645874,
      "epoch": 1.9839949590422181,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046045044015011975,
      "loss": 5.2292,
      "mean_token_accuracy": 0.1795830398797989,
      "num_tokens": 43576275.0,
      "step": 23615
    },
    {
      "entropy": 5.585061740875244,
      "epoch": 1.984415038857383,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004604334808658081,
      "loss": 5.3526,
      "mean_token_accuracy": 0.17323249578475952,
      "num_tokens": 43585480.0,
      "step": 23620
    },
    {
      "entropy": 5.639600992202759,
      "epoch": 1.9848351186725477,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046041651829663787,
      "loss": 5.3727,
      "mean_token_accuracy": 0.17491732090711593,
      "num_tokens": 43593911.0,
      "step": 23625
    },
    {
      "entropy": 5.6159913539886475,
      "epoch": 1.9852551984877127,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046039955244290957,
      "loss": 5.2976,
      "mean_token_accuracy": 0.1727191761136055,
      "num_tokens": 43604029.0,
      "step": 23630
    },
    {
      "entropy": 5.63344144821167,
      "epoch": 1.9856752783028777,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046038258330492363,
      "loss": 5.3392,
      "mean_token_accuracy": 0.18076819628477098,
      "num_tokens": 43613248.0,
      "step": 23635
    },
    {
      "entropy": 5.627498912811279,
      "epoch": 1.9860953581180425,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004603656108829806,
      "loss": 5.2829,
      "mean_token_accuracy": 0.1781997114419937,
      "num_tokens": 43623232.0,
      "step": 23640
    },
    {
      "entropy": 5.689774131774902,
      "epoch": 1.9865154379332073,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046034863517738136,
      "loss": 5.3511,
      "mean_token_accuracy": 0.16521923393011093,
      "num_tokens": 43632999.0,
      "step": 23645
    },
    {
      "entropy": 5.593861246109009,
      "epoch": 1.986935517748372,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046033165618842637,
      "loss": 5.3079,
      "mean_token_accuracy": 0.176187202334404,
      "num_tokens": 43641492.0,
      "step": 23650
    },
    {
      "entropy": 5.703560924530029,
      "epoch": 1.987355597563537,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046031467391641657,
      "loss": 5.2908,
      "mean_token_accuracy": 0.18088617026805878,
      "num_tokens": 43650999.0,
      "step": 23655
    },
    {
      "entropy": 5.671897935867309,
      "epoch": 1.987775677378702,
      "grad_norm": 1.25,
      "learning_rate": 0.0004602976883616527,
      "loss": 5.3826,
      "mean_token_accuracy": 0.16539280563592912,
      "num_tokens": 43660777.0,
      "step": 23660
    },
    {
      "entropy": 5.545668268203736,
      "epoch": 1.9881957571938669,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046028069952443575,
      "loss": 5.281,
      "mean_token_accuracy": 0.18010626286268233,
      "num_tokens": 43670404.0,
      "step": 23665
    },
    {
      "entropy": 5.5640342235565186,
      "epoch": 1.9886158370090317,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046026370740506663,
      "loss": 5.2254,
      "mean_token_accuracy": 0.18301359862089156,
      "num_tokens": 43679183.0,
      "step": 23670
    },
    {
      "entropy": 5.541779041290283,
      "epoch": 1.9890359168241964,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004602467120038463,
      "loss": 5.2292,
      "mean_token_accuracy": 0.18373605161905288,
      "num_tokens": 43688080.0,
      "step": 23675
    },
    {
      "entropy": 5.61362624168396,
      "epoch": 1.9894559966393615,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046022971332107586,
      "loss": 5.2172,
      "mean_token_accuracy": 0.1879374623298645,
      "num_tokens": 43697271.0,
      "step": 23680
    },
    {
      "entropy": 5.563038635253906,
      "epoch": 1.9898760764545265,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046021271135705637,
      "loss": 5.2273,
      "mean_token_accuracy": 0.18289953768253325,
      "num_tokens": 43705541.0,
      "step": 23685
    },
    {
      "entropy": 5.5623798847198485,
      "epoch": 1.9902961562696913,
      "grad_norm": 1.375,
      "learning_rate": 0.0004601957061120891,
      "loss": 5.362,
      "mean_token_accuracy": 0.17016636431217194,
      "num_tokens": 43713701.0,
      "step": 23690
    },
    {
      "entropy": 5.582708740234375,
      "epoch": 1.990716236084856,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004601786975864753,
      "loss": 5.3175,
      "mean_token_accuracy": 0.18513188362121583,
      "num_tokens": 43723050.0,
      "step": 23695
    },
    {
      "entropy": 5.59705867767334,
      "epoch": 1.991136315900021,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004601616857805162,
      "loss": 5.2862,
      "mean_token_accuracy": 0.18229536563158036,
      "num_tokens": 43733029.0,
      "step": 23700
    },
    {
      "entropy": 5.536784029006958,
      "epoch": 1.9915563957151858,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004601446706945132,
      "loss": 5.2666,
      "mean_token_accuracy": 0.17475199550390244,
      "num_tokens": 43741818.0,
      "step": 23705
    },
    {
      "entropy": 5.58607292175293,
      "epoch": 1.9919764755303508,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046012765232876767,
      "loss": 5.295,
      "mean_token_accuracy": 0.18050657361745834,
      "num_tokens": 43750755.0,
      "step": 23710
    },
    {
      "entropy": 5.502914237976074,
      "epoch": 1.9923965553455156,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004601106306835811,
      "loss": 5.1978,
      "mean_token_accuracy": 0.18477195501327515,
      "num_tokens": 43759135.0,
      "step": 23715
    },
    {
      "entropy": 5.552498292922974,
      "epoch": 1.9928166351606804,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004600936057592551,
      "loss": 5.1403,
      "mean_token_accuracy": 0.1892944946885109,
      "num_tokens": 43767629.0,
      "step": 23720
    },
    {
      "entropy": 5.607378053665161,
      "epoch": 1.9932367149758454,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046007657755609113,
      "loss": 5.3316,
      "mean_token_accuracy": 0.17264370173215865,
      "num_tokens": 43776561.0,
      "step": 23725
    },
    {
      "entropy": 5.667064476013183,
      "epoch": 1.9936567947910104,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004600595460743908,
      "loss": 5.4089,
      "mean_token_accuracy": 0.16604346483945848,
      "num_tokens": 43786569.0,
      "step": 23730
    },
    {
      "entropy": 5.597362327575683,
      "epoch": 1.9940768746061752,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000460042511314456,
      "loss": 5.351,
      "mean_token_accuracy": 0.17066410183906555,
      "num_tokens": 43795621.0,
      "step": 23735
    },
    {
      "entropy": 5.693417549133301,
      "epoch": 1.99449695442134,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046002547327658847,
      "loss": 5.3389,
      "mean_token_accuracy": 0.17495685517787934,
      "num_tokens": 43804728.0,
      "step": 23740
    },
    {
      "entropy": 5.567583084106445,
      "epoch": 1.9949170342365048,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004600084319610898,
      "loss": 5.2516,
      "mean_token_accuracy": 0.1847442075610161,
      "num_tokens": 43813495.0,
      "step": 23745
    },
    {
      "entropy": 5.5047966003417965,
      "epoch": 1.9953371140516698,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004599913873682621,
      "loss": 5.1783,
      "mean_token_accuracy": 0.1822268098592758,
      "num_tokens": 43823791.0,
      "step": 23750
    },
    {
      "entropy": 5.591089963912964,
      "epoch": 1.9957571938668348,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045997433949840724,
      "loss": 5.2532,
      "mean_token_accuracy": 0.18184020668268203,
      "num_tokens": 43833904.0,
      "step": 23755
    },
    {
      "entropy": 5.637771415710449,
      "epoch": 1.9961772736819996,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00045995728835182716,
      "loss": 5.3445,
      "mean_token_accuracy": 0.17545148134231567,
      "num_tokens": 43843430.0,
      "step": 23760
    },
    {
      "entropy": 5.651317596435547,
      "epoch": 1.9965973534971644,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045994023392882395,
      "loss": 5.2857,
      "mean_token_accuracy": 0.1868788331747055,
      "num_tokens": 43851405.0,
      "step": 23765
    },
    {
      "entropy": 5.652814865112305,
      "epoch": 1.9970174333123294,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045992317622969977,
      "loss": 5.373,
      "mean_token_accuracy": 0.17030868381261827,
      "num_tokens": 43860034.0,
      "step": 23770
    },
    {
      "entropy": 5.5887829780578615,
      "epoch": 1.9974375131274942,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045990611525475675,
      "loss": 5.3062,
      "mean_token_accuracy": 0.17263094931840897,
      "num_tokens": 43869371.0,
      "step": 23775
    },
    {
      "entropy": 5.672349834442139,
      "epoch": 1.9978575929426592,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004598890510042971,
      "loss": 5.3547,
      "mean_token_accuracy": 0.17403602600097656,
      "num_tokens": 43878462.0,
      "step": 23780
    },
    {
      "entropy": 5.709714889526367,
      "epoch": 1.998277672757824,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000459871983478623,
      "loss": 5.2695,
      "mean_token_accuracy": 0.17943592071533204,
      "num_tokens": 43887435.0,
      "step": 23785
    },
    {
      "entropy": 5.636635160446167,
      "epoch": 1.9986977525729888,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00045985491267803703,
      "loss": 5.3252,
      "mean_token_accuracy": 0.17516879886388778,
      "num_tokens": 43896720.0,
      "step": 23790
    },
    {
      "entropy": 5.6023845195770265,
      "epoch": 1.9991178323881538,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045983783860284146,
      "loss": 5.3364,
      "mean_token_accuracy": 0.17209494560956956,
      "num_tokens": 43906403.0,
      "step": 23795
    },
    {
      "entropy": 5.630288171768188,
      "epoch": 1.9995379122033188,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045982076125333874,
      "loss": 5.3652,
      "mean_token_accuracy": 0.17041498869657518,
      "num_tokens": 43915059.0,
      "step": 23800
    },
    {
      "entropy": 5.730317735671997,
      "epoch": 1.9999579920184836,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045980368062983147,
      "loss": 5.3868,
      "mean_token_accuracy": 0.17035310864448547,
      "num_tokens": 43925598.0,
      "step": 23805
    },
    {
      "entropy": 5.6578372319539385,
      "epoch": 2.000336063852132,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004597865967326221,
      "loss": 5.2007,
      "mean_token_accuracy": 0.17507648799154493,
      "num_tokens": 43934471.0,
      "step": 23810
    },
    {
      "entropy": 5.58235182762146,
      "epoch": 2.0007561436672967,
      "grad_norm": 1.375,
      "learning_rate": 0.00045976950956201325,
      "loss": 5.2865,
      "mean_token_accuracy": 0.17469087839126587,
      "num_tokens": 43944451.0,
      "step": 23815
    },
    {
      "entropy": 5.694094800949097,
      "epoch": 2.0011762234824615,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004597524191183078,
      "loss": 5.2524,
      "mean_token_accuracy": 0.1853089526295662,
      "num_tokens": 43953892.0,
      "step": 23820
    },
    {
      "entropy": 5.657417249679566,
      "epoch": 2.0015963032976267,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004597353254018082,
      "loss": 5.3222,
      "mean_token_accuracy": 0.1727910190820694,
      "num_tokens": 43963155.0,
      "step": 23825
    },
    {
      "entropy": 5.59494309425354,
      "epoch": 2.0020163831127915,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004597182284128177,
      "loss": 5.1901,
      "mean_token_accuracy": 0.18787469416856767,
      "num_tokens": 43972468.0,
      "step": 23830
    },
    {
      "entropy": 5.748509216308594,
      "epoch": 2.0024364629279563,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004597011281516387,
      "loss": 5.4383,
      "mean_token_accuracy": 0.16790467798709868,
      "num_tokens": 43982709.0,
      "step": 23835
    },
    {
      "entropy": 5.58790955543518,
      "epoch": 2.002856542743121,
      "grad_norm": 1.546875,
      "learning_rate": 0.00045968402461857435,
      "loss": 5.199,
      "mean_token_accuracy": 0.18892733305692672,
      "num_tokens": 43992607.0,
      "step": 23840
    },
    {
      "entropy": 5.578325796127319,
      "epoch": 2.003276622558286,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045966691781392763,
      "loss": 5.1734,
      "mean_token_accuracy": 0.18043159544467927,
      "num_tokens": 44001265.0,
      "step": 23845
    },
    {
      "entropy": 5.653204393386841,
      "epoch": 2.003696702373451,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045964980773800156,
      "loss": 5.3648,
      "mean_token_accuracy": 0.1742731049656868,
      "num_tokens": 44010440.0,
      "step": 23850
    },
    {
      "entropy": 5.621107196807861,
      "epoch": 2.004116782188616,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004596326943910993,
      "loss": 5.2092,
      "mean_token_accuracy": 0.1796395882964134,
      "num_tokens": 44020237.0,
      "step": 23855
    },
    {
      "entropy": 5.617325401306152,
      "epoch": 2.0045368620037807,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045961557777352376,
      "loss": 5.3118,
      "mean_token_accuracy": 0.17234931290149688,
      "num_tokens": 44028976.0,
      "step": 23860
    },
    {
      "entropy": 5.648115825653076,
      "epoch": 2.0049569418189455,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045959845788557844,
      "loss": 5.2939,
      "mean_token_accuracy": 0.17594058066606522,
      "num_tokens": 44038186.0,
      "step": 23865
    },
    {
      "entropy": 5.613311719894409,
      "epoch": 2.0053770216341107,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004595813347275665,
      "loss": 5.2367,
      "mean_token_accuracy": 0.17081933170557023,
      "num_tokens": 44047780.0,
      "step": 23870
    },
    {
      "entropy": 5.602234649658203,
      "epoch": 2.0057971014492755,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004595642082997912,
      "loss": 5.196,
      "mean_token_accuracy": 0.1816309615969658,
      "num_tokens": 44056678.0,
      "step": 23875
    },
    {
      "entropy": 5.6001729488372805,
      "epoch": 2.0062171812644403,
      "grad_norm": 1.265625,
      "learning_rate": 0.000459547078602556,
      "loss": 5.2542,
      "mean_token_accuracy": 0.1733650431036949,
      "num_tokens": 44066428.0,
      "step": 23880
    },
    {
      "entropy": 5.602000761032104,
      "epoch": 2.006637261079605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045952994563616434,
      "loss": 5.2274,
      "mean_token_accuracy": 0.1813978970050812,
      "num_tokens": 44075285.0,
      "step": 23885
    },
    {
      "entropy": 5.62816047668457,
      "epoch": 2.00705734089477,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004595128094009197,
      "loss": 5.2325,
      "mean_token_accuracy": 0.17623593807220458,
      "num_tokens": 44084333.0,
      "step": 23890
    },
    {
      "entropy": 5.625877332687378,
      "epoch": 2.007477420709935,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004594956698971256,
      "loss": 5.2512,
      "mean_token_accuracy": 0.17474242001771928,
      "num_tokens": 44093504.0,
      "step": 23895
    },
    {
      "entropy": 5.6584666728973385,
      "epoch": 2.0078975005251,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004594785271250858,
      "loss": 5.2595,
      "mean_token_accuracy": 0.1743064358830452,
      "num_tokens": 44102887.0,
      "step": 23900
    },
    {
      "entropy": 5.561724042892456,
      "epoch": 2.0083175803402646,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004594613810851039,
      "loss": 5.2499,
      "mean_token_accuracy": 0.17559059709310532,
      "num_tokens": 44113074.0,
      "step": 23905
    },
    {
      "entropy": 5.574095249176025,
      "epoch": 2.0087376601554294,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045944423177748353,
      "loss": 5.2384,
      "mean_token_accuracy": 0.18805206567049026,
      "num_tokens": 44122557.0,
      "step": 23910
    },
    {
      "entropy": 5.5974695682525635,
      "epoch": 2.009157739970594,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045942707920252864,
      "loss": 5.2718,
      "mean_token_accuracy": 0.17523897886276246,
      "num_tokens": 44130198.0,
      "step": 23915
    },
    {
      "entropy": 5.598403835296631,
      "epoch": 2.0095778197857594,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000459409923360543,
      "loss": 5.2272,
      "mean_token_accuracy": 0.185373891890049,
      "num_tokens": 44139267.0,
      "step": 23920
    },
    {
      "entropy": 5.604103994369507,
      "epoch": 2.0099978996009242,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004593927642518305,
      "loss": 5.2862,
      "mean_token_accuracy": 0.1756848230957985,
      "num_tokens": 44149620.0,
      "step": 23925
    },
    {
      "entropy": 5.5682422637939455,
      "epoch": 2.010417979416089,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004593756018766951,
      "loss": 5.1411,
      "mean_token_accuracy": 0.181699176132679,
      "num_tokens": 44158678.0,
      "step": 23930
    },
    {
      "entropy": 5.500216627120972,
      "epoch": 2.010838059231254,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045935843623544093,
      "loss": 5.1303,
      "mean_token_accuracy": 0.17824358344078065,
      "num_tokens": 44167376.0,
      "step": 23935
    },
    {
      "entropy": 5.5720765590667725,
      "epoch": 2.011258139046419,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004593412673283719,
      "loss": 5.2524,
      "mean_token_accuracy": 0.18108217418193817,
      "num_tokens": 44176001.0,
      "step": 23940
    },
    {
      "entropy": 5.652003765106201,
      "epoch": 2.011678218861584,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045932409515579226,
      "loss": 5.3046,
      "mean_token_accuracy": 0.17257492542266845,
      "num_tokens": 44185132.0,
      "step": 23945
    },
    {
      "entropy": 5.577314758300782,
      "epoch": 2.0120982986767486,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045930691971800627,
      "loss": 5.2499,
      "mean_token_accuracy": 0.18168537318706512,
      "num_tokens": 44193256.0,
      "step": 23950
    },
    {
      "entropy": 5.700579309463501,
      "epoch": 2.0125183784919134,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045928974101531805,
      "loss": 5.3447,
      "mean_token_accuracy": 0.1787986636161804,
      "num_tokens": 44202884.0,
      "step": 23955
    },
    {
      "entropy": 5.677771139144897,
      "epoch": 2.012938458307078,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004592725590480319,
      "loss": 5.3188,
      "mean_token_accuracy": 0.17306433022022247,
      "num_tokens": 44212826.0,
      "step": 23960
    },
    {
      "entropy": 5.5983131408691404,
      "epoch": 2.0133585381222434,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004592553738164524,
      "loss": 5.2821,
      "mean_token_accuracy": 0.17070496827363968,
      "num_tokens": 44222369.0,
      "step": 23965
    },
    {
      "entropy": 5.596569299697876,
      "epoch": 2.013778617937408,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004592381853208837,
      "loss": 5.1992,
      "mean_token_accuracy": 0.18120295703411102,
      "num_tokens": 44230964.0,
      "step": 23970
    },
    {
      "entropy": 5.63546233177185,
      "epoch": 2.014198697752573,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004592209935616304,
      "loss": 5.2814,
      "mean_token_accuracy": 0.1777893841266632,
      "num_tokens": 44240199.0,
      "step": 23975
    },
    {
      "entropy": 5.612800645828247,
      "epoch": 2.0146187775677378,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004592037985389971,
      "loss": 5.2419,
      "mean_token_accuracy": 0.18401461243629455,
      "num_tokens": 44249857.0,
      "step": 23980
    },
    {
      "entropy": 5.5419095993042,
      "epoch": 2.0150388573829026,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004591866002532885,
      "loss": 5.2237,
      "mean_token_accuracy": 0.17834097295999526,
      "num_tokens": 44258364.0,
      "step": 23985
    },
    {
      "entropy": 5.514514636993408,
      "epoch": 2.015458937198068,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045916939870480896,
      "loss": 5.1397,
      "mean_token_accuracy": 0.17798831462860107,
      "num_tokens": 44267473.0,
      "step": 23990
    },
    {
      "entropy": 5.623774576187134,
      "epoch": 2.0158790170132326,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045915219389386336,
      "loss": 5.1893,
      "mean_token_accuracy": 0.1826447308063507,
      "num_tokens": 44276665.0,
      "step": 23995
    },
    {
      "entropy": 5.54183259010315,
      "epoch": 2.0162990968283974,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004591349858207565,
      "loss": 5.2364,
      "mean_token_accuracy": 0.17832147628068923,
      "num_tokens": 44285928.0,
      "step": 24000
    },
    {
      "epoch": 2.0162990968283974,
      "eval_entropy": 5.364241007129275,
      "eval_loss": 5.346562385559082,
      "eval_mean_token_accuracy": 0.18400972061301038,
      "eval_num_tokens": 44285928.0,
      "eval_runtime": 27.3313,
      "eval_samples_per_second": 1367.153,
      "eval_steps_per_second": 170.903,
      "step": 24000
    },
    {
      "entropy": 5.580892705917359,
      "epoch": 2.016719176643562,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045911777448579325,
      "loss": 5.1652,
      "mean_token_accuracy": 0.1798853576183319,
      "num_tokens": 44295189.0,
      "step": 24005
    },
    {
      "entropy": 5.6505128860473635,
      "epoch": 2.0171392564587274,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045910055988927827,
      "loss": 5.2673,
      "mean_token_accuracy": 0.173286309838295,
      "num_tokens": 44305757.0,
      "step": 24010
    },
    {
      "entropy": 5.588227319717407,
      "epoch": 2.017559336273892,
      "grad_norm": 1.25,
      "learning_rate": 0.0004590833420315169,
      "loss": 5.2778,
      "mean_token_accuracy": 0.17095820009708404,
      "num_tokens": 44314661.0,
      "step": 24015
    },
    {
      "entropy": 5.637849998474121,
      "epoch": 2.017979416089057,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045906612091281367,
      "loss": 5.3357,
      "mean_token_accuracy": 0.17593427151441574,
      "num_tokens": 44324056.0,
      "step": 24020
    },
    {
      "entropy": 5.560318803787231,
      "epoch": 2.0183994959042217,
      "grad_norm": 1.546875,
      "learning_rate": 0.000459048896533474,
      "loss": 5.1611,
      "mean_token_accuracy": 0.1816700890660286,
      "num_tokens": 44332750.0,
      "step": 24025
    },
    {
      "entropy": 5.5846991539001465,
      "epoch": 2.0188195757193865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045903166889380286,
      "loss": 5.2415,
      "mean_token_accuracy": 0.18128395974636077,
      "num_tokens": 44341808.0,
      "step": 24030
    },
    {
      "entropy": 5.522727537155151,
      "epoch": 2.0192396555345518,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004590144379941056,
      "loss": 5.1938,
      "mean_token_accuracy": 0.1836376190185547,
      "num_tokens": 44350740.0,
      "step": 24035
    },
    {
      "entropy": 5.63276858329773,
      "epoch": 2.0196597353497165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004589972038346872,
      "loss": 5.3333,
      "mean_token_accuracy": 0.1727972224354744,
      "num_tokens": 44360100.0,
      "step": 24040
    },
    {
      "entropy": 5.620118188858032,
      "epoch": 2.0200798151648813,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004589799664158531,
      "loss": 5.182,
      "mean_token_accuracy": 0.18031061738729476,
      "num_tokens": 44369448.0,
      "step": 24045
    },
    {
      "entropy": 5.636735486984253,
      "epoch": 2.020499894980046,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004589627257379086,
      "loss": 5.1939,
      "mean_token_accuracy": 0.18703571259975432,
      "num_tokens": 44378286.0,
      "step": 24050
    },
    {
      "entropy": 5.649603223800659,
      "epoch": 2.020919974795211,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004589454818011592,
      "loss": 5.465,
      "mean_token_accuracy": 0.17218211889266968,
      "num_tokens": 44389419.0,
      "step": 24055
    },
    {
      "entropy": 5.757102060317993,
      "epoch": 2.021340054610376,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045892823460591027,
      "loss": 5.3723,
      "mean_token_accuracy": 0.16836898028850555,
      "num_tokens": 44400509.0,
      "step": 24060
    },
    {
      "entropy": 5.675226879119873,
      "epoch": 2.021760134425541,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045891098415246735,
      "loss": 5.3042,
      "mean_token_accuracy": 0.18032456785440446,
      "num_tokens": 44409638.0,
      "step": 24065
    },
    {
      "entropy": 5.495266103744507,
      "epoch": 2.0221802142407057,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004588937304411361,
      "loss": 5.1453,
      "mean_token_accuracy": 0.1939349964261055,
      "num_tokens": 44418367.0,
      "step": 24070
    },
    {
      "entropy": 5.4988878726959225,
      "epoch": 2.0226002940558705,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045887647347222214,
      "loss": 5.2018,
      "mean_token_accuracy": 0.18447556346654892,
      "num_tokens": 44427267.0,
      "step": 24075
    },
    {
      "entropy": 5.558732366561889,
      "epoch": 2.0230203738710353,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000458859213246031,
      "loss": 5.2121,
      "mean_token_accuracy": 0.18106286823749543,
      "num_tokens": 44437301.0,
      "step": 24080
    },
    {
      "entropy": 5.5994490623474125,
      "epoch": 2.0234404536862005,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004588419497628687,
      "loss": 5.2824,
      "mean_token_accuracy": 0.17776345759630202,
      "num_tokens": 44447041.0,
      "step": 24085
    },
    {
      "entropy": 5.579372835159302,
      "epoch": 2.0238605335013653,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004588246830230409,
      "loss": 5.2111,
      "mean_token_accuracy": 0.18483465015888215,
      "num_tokens": 44456366.0,
      "step": 24090
    },
    {
      "entropy": 5.524551773071289,
      "epoch": 2.02428061331653,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004588074130268535,
      "loss": 5.1701,
      "mean_token_accuracy": 0.18263016045093536,
      "num_tokens": 44465092.0,
      "step": 24095
    },
    {
      "entropy": 5.550295305252075,
      "epoch": 2.024700693131695,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004587901397746124,
      "loss": 5.2204,
      "mean_token_accuracy": 0.17631825506687165,
      "num_tokens": 44474202.0,
      "step": 24100
    },
    {
      "entropy": 5.599558687210083,
      "epoch": 2.02512077294686,
      "grad_norm": 1.375,
      "learning_rate": 0.0004587728632666236,
      "loss": 5.2857,
      "mean_token_accuracy": 0.175230011343956,
      "num_tokens": 44483466.0,
      "step": 24105
    },
    {
      "entropy": 5.502787494659424,
      "epoch": 2.025540852762025,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004587555835031931,
      "loss": 5.1472,
      "mean_token_accuracy": 0.18480588495731354,
      "num_tokens": 44492046.0,
      "step": 24110
    },
    {
      "entropy": 5.568628358840942,
      "epoch": 2.0259609325771897,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004587383004846272,
      "loss": 5.2668,
      "mean_token_accuracy": 0.17492343932390214,
      "num_tokens": 44502657.0,
      "step": 24115
    },
    {
      "entropy": 5.582848978042603,
      "epoch": 2.0263810123923545,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004587210142112318,
      "loss": 5.1572,
      "mean_token_accuracy": 0.18730162233114242,
      "num_tokens": 44512144.0,
      "step": 24120
    },
    {
      "entropy": 5.62736873626709,
      "epoch": 2.0268010922075193,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004587037246833133,
      "loss": 5.2712,
      "mean_token_accuracy": 0.1746353566646576,
      "num_tokens": 44522193.0,
      "step": 24125
    },
    {
      "entropy": 5.534616470336914,
      "epoch": 2.0272211720226845,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004586864319011779,
      "loss": 5.2136,
      "mean_token_accuracy": 0.18072507083415984,
      "num_tokens": 44531112.0,
      "step": 24130
    },
    {
      "entropy": 5.542225694656372,
      "epoch": 2.0276412518378493,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045866913586513184,
      "loss": 5.1793,
      "mean_token_accuracy": 0.17790348529815675,
      "num_tokens": 44540558.0,
      "step": 24135
    },
    {
      "entropy": 5.516228771209716,
      "epoch": 2.028061331653014,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004586518365754818,
      "loss": 5.1406,
      "mean_token_accuracy": 0.18717459291219712,
      "num_tokens": 44549592.0,
      "step": 24140
    },
    {
      "entropy": 5.581379175186157,
      "epoch": 2.028481411468179,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045863453403253393,
      "loss": 5.2547,
      "mean_token_accuracy": 0.18513282090425492,
      "num_tokens": 44559551.0,
      "step": 24145
    },
    {
      "entropy": 5.545409727096557,
      "epoch": 2.0289014912833436,
      "grad_norm": 1.484375,
      "learning_rate": 0.00045861722823659486,
      "loss": 5.2392,
      "mean_token_accuracy": 0.18263183683156967,
      "num_tokens": 44568425.0,
      "step": 24150
    },
    {
      "entropy": 5.597374296188354,
      "epoch": 2.029321571098509,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045859991918797114,
      "loss": 5.2773,
      "mean_token_accuracy": 0.17473794221878053,
      "num_tokens": 44577706.0,
      "step": 24155
    },
    {
      "entropy": 5.5031280517578125,
      "epoch": 2.0297416509136736,
      "grad_norm": 1.5,
      "learning_rate": 0.00045858260688696947,
      "loss": 5.1727,
      "mean_token_accuracy": 0.18893363922834397,
      "num_tokens": 44586926.0,
      "step": 24160
    },
    {
      "entropy": 5.669294738769532,
      "epoch": 2.0301617307288384,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045856529133389627,
      "loss": 5.3039,
      "mean_token_accuracy": 0.17815970331430436,
      "num_tokens": 44596541.0,
      "step": 24165
    },
    {
      "entropy": 5.537741184234619,
      "epoch": 2.030581810544003,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004585479725290586,
      "loss": 5.2326,
      "mean_token_accuracy": 0.1836979866027832,
      "num_tokens": 44605111.0,
      "step": 24170
    },
    {
      "entropy": 5.6495153427124025,
      "epoch": 2.0310018903591684,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00045853065047276307,
      "loss": 5.3776,
      "mean_token_accuracy": 0.1673561617732048,
      "num_tokens": 44614849.0,
      "step": 24175
    },
    {
      "entropy": 5.537764358520508,
      "epoch": 2.0314219701743332,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045851332516531657,
      "loss": 5.2073,
      "mean_token_accuracy": 0.17983455657958985,
      "num_tokens": 44623903.0,
      "step": 24180
    },
    {
      "entropy": 5.6171129703521725,
      "epoch": 2.031842049989498,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000458495996607026,
      "loss": 5.2732,
      "mean_token_accuracy": 0.17854902744293213,
      "num_tokens": 44633113.0,
      "step": 24185
    },
    {
      "entropy": 5.666884279251098,
      "epoch": 2.032262129804663,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004584786647981984,
      "loss": 5.2461,
      "mean_token_accuracy": 0.18106930702924728,
      "num_tokens": 44642121.0,
      "step": 24190
    },
    {
      "entropy": 5.6153614044189455,
      "epoch": 2.0326822096198276,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004584613297391407,
      "loss": 5.2897,
      "mean_token_accuracy": 0.17714892327785492,
      "num_tokens": 44651836.0,
      "step": 24195
    },
    {
      "entropy": 5.469574737548828,
      "epoch": 2.033102289434993,
      "grad_norm": 1.5,
      "learning_rate": 0.0004584439914301599,
      "loss": 5.1937,
      "mean_token_accuracy": 0.18004366308450698,
      "num_tokens": 44660606.0,
      "step": 24200
    },
    {
      "entropy": 5.6829029560089115,
      "epoch": 2.0335223692501576,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004584266498715634,
      "loss": 5.2686,
      "mean_token_accuracy": 0.18643833249807357,
      "num_tokens": 44669753.0,
      "step": 24205
    },
    {
      "entropy": 5.535539865493774,
      "epoch": 2.0339424490653224,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045840930506365815,
      "loss": 5.096,
      "mean_token_accuracy": 0.18480514287948607,
      "num_tokens": 44677795.0,
      "step": 24210
    },
    {
      "entropy": 5.492590236663818,
      "epoch": 2.034362528880487,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045839195700675156,
      "loss": 5.178,
      "mean_token_accuracy": 0.18101312816143036,
      "num_tokens": 44687706.0,
      "step": 24215
    },
    {
      "entropy": 5.6262870788574215,
      "epoch": 2.034782608695652,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004583746057011509,
      "loss": 5.2438,
      "mean_token_accuracy": 0.1870307981967926,
      "num_tokens": 44696238.0,
      "step": 24220
    },
    {
      "entropy": 5.569596481323242,
      "epoch": 2.035202688510817,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045835725114716345,
      "loss": 5.172,
      "mean_token_accuracy": 0.18598684519529343,
      "num_tokens": 44705461.0,
      "step": 24225
    },
    {
      "entropy": 5.619994354248047,
      "epoch": 2.035622768325982,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004583398933450967,
      "loss": 5.2725,
      "mean_token_accuracy": 0.18229412734508516,
      "num_tokens": 44714469.0,
      "step": 24230
    },
    {
      "entropy": 5.5049072265625,
      "epoch": 2.0360428481411468,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004583225322952583,
      "loss": 5.2122,
      "mean_token_accuracy": 0.18438867777585982,
      "num_tokens": 44723681.0,
      "step": 24235
    },
    {
      "entropy": 5.484808349609375,
      "epoch": 2.0364629279563116,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004583051679979555,
      "loss": 5.1058,
      "mean_token_accuracy": 0.18528782576322556,
      "num_tokens": 44732521.0,
      "step": 24240
    },
    {
      "entropy": 5.681520271301269,
      "epoch": 2.036883007771477,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004582878004534961,
      "loss": 5.3046,
      "mean_token_accuracy": 0.17870566993951797,
      "num_tokens": 44741513.0,
      "step": 24245
    },
    {
      "entropy": 5.628424215316772,
      "epoch": 2.0373030875866416,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045827042966218765,
      "loss": 5.1645,
      "mean_token_accuracy": 0.18668244630098343,
      "num_tokens": 44750279.0,
      "step": 24250
    },
    {
      "entropy": 5.602957630157471,
      "epoch": 2.0377231674018064,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004582530556243379,
      "loss": 5.3249,
      "mean_token_accuracy": 0.17166158854961394,
      "num_tokens": 44758787.0,
      "step": 24255
    },
    {
      "entropy": 5.501754474639893,
      "epoch": 2.038143247216971,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045823567834025477,
      "loss": 5.1649,
      "mean_token_accuracy": 0.18621571958065034,
      "num_tokens": 44767247.0,
      "step": 24260
    },
    {
      "entropy": 5.529826593399048,
      "epoch": 2.038563327032136,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045821829781024574,
      "loss": 5.2606,
      "mean_token_accuracy": 0.1758716180920601,
      "num_tokens": 44776568.0,
      "step": 24265
    },
    {
      "entropy": 5.623200798034668,
      "epoch": 2.038983406847301,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045820091403461907,
      "loss": 5.2696,
      "mean_token_accuracy": 0.17897906005382538,
      "num_tokens": 44785757.0,
      "step": 24270
    },
    {
      "entropy": 5.650467777252198,
      "epoch": 2.039403486662466,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045818352701368245,
      "loss": 5.2505,
      "mean_token_accuracy": 0.18044525682926177,
      "num_tokens": 44794390.0,
      "step": 24275
    },
    {
      "entropy": 5.642752552032471,
      "epoch": 2.0398235664776307,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045816613674774396,
      "loss": 5.2361,
      "mean_token_accuracy": 0.18200088590383529,
      "num_tokens": 44802909.0,
      "step": 24280
    },
    {
      "entropy": 5.537203121185303,
      "epoch": 2.0402436462927955,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045814874323711174,
      "loss": 5.2447,
      "mean_token_accuracy": 0.1758776530623436,
      "num_tokens": 44811844.0,
      "step": 24285
    },
    {
      "entropy": 5.568144989013672,
      "epoch": 2.0406637261079603,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004581313464820938,
      "loss": 5.1936,
      "mean_token_accuracy": 0.18509511053562164,
      "num_tokens": 44820661.0,
      "step": 24290
    },
    {
      "entropy": 5.631122016906739,
      "epoch": 2.0410838059231255,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004581139464829985,
      "loss": 5.2856,
      "mean_token_accuracy": 0.17662902772426606,
      "num_tokens": 44830185.0,
      "step": 24295
    },
    {
      "entropy": 5.547418928146362,
      "epoch": 2.0415038857382903,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004580965432401337,
      "loss": 5.1973,
      "mean_token_accuracy": 0.17874108850955964,
      "num_tokens": 44839546.0,
      "step": 24300
    },
    {
      "entropy": 5.576404809951782,
      "epoch": 2.041923965553455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000458079136753808,
      "loss": 5.2264,
      "mean_token_accuracy": 0.1850030317902565,
      "num_tokens": 44848824.0,
      "step": 24305
    },
    {
      "entropy": 5.622378158569336,
      "epoch": 2.04234404536862,
      "grad_norm": 1.25,
      "learning_rate": 0.00045806172702432974,
      "loss": 5.1205,
      "mean_token_accuracy": 0.1887942746281624,
      "num_tokens": 44857110.0,
      "step": 24310
    },
    {
      "entropy": 5.6489293575286865,
      "epoch": 2.042764125183785,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045804431405200717,
      "loss": 5.2972,
      "mean_token_accuracy": 0.17722827941179276,
      "num_tokens": 44866429.0,
      "step": 24315
    },
    {
      "entropy": 5.574634695053101,
      "epoch": 2.04318420499895,
      "grad_norm": 2.140625,
      "learning_rate": 0.00045802689783714886,
      "loss": 5.301,
      "mean_token_accuracy": 0.18373814225196838,
      "num_tokens": 44874893.0,
      "step": 24320
    },
    {
      "entropy": 5.5713441371917725,
      "epoch": 2.0436042848141147,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045800947838006326,
      "loss": 5.2563,
      "mean_token_accuracy": 0.1825754836201668,
      "num_tokens": 44883778.0,
      "step": 24325
    },
    {
      "entropy": 5.720736837387085,
      "epoch": 2.0440243646292795,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045799205568105905,
      "loss": 5.3805,
      "mean_token_accuracy": 0.17865391522645951,
      "num_tokens": 44892996.0,
      "step": 24330
    },
    {
      "entropy": 5.589629364013672,
      "epoch": 2.0444444444444443,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045797462974044473,
      "loss": 5.276,
      "mean_token_accuracy": 0.17736382335424422,
      "num_tokens": 44901752.0,
      "step": 24335
    },
    {
      "entropy": 5.572343158721924,
      "epoch": 2.0448645242596095,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045795720055852914,
      "loss": 5.3249,
      "mean_token_accuracy": 0.17196996957063676,
      "num_tokens": 44912078.0,
      "step": 24340
    },
    {
      "entropy": 5.761699438095093,
      "epoch": 2.0452846040747743,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004579397681356209,
      "loss": 5.3807,
      "mean_token_accuracy": 0.17622655928134917,
      "num_tokens": 44921849.0,
      "step": 24345
    },
    {
      "entropy": 5.606022930145263,
      "epoch": 2.045704683889939,
      "grad_norm": 1.125,
      "learning_rate": 0.00045792233247202884,
      "loss": 5.233,
      "mean_token_accuracy": 0.18701769560575485,
      "num_tokens": 44931696.0,
      "step": 24350
    },
    {
      "entropy": 5.603670358657837,
      "epoch": 2.046124763705104,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045790489356806196,
      "loss": 5.2877,
      "mean_token_accuracy": 0.1727035164833069,
      "num_tokens": 44940702.0,
      "step": 24355
    },
    {
      "entropy": 5.570962762832641,
      "epoch": 2.0465448435202687,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045788745142402894,
      "loss": 5.2269,
      "mean_token_accuracy": 0.18168613016605378,
      "num_tokens": 44950468.0,
      "step": 24360
    },
    {
      "entropy": 5.541432476043701,
      "epoch": 2.046964923335434,
      "grad_norm": 1.234375,
      "learning_rate": 0.000457870006040239,
      "loss": 5.199,
      "mean_token_accuracy": 0.18538538962602616,
      "num_tokens": 44960524.0,
      "step": 24365
    },
    {
      "entropy": 5.556615114212036,
      "epoch": 2.0473850031505987,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045785255741700103,
      "loss": 5.1973,
      "mean_token_accuracy": 0.18668646812438966,
      "num_tokens": 44969542.0,
      "step": 24370
    },
    {
      "entropy": 5.489857530593872,
      "epoch": 2.0478050829657635,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004578351055546241,
      "loss": 5.1893,
      "mean_token_accuracy": 0.1853371351957321,
      "num_tokens": 44979033.0,
      "step": 24375
    },
    {
      "entropy": 5.500544452667237,
      "epoch": 2.0482251627809283,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045781765045341743,
      "loss": 5.1732,
      "mean_token_accuracy": 0.18391168266534805,
      "num_tokens": 44987775.0,
      "step": 24380
    },
    {
      "entropy": 5.564240550994873,
      "epoch": 2.048645242596093,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004578001921136903,
      "loss": 5.1823,
      "mean_token_accuracy": 0.18386925011873245,
      "num_tokens": 44996232.0,
      "step": 24385
    },
    {
      "entropy": 5.558294057846069,
      "epoch": 2.0490653224112583,
      "grad_norm": 1.25,
      "learning_rate": 0.0004577827305357519,
      "loss": 5.2187,
      "mean_token_accuracy": 0.1824587568640709,
      "num_tokens": 45005923.0,
      "step": 24390
    },
    {
      "entropy": 5.616902112960815,
      "epoch": 2.049485402226423,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045776526571991147,
      "loss": 5.203,
      "mean_token_accuracy": 0.1801037847995758,
      "num_tokens": 45014684.0,
      "step": 24395
    },
    {
      "entropy": 5.565790128707886,
      "epoch": 2.049905482041588,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045774779766647854,
      "loss": 5.1984,
      "mean_token_accuracy": 0.18141575604677201,
      "num_tokens": 45023729.0,
      "step": 24400
    },
    {
      "entropy": 5.66400408744812,
      "epoch": 2.0503255618567526,
      "grad_norm": 1.375,
      "learning_rate": 0.0004577303263757624,
      "loss": 5.3392,
      "mean_token_accuracy": 0.17666908651590346,
      "num_tokens": 45034064.0,
      "step": 24405
    },
    {
      "entropy": 5.6612437725067135,
      "epoch": 2.050745641671918,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045771285184807264,
      "loss": 5.2144,
      "mean_token_accuracy": 0.1818854808807373,
      "num_tokens": 45043650.0,
      "step": 24410
    },
    {
      "entropy": 5.428490209579468,
      "epoch": 2.0511657214870826,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045769537408371885,
      "loss": 5.1325,
      "mean_token_accuracy": 0.17986335307359697,
      "num_tokens": 45052925.0,
      "step": 24415
    },
    {
      "entropy": 5.558942985534668,
      "epoch": 2.0515858013022474,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045767789308301057,
      "loss": 5.2558,
      "mean_token_accuracy": 0.17838992029428483,
      "num_tokens": 45062505.0,
      "step": 24420
    },
    {
      "entropy": 5.57352032661438,
      "epoch": 2.0520058811174122,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004576604088462575,
      "loss": 5.2696,
      "mean_token_accuracy": 0.1791820377111435,
      "num_tokens": 45072134.0,
      "step": 24425
    },
    {
      "entropy": 5.635435152053833,
      "epoch": 2.052425960932577,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045764292137376924,
      "loss": 5.2247,
      "mean_token_accuracy": 0.1791029155254364,
      "num_tokens": 45081438.0,
      "step": 24430
    },
    {
      "entropy": 5.609773302078247,
      "epoch": 2.0528460407477422,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004576254306658557,
      "loss": 5.214,
      "mean_token_accuracy": 0.17731148898601531,
      "num_tokens": 45091601.0,
      "step": 24435
    },
    {
      "entropy": 5.545984888076783,
      "epoch": 2.053266120562907,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045760793672282676,
      "loss": 5.2322,
      "mean_token_accuracy": 0.18312591314315796,
      "num_tokens": 45101721.0,
      "step": 24440
    },
    {
      "entropy": 5.587225770950317,
      "epoch": 2.053686200378072,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004575904395449921,
      "loss": 5.3445,
      "mean_token_accuracy": 0.17362598776817323,
      "num_tokens": 45110614.0,
      "step": 24445
    },
    {
      "entropy": 5.605717658996582,
      "epoch": 2.0541062801932366,
      "grad_norm": 1.15625,
      "learning_rate": 0.000457572939132662,
      "loss": 5.205,
      "mean_token_accuracy": 0.18611571937799454,
      "num_tokens": 45120103.0,
      "step": 24450
    },
    {
      "entropy": 5.641632080078125,
      "epoch": 2.0545263600084014,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045755543548614623,
      "loss": 5.2136,
      "mean_token_accuracy": 0.18680927753448487,
      "num_tokens": 45128811.0,
      "step": 24455
    },
    {
      "entropy": 5.527692890167236,
      "epoch": 2.0549464398235666,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000457537928605755,
      "loss": 5.1198,
      "mean_token_accuracy": 0.1895752042531967,
      "num_tokens": 45137351.0,
      "step": 24460
    },
    {
      "entropy": 5.660110330581665,
      "epoch": 2.0553665196387314,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045752041849179823,
      "loss": 5.2321,
      "mean_token_accuracy": 0.18181769102811812,
      "num_tokens": 45146504.0,
      "step": 24465
    },
    {
      "entropy": 5.536942481994629,
      "epoch": 2.055786599453896,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004575029051445863,
      "loss": 5.1922,
      "mean_token_accuracy": 0.18739402741193772,
      "num_tokens": 45155671.0,
      "step": 24470
    },
    {
      "entropy": 5.660780954360962,
      "epoch": 2.056206679269061,
      "grad_norm": 1.375,
      "learning_rate": 0.0004574853885644293,
      "loss": 5.3292,
      "mean_token_accuracy": 0.17006754428148269,
      "num_tokens": 45164482.0,
      "step": 24475
    },
    {
      "entropy": 5.645761489868164,
      "epoch": 2.056626759084226,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004574678687516377,
      "loss": 5.3035,
      "mean_token_accuracy": 0.17532514035701752,
      "num_tokens": 45174185.0,
      "step": 24480
    },
    {
      "entropy": 5.618024301528931,
      "epoch": 2.057046838899391,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004574503457065217,
      "loss": 5.252,
      "mean_token_accuracy": 0.1789383754134178,
      "num_tokens": 45184498.0,
      "step": 24485
    },
    {
      "entropy": 5.623136901855469,
      "epoch": 2.057466918714556,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004574328194293919,
      "loss": 5.2752,
      "mean_token_accuracy": 0.18183509707450868,
      "num_tokens": 45194297.0,
      "step": 24490
    },
    {
      "entropy": 5.557909250259399,
      "epoch": 2.0578869985297206,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004574152899205585,
      "loss": 5.2129,
      "mean_token_accuracy": 0.1737196624279022,
      "num_tokens": 45204930.0,
      "step": 24495
    },
    {
      "entropy": 5.52739839553833,
      "epoch": 2.0583070783448854,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004573977571803322,
      "loss": 5.2167,
      "mean_token_accuracy": 0.18684130162000656,
      "num_tokens": 45213569.0,
      "step": 24500
    },
    {
      "entropy": 5.57692403793335,
      "epoch": 2.0587271581600506,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045738022120902355,
      "loss": 5.2152,
      "mean_token_accuracy": 0.1801399201154709,
      "num_tokens": 45223501.0,
      "step": 24505
    },
    {
      "entropy": 5.54924955368042,
      "epoch": 2.0591472379752154,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004573626820069433,
      "loss": 5.2384,
      "mean_token_accuracy": 0.18109158128499986,
      "num_tokens": 45231919.0,
      "step": 24510
    },
    {
      "entropy": 5.511261081695556,
      "epoch": 2.05956731779038,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004573451395744019,
      "loss": 5.1382,
      "mean_token_accuracy": 0.18450634479522704,
      "num_tokens": 45240313.0,
      "step": 24515
    },
    {
      "entropy": 5.550677871704101,
      "epoch": 2.059987397605545,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004573275939117104,
      "loss": 5.2498,
      "mean_token_accuracy": 0.17467723190784454,
      "num_tokens": 45249509.0,
      "step": 24520
    },
    {
      "entropy": 5.559187507629394,
      "epoch": 2.0604074774207097,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004573100450191793,
      "loss": 5.2603,
      "mean_token_accuracy": 0.17894843220710754,
      "num_tokens": 45258644.0,
      "step": 24525
    },
    {
      "entropy": 5.59469838142395,
      "epoch": 2.060827557235875,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045729249289711964,
      "loss": 5.2664,
      "mean_token_accuracy": 0.18331752717494965,
      "num_tokens": 45267898.0,
      "step": 24530
    },
    {
      "entropy": 5.6508166790008545,
      "epoch": 2.0612476370510397,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045727493754584237,
      "loss": 5.3102,
      "mean_token_accuracy": 0.17394756227731706,
      "num_tokens": 45278460.0,
      "step": 24535
    },
    {
      "entropy": 5.585290575027466,
      "epoch": 2.0616677168662045,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004572573789656584,
      "loss": 5.1884,
      "mean_token_accuracy": 0.19330764710903167,
      "num_tokens": 45287850.0,
      "step": 24540
    },
    {
      "entropy": 5.4788306713104244,
      "epoch": 2.0620877966813693,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004572398171568789,
      "loss": 5.2319,
      "mean_token_accuracy": 0.18234034180641173,
      "num_tokens": 45297835.0,
      "step": 24545
    },
    {
      "entropy": 5.5157910823822025,
      "epoch": 2.0625078764965346,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004572222521198149,
      "loss": 5.1985,
      "mean_token_accuracy": 0.18824953883886336,
      "num_tokens": 45305439.0,
      "step": 24550
    },
    {
      "entropy": 5.648874664306641,
      "epoch": 2.0629279563116993,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045720468385477745,
      "loss": 5.2826,
      "mean_token_accuracy": 0.1790318086743355,
      "num_tokens": 45314445.0,
      "step": 24555
    },
    {
      "entropy": 5.522325754165649,
      "epoch": 2.063348036126864,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004571871123620778,
      "loss": 5.152,
      "mean_token_accuracy": 0.18670125007629396,
      "num_tokens": 45323402.0,
      "step": 24560
    },
    {
      "entropy": 5.500130128860474,
      "epoch": 2.063768115942029,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004571695376420274,
      "loss": 5.2864,
      "mean_token_accuracy": 0.17911235392093658,
      "num_tokens": 45332898.0,
      "step": 24565
    },
    {
      "entropy": 5.593569898605347,
      "epoch": 2.0641881957571937,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004571519596949374,
      "loss": 5.2417,
      "mean_token_accuracy": 0.17707732766866685,
      "num_tokens": 45342470.0,
      "step": 24570
    },
    {
      "entropy": 5.572461414337158,
      "epoch": 2.064608275572359,
      "grad_norm": 1.375,
      "learning_rate": 0.0004571343785211192,
      "loss": 5.1941,
      "mean_token_accuracy": 0.18562103658914567,
      "num_tokens": 45351241.0,
      "step": 24575
    },
    {
      "entropy": 5.4869222164154055,
      "epoch": 2.0650283553875237,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004571167941208843,
      "loss": 5.1128,
      "mean_token_accuracy": 0.19778698831796646,
      "num_tokens": 45360002.0,
      "step": 24580
    },
    {
      "entropy": 5.568949413299561,
      "epoch": 2.0654484352026885,
      "grad_norm": 1.25,
      "learning_rate": 0.0004570992064945441,
      "loss": 5.2937,
      "mean_token_accuracy": 0.17980446070432662,
      "num_tokens": 45369326.0,
      "step": 24585
    },
    {
      "entropy": 5.566601324081421,
      "epoch": 2.0658685150178533,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004570816156424103,
      "loss": 5.2758,
      "mean_token_accuracy": 0.18163792490959169,
      "num_tokens": 45378987.0,
      "step": 24590
    },
    {
      "entropy": 5.668241405487061,
      "epoch": 2.066288594833018,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004570640215647944,
      "loss": 5.3797,
      "mean_token_accuracy": 0.17765216678380966,
      "num_tokens": 45388845.0,
      "step": 24595
    },
    {
      "entropy": 5.565196943283081,
      "epoch": 2.0667086746481833,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004570464242620081,
      "loss": 5.1415,
      "mean_token_accuracy": 0.17886368185281754,
      "num_tokens": 45397794.0,
      "step": 24600
    },
    {
      "entropy": 5.557122421264649,
      "epoch": 2.067128754463348,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045702882373436317,
      "loss": 5.2033,
      "mean_token_accuracy": 0.17947273701429367,
      "num_tokens": 45406318.0,
      "step": 24605
    },
    {
      "entropy": 5.575728273391723,
      "epoch": 2.067548834278513,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004570112199821713,
      "loss": 5.3059,
      "mean_token_accuracy": 0.17552610337734223,
      "num_tokens": 45415573.0,
      "step": 24610
    },
    {
      "entropy": 5.501656866073608,
      "epoch": 2.0679689140936777,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045699361300574447,
      "loss": 5.1138,
      "mean_token_accuracy": 0.18627094030380248,
      "num_tokens": 45424607.0,
      "step": 24615
    },
    {
      "entropy": 5.607026767730713,
      "epoch": 2.068388993908843,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004569760028053944,
      "loss": 5.2574,
      "mean_token_accuracy": 0.17425691187381745,
      "num_tokens": 45434031.0,
      "step": 24620
    },
    {
      "entropy": 5.615068960189819,
      "epoch": 2.0688090737240077,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004569583893814333,
      "loss": 5.2849,
      "mean_token_accuracy": 0.17995006293058396,
      "num_tokens": 45443158.0,
      "step": 24625
    },
    {
      "entropy": 5.594398927688599,
      "epoch": 2.0692291535391725,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045694077273417295,
      "loss": 5.2169,
      "mean_token_accuracy": 0.17951955944299697,
      "num_tokens": 45452500.0,
      "step": 24630
    },
    {
      "entropy": 5.5129790782928465,
      "epoch": 2.0696492333543373,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004569231528639254,
      "loss": 5.18,
      "mean_token_accuracy": 0.1865200862288475,
      "num_tokens": 45461613.0,
      "step": 24635
    },
    {
      "entropy": 5.618472242355347,
      "epoch": 2.070069313169502,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045690552977100296,
      "loss": 5.3265,
      "mean_token_accuracy": 0.17714642733335495,
      "num_tokens": 45470799.0,
      "step": 24640
    },
    {
      "entropy": 5.623915338516236,
      "epoch": 2.0704893929846673,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045688790345571774,
      "loss": 5.2498,
      "mean_token_accuracy": 0.17689600735902786,
      "num_tokens": 45480780.0,
      "step": 24645
    },
    {
      "entropy": 5.460437631607055,
      "epoch": 2.070909472799832,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004568702739183819,
      "loss": 5.1373,
      "mean_token_accuracy": 0.18301567286252976,
      "num_tokens": 45489401.0,
      "step": 24650
    },
    {
      "entropy": 5.530411148071289,
      "epoch": 2.071329552614997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004568526411593079,
      "loss": 5.2187,
      "mean_token_accuracy": 0.1799318626523018,
      "num_tokens": 45498420.0,
      "step": 24655
    },
    {
      "entropy": 5.579149580001831,
      "epoch": 2.0717496324301616,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045683500517880796,
      "loss": 5.1726,
      "mean_token_accuracy": 0.18592991530895234,
      "num_tokens": 45507417.0,
      "step": 24660
    },
    {
      "entropy": 5.604370832443237,
      "epoch": 2.0721697122453264,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00045681736597719455,
      "loss": 5.3219,
      "mean_token_accuracy": 0.1745927855372429,
      "num_tokens": 45516518.0,
      "step": 24665
    },
    {
      "entropy": 5.555382919311524,
      "epoch": 2.0725897920604917,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00045679972355478003,
      "loss": 5.2308,
      "mean_token_accuracy": 0.17780272662639618,
      "num_tokens": 45526000.0,
      "step": 24670
    },
    {
      "entropy": 5.632692432403564,
      "epoch": 2.0730098718756564,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045678207791187707,
      "loss": 5.3228,
      "mean_token_accuracy": 0.17965226471424103,
      "num_tokens": 45534429.0,
      "step": 24675
    },
    {
      "entropy": 5.600115442276001,
      "epoch": 2.0734299516908212,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004567644290487983,
      "loss": 5.3045,
      "mean_token_accuracy": 0.17200569063425064,
      "num_tokens": 45543165.0,
      "step": 24680
    },
    {
      "entropy": 5.624059104919434,
      "epoch": 2.073850031505986,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045674677696585614,
      "loss": 5.221,
      "mean_token_accuracy": 0.18597962856292724,
      "num_tokens": 45551972.0,
      "step": 24685
    },
    {
      "entropy": 5.62484221458435,
      "epoch": 2.074270111321151,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004567291216633635,
      "loss": 5.2476,
      "mean_token_accuracy": 0.1830361306667328,
      "num_tokens": 45560903.0,
      "step": 24690
    },
    {
      "entropy": 5.628470945358276,
      "epoch": 2.074690191136316,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045671146314163295,
      "loss": 5.3988,
      "mean_token_accuracy": 0.17515729665756224,
      "num_tokens": 45571117.0,
      "step": 24695
    },
    {
      "entropy": 5.589072370529175,
      "epoch": 2.075110270951481,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045669380140097747,
      "loss": 5.2202,
      "mean_token_accuracy": 0.1762699380517006,
      "num_tokens": 45579961.0,
      "step": 24700
    },
    {
      "entropy": 5.633452224731445,
      "epoch": 2.0755303507666456,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004566761364417099,
      "loss": 5.1873,
      "mean_token_accuracy": 0.17699847519397735,
      "num_tokens": 45587968.0,
      "step": 24705
    },
    {
      "entropy": 5.494909715652466,
      "epoch": 2.0759504305818104,
      "grad_norm": 1.359375,
      "learning_rate": 0.000456658468264143,
      "loss": 5.1706,
      "mean_token_accuracy": 0.1861261323094368,
      "num_tokens": 45597410.0,
      "step": 24710
    },
    {
      "entropy": 5.478945112228393,
      "epoch": 2.0763705103969756,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045664079686858996,
      "loss": 5.1545,
      "mean_token_accuracy": 0.1823420152068138,
      "num_tokens": 45605545.0,
      "step": 24715
    },
    {
      "entropy": 5.624455833435059,
      "epoch": 2.0767905902121404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045662312225536373,
      "loss": 5.284,
      "mean_token_accuracy": 0.18170335739850998,
      "num_tokens": 45614904.0,
      "step": 24720
    },
    {
      "entropy": 5.594312334060669,
      "epoch": 2.077210670027305,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045660544442477737,
      "loss": 5.2627,
      "mean_token_accuracy": 0.18582610934972763,
      "num_tokens": 45624160.0,
      "step": 24725
    },
    {
      "entropy": 5.552125549316406,
      "epoch": 2.07763074984247,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004565877633771441,
      "loss": 5.2269,
      "mean_token_accuracy": 0.1772698238492012,
      "num_tokens": 45632931.0,
      "step": 24730
    },
    {
      "entropy": 5.588324308395386,
      "epoch": 2.0780508296576348,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004565700791127771,
      "loss": 5.2424,
      "mean_token_accuracy": 0.17546268552541733,
      "num_tokens": 45641280.0,
      "step": 24735
    },
    {
      "entropy": 5.553592681884766,
      "epoch": 2.0784709094728,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004565523916319897,
      "loss": 5.1518,
      "mean_token_accuracy": 0.17874136865139006,
      "num_tokens": 45650292.0,
      "step": 24740
    },
    {
      "entropy": 5.634365892410278,
      "epoch": 2.078890989287965,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004565347009350951,
      "loss": 5.2973,
      "mean_token_accuracy": 0.178208526968956,
      "num_tokens": 45659680.0,
      "step": 24745
    },
    {
      "entropy": 5.551399850845337,
      "epoch": 2.0793110691031296,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004565170070224068,
      "loss": 5.2069,
      "mean_token_accuracy": 0.18280464559793472,
      "num_tokens": 45668398.0,
      "step": 24750
    },
    {
      "entropy": 5.460042095184326,
      "epoch": 2.0797311489182944,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045649930989423806,
      "loss": 5.1411,
      "mean_token_accuracy": 0.1903093695640564,
      "num_tokens": 45677221.0,
      "step": 24755
    },
    {
      "entropy": 5.499493980407715,
      "epoch": 2.080151228733459,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004564816095509026,
      "loss": 5.1727,
      "mean_token_accuracy": 0.18767403364181517,
      "num_tokens": 45685946.0,
      "step": 24760
    },
    {
      "entropy": 5.559024238586426,
      "epoch": 2.0805713085486244,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004564639059927139,
      "loss": 5.1957,
      "mean_token_accuracy": 0.18611965477466583,
      "num_tokens": 45694690.0,
      "step": 24765
    },
    {
      "entropy": 5.611278676986695,
      "epoch": 2.080991388363789,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045644619921998556,
      "loss": 5.2765,
      "mean_token_accuracy": 0.17320101261138915,
      "num_tokens": 45704426.0,
      "step": 24770
    },
    {
      "entropy": 5.672424936294556,
      "epoch": 2.081411468178954,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004564284892330311,
      "loss": 5.2843,
      "mean_token_accuracy": 0.1706093892455101,
      "num_tokens": 45714044.0,
      "step": 24775
    },
    {
      "entropy": 5.604932165145874,
      "epoch": 2.0818315479941187,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004564107760321645,
      "loss": 5.2515,
      "mean_token_accuracy": 0.17845078706741332,
      "num_tokens": 45722860.0,
      "step": 24780
    },
    {
      "entropy": 5.540560007095337,
      "epoch": 2.082251627809284,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004563930596176994,
      "loss": 5.2354,
      "mean_token_accuracy": 0.1849985897541046,
      "num_tokens": 45731606.0,
      "step": 24785
    },
    {
      "entropy": 5.520496654510498,
      "epoch": 2.0826717076244488,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004563753399899496,
      "loss": 5.1897,
      "mean_token_accuracy": 0.1822981968522072,
      "num_tokens": 45741189.0,
      "step": 24790
    },
    {
      "entropy": 5.519714736938477,
      "epoch": 2.0830917874396135,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004563576171492291,
      "loss": 5.1293,
      "mean_token_accuracy": 0.18937772363424302,
      "num_tokens": 45750059.0,
      "step": 24795
    },
    {
      "entropy": 5.496574831008911,
      "epoch": 2.0835118672547783,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004563398910958518,
      "loss": 5.1425,
      "mean_token_accuracy": 0.18773368299007415,
      "num_tokens": 45758675.0,
      "step": 24800
    },
    {
      "entropy": 5.523564767837525,
      "epoch": 2.083931947069943,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045632216183013165,
      "loss": 5.2322,
      "mean_token_accuracy": 0.18414737284183502,
      "num_tokens": 45767886.0,
      "step": 24805
    },
    {
      "entropy": 5.50952467918396,
      "epoch": 2.0843520268851083,
      "grad_norm": 1.375,
      "learning_rate": 0.0004563044293523828,
      "loss": 5.1723,
      "mean_token_accuracy": 0.18770441114902497,
      "num_tokens": 45777208.0,
      "step": 24810
    },
    {
      "entropy": 5.513107681274414,
      "epoch": 2.084772106700273,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045628669366291934,
      "loss": 5.1251,
      "mean_token_accuracy": 0.18763090223073958,
      "num_tokens": 45787422.0,
      "step": 24815
    },
    {
      "entropy": 5.7077250480651855,
      "epoch": 2.085192186515438,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004562689547620555,
      "loss": 5.3828,
      "mean_token_accuracy": 0.1718765527009964,
      "num_tokens": 45797748.0,
      "step": 24820
    },
    {
      "entropy": 5.585168313980103,
      "epoch": 2.0856122663306027,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004562512126501054,
      "loss": 5.1788,
      "mean_token_accuracy": 0.18911453187465668,
      "num_tokens": 45806378.0,
      "step": 24825
    },
    {
      "entropy": 5.522868347167969,
      "epoch": 2.0860323461457675,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045623346732738345,
      "loss": 5.2261,
      "mean_token_accuracy": 0.17950569689273835,
      "num_tokens": 45815003.0,
      "step": 24830
    },
    {
      "entropy": 5.490697908401489,
      "epoch": 2.0864524259609327,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004562157187942039,
      "loss": 5.1575,
      "mean_token_accuracy": 0.1838339611887932,
      "num_tokens": 45823839.0,
      "step": 24835
    },
    {
      "entropy": 5.561507320404052,
      "epoch": 2.0868725057760975,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004561979670508812,
      "loss": 5.219,
      "mean_token_accuracy": 0.186343914270401,
      "num_tokens": 45832139.0,
      "step": 24840
    },
    {
      "entropy": 5.53038067817688,
      "epoch": 2.0872925855912623,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045618021209772983,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1863807335495949,
      "num_tokens": 45840792.0,
      "step": 24845
    },
    {
      "entropy": 5.592167472839355,
      "epoch": 2.087712665406427,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004561624539350643,
      "loss": 5.2048,
      "mean_token_accuracy": 0.1782603979110718,
      "num_tokens": 45849522.0,
      "step": 24850
    },
    {
      "entropy": 5.497398948669433,
      "epoch": 2.0881327452215923,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004561446925631992,
      "loss": 5.1737,
      "mean_token_accuracy": 0.1837841659784317,
      "num_tokens": 45858581.0,
      "step": 24855
    },
    {
      "entropy": 5.646769571304321,
      "epoch": 2.088552825036757,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00045612692798244913,
      "loss": 5.2435,
      "mean_token_accuracy": 0.17448375076055528,
      "num_tokens": 45868430.0,
      "step": 24860
    },
    {
      "entropy": 5.529869031906128,
      "epoch": 2.088972904851922,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004561091601931288,
      "loss": 5.2052,
      "mean_token_accuracy": 0.1850036010146141,
      "num_tokens": 45878332.0,
      "step": 24865
    },
    {
      "entropy": 5.5696910381317135,
      "epoch": 2.0893929846670867,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045609138919555295,
      "loss": 5.3418,
      "mean_token_accuracy": 0.1716714859008789,
      "num_tokens": 45887571.0,
      "step": 24870
    },
    {
      "entropy": 5.535857152938843,
      "epoch": 2.0898130644822515,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004560736149900364,
      "loss": 5.1766,
      "mean_token_accuracy": 0.18250393867492676,
      "num_tokens": 45896967.0,
      "step": 24875
    },
    {
      "entropy": 5.542154169082641,
      "epoch": 2.0902331442974167,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045605583757689393,
      "loss": 5.236,
      "mean_token_accuracy": 0.17983919978141785,
      "num_tokens": 45905935.0,
      "step": 24880
    },
    {
      "entropy": 5.565347385406494,
      "epoch": 2.0906532241125815,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004560380569564406,
      "loss": 5.1904,
      "mean_token_accuracy": 0.19417389035224913,
      "num_tokens": 45914202.0,
      "step": 24885
    },
    {
      "entropy": 5.489102125167847,
      "epoch": 2.0910733039277463,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045602027312899134,
      "loss": 5.1368,
      "mean_token_accuracy": 0.19084442555904388,
      "num_tokens": 45923700.0,
      "step": 24890
    },
    {
      "entropy": 5.5385582447052,
      "epoch": 2.091493383742911,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004560024860948611,
      "loss": 5.2829,
      "mean_token_accuracy": 0.1735875904560089,
      "num_tokens": 45932885.0,
      "step": 24895
    },
    {
      "entropy": 5.506993675231934,
      "epoch": 2.091913463558076,
      "grad_norm": 1.484375,
      "learning_rate": 0.000455984695854365,
      "loss": 5.1623,
      "mean_token_accuracy": 0.18380543440580369,
      "num_tokens": 45942039.0,
      "step": 24900
    },
    {
      "entropy": 5.514059543609619,
      "epoch": 2.092333543373241,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004559669024078183,
      "loss": 5.1355,
      "mean_token_accuracy": 0.18962627351284028,
      "num_tokens": 45950893.0,
      "step": 24905
    },
    {
      "entropy": 5.546408843994141,
      "epoch": 2.092753623188406,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000455949105755536,
      "loss": 5.2347,
      "mean_token_accuracy": 0.18201974928379058,
      "num_tokens": 45960225.0,
      "step": 24910
    },
    {
      "entropy": 5.543997049331665,
      "epoch": 2.0931737030035706,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045593130589783356,
      "loss": 5.1642,
      "mean_token_accuracy": 0.18111738860607146,
      "num_tokens": 45969002.0,
      "step": 24915
    },
    {
      "entropy": 5.603867244720459,
      "epoch": 2.0935937828187354,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004559135028350262,
      "loss": 5.3574,
      "mean_token_accuracy": 0.17460740357637405,
      "num_tokens": 45979088.0,
      "step": 24920
    },
    {
      "entropy": 5.614394474029541,
      "epoch": 2.0940138626339007,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004558956965674292,
      "loss": 5.1867,
      "mean_token_accuracy": 0.18271155655384064,
      "num_tokens": 45988308.0,
      "step": 24925
    },
    {
      "entropy": 5.554372262954712,
      "epoch": 2.0944339424490654,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004558778870953582,
      "loss": 5.2233,
      "mean_token_accuracy": 0.18544308394193648,
      "num_tokens": 45997758.0,
      "step": 24930
    },
    {
      "entropy": 5.5245012760162355,
      "epoch": 2.0948540222642302,
      "grad_norm": 1.46875,
      "learning_rate": 0.00045586007441912846,
      "loss": 5.2022,
      "mean_token_accuracy": 0.18122695088386537,
      "num_tokens": 46006648.0,
      "step": 24935
    },
    {
      "entropy": 5.620496368408203,
      "epoch": 2.095274102079395,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045584225853905565,
      "loss": 5.293,
      "mean_token_accuracy": 0.17281267344951629,
      "num_tokens": 46015693.0,
      "step": 24940
    },
    {
      "entropy": 5.648223781585694,
      "epoch": 2.09569418189456,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004558244394554554,
      "loss": 5.2603,
      "mean_token_accuracy": 0.18465479761362075,
      "num_tokens": 46024748.0,
      "step": 24945
    },
    {
      "entropy": 5.57706618309021,
      "epoch": 2.096114261709725,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004558066171686433,
      "loss": 5.2482,
      "mean_token_accuracy": 0.18186000883579254,
      "num_tokens": 46033613.0,
      "step": 24950
    },
    {
      "entropy": 5.611463117599487,
      "epoch": 2.09653434152489,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004557887916789351,
      "loss": 5.3563,
      "mean_token_accuracy": 0.17672063857316972,
      "num_tokens": 46042749.0,
      "step": 24955
    },
    {
      "entropy": 5.614355134963989,
      "epoch": 2.0969544213400546,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045577096298664646,
      "loss": 5.2231,
      "mean_token_accuracy": 0.18338105380535125,
      "num_tokens": 46051969.0,
      "step": 24960
    },
    {
      "entropy": 5.632194852828979,
      "epoch": 2.0973745011552194,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004557531310920934,
      "loss": 5.2465,
      "mean_token_accuracy": 0.18719875812530518,
      "num_tokens": 46061115.0,
      "step": 24965
    },
    {
      "entropy": 5.524462938308716,
      "epoch": 2.097794580970384,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004557352959955916,
      "loss": 5.1938,
      "mean_token_accuracy": 0.1834063783288002,
      "num_tokens": 46071283.0,
      "step": 24970
    },
    {
      "entropy": 5.562889051437378,
      "epoch": 2.0982146607855494,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045571745769745715,
      "loss": 5.2211,
      "mean_token_accuracy": 0.18331777304410934,
      "num_tokens": 46080730.0,
      "step": 24975
    },
    {
      "entropy": 5.529333209991455,
      "epoch": 2.098634740600714,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000455699616198006,
      "loss": 5.1758,
      "mean_token_accuracy": 0.17879880368709564,
      "num_tokens": 46090717.0,
      "step": 24980
    },
    {
      "entropy": 5.5903397560119625,
      "epoch": 2.099054820415879,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004556817714975542,
      "loss": 5.3098,
      "mean_token_accuracy": 0.17457730770111085,
      "num_tokens": 46099212.0,
      "step": 24985
    },
    {
      "entropy": 5.662271118164062,
      "epoch": 2.0994749002310438,
      "grad_norm": 1.25,
      "learning_rate": 0.0004556639235964178,
      "loss": 5.2998,
      "mean_token_accuracy": 0.1736871376633644,
      "num_tokens": 46109129.0,
      "step": 24990
    },
    {
      "entropy": 5.623859024047851,
      "epoch": 2.0998949800462086,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004556460724949131,
      "loss": 5.256,
      "mean_token_accuracy": 0.18066179007291794,
      "num_tokens": 46118217.0,
      "step": 24995
    },
    {
      "entropy": 5.575393295288086,
      "epoch": 2.100315059861374,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045562821819335615,
      "loss": 5.2223,
      "mean_token_accuracy": 0.18051331788301467,
      "num_tokens": 46127472.0,
      "step": 25000
    },
    {
      "entropy": 5.5622560501098635,
      "epoch": 2.1007351396765386,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004556103606920634,
      "loss": 5.2561,
      "mean_token_accuracy": 0.17296084016561508,
      "num_tokens": 46136899.0,
      "step": 25005
    },
    {
      "entropy": 5.654111385345459,
      "epoch": 2.1011552194917034,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045559249999135105,
      "loss": 5.253,
      "mean_token_accuracy": 0.1824309468269348,
      "num_tokens": 46145916.0,
      "step": 25010
    },
    {
      "entropy": 5.50651593208313,
      "epoch": 2.101575299306868,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004555746360915356,
      "loss": 5.1276,
      "mean_token_accuracy": 0.18131841719150543,
      "num_tokens": 46155434.0,
      "step": 25015
    },
    {
      "entropy": 5.450176525115967,
      "epoch": 2.1019953791220334,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004555567689929335,
      "loss": 5.1847,
      "mean_token_accuracy": 0.18219584375619888,
      "num_tokens": 46163973.0,
      "step": 25020
    },
    {
      "entropy": 5.6124646186828615,
      "epoch": 2.102415458937198,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004555388986958611,
      "loss": 5.3772,
      "mean_token_accuracy": 0.17689626961946486,
      "num_tokens": 46174359.0,
      "step": 25025
    },
    {
      "entropy": 5.537656545639038,
      "epoch": 2.102835538752363,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045552102520063516,
      "loss": 5.0944,
      "mean_token_accuracy": 0.18438531309366227,
      "num_tokens": 46182913.0,
      "step": 25030
    },
    {
      "entropy": 5.551927947998047,
      "epoch": 2.1032556185675277,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004555031485075722,
      "loss": 5.2106,
      "mean_token_accuracy": 0.18115858882665634,
      "num_tokens": 46192131.0,
      "step": 25035
    },
    {
      "entropy": 5.566977071762085,
      "epoch": 2.1036756983826925,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004554852686169889,
      "loss": 5.2143,
      "mean_token_accuracy": 0.18304333388805388,
      "num_tokens": 46201988.0,
      "step": 25040
    },
    {
      "entropy": 5.429478883743286,
      "epoch": 2.1040957781978578,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004554673855292019,
      "loss": 5.1021,
      "mean_token_accuracy": 0.18505032807588578,
      "num_tokens": 46210677.0,
      "step": 25045
    },
    {
      "entropy": 5.614066505432129,
      "epoch": 2.1045158580130225,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045544949924452823,
      "loss": 5.2251,
      "mean_token_accuracy": 0.18117182999849318,
      "num_tokens": 46219489.0,
      "step": 25050
    },
    {
      "entropy": 5.582338619232178,
      "epoch": 2.1049359378281873,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004554316097632846,
      "loss": 5.2678,
      "mean_token_accuracy": 0.18271519541740416,
      "num_tokens": 46229494.0,
      "step": 25055
    },
    {
      "entropy": 5.475341749191284,
      "epoch": 2.105356017643352,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045541371708578783,
      "loss": 5.1036,
      "mean_token_accuracy": 0.19027747064828873,
      "num_tokens": 46238190.0,
      "step": 25060
    },
    {
      "entropy": 5.59563307762146,
      "epoch": 2.105776097458517,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045539582121235494,
      "loss": 5.2475,
      "mean_token_accuracy": 0.1791289046406746,
      "num_tokens": 46247492.0,
      "step": 25065
    },
    {
      "entropy": 5.659219741821289,
      "epoch": 2.106196177273682,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045537792214330304,
      "loss": 5.2945,
      "mean_token_accuracy": 0.18187092244625092,
      "num_tokens": 46257014.0,
      "step": 25070
    },
    {
      "entropy": 5.508939599990844,
      "epoch": 2.106616257088847,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045536001987894916,
      "loss": 5.2168,
      "mean_token_accuracy": 0.18076327443122864,
      "num_tokens": 46266450.0,
      "step": 25075
    },
    {
      "entropy": 5.585366582870483,
      "epoch": 2.1070363369040117,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004553421144196103,
      "loss": 5.3144,
      "mean_token_accuracy": 0.17510906904935836,
      "num_tokens": 46275118.0,
      "step": 25080
    },
    {
      "entropy": 5.614615964889526,
      "epoch": 2.1074564167191765,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045532420576560384,
      "loss": 5.2395,
      "mean_token_accuracy": 0.1840244397521019,
      "num_tokens": 46283617.0,
      "step": 25085
    },
    {
      "entropy": 5.573185396194458,
      "epoch": 2.1078764965343417,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004553062939172469,
      "loss": 5.2286,
      "mean_token_accuracy": 0.18324063569307328,
      "num_tokens": 46293456.0,
      "step": 25090
    },
    {
      "entropy": 5.536851024627685,
      "epoch": 2.1082965763495065,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004552883788748568,
      "loss": 5.2461,
      "mean_token_accuracy": 0.18436174392700194,
      "num_tokens": 46302579.0,
      "step": 25095
    },
    {
      "entropy": 5.5842077255249025,
      "epoch": 2.1087166561646713,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004552704606387509,
      "loss": 5.3352,
      "mean_token_accuracy": 0.17715413719415665,
      "num_tokens": 46311837.0,
      "step": 25100
    },
    {
      "entropy": 5.725308084487915,
      "epoch": 2.109136735979836,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045525253920924665,
      "loss": 5.3293,
      "mean_token_accuracy": 0.17480026930570602,
      "num_tokens": 46321697.0,
      "step": 25105
    },
    {
      "entropy": 5.534957504272461,
      "epoch": 2.109556815795001,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004552346145866614,
      "loss": 5.0485,
      "mean_token_accuracy": 0.198176771402359,
      "num_tokens": 46330626.0,
      "step": 25110
    },
    {
      "entropy": 5.641889524459839,
      "epoch": 2.109976895610166,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045521668677131277,
      "loss": 5.4047,
      "mean_token_accuracy": 0.17652053087949754,
      "num_tokens": 46340691.0,
      "step": 25115
    },
    {
      "entropy": 5.477499151229859,
      "epoch": 2.110396975425331,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004551987557635184,
      "loss": 5.1552,
      "mean_token_accuracy": 0.19363771080970765,
      "num_tokens": 46350153.0,
      "step": 25120
    },
    {
      "entropy": 5.500189542770386,
      "epoch": 2.1108170552404957,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004551808215635958,
      "loss": 5.2163,
      "mean_token_accuracy": 0.18013697266578674,
      "num_tokens": 46359399.0,
      "step": 25125
    },
    {
      "entropy": 5.588318967819214,
      "epoch": 2.1112371350556605,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045516288417186274,
      "loss": 5.2544,
      "mean_token_accuracy": 0.17766411900520324,
      "num_tokens": 46369026.0,
      "step": 25130
    },
    {
      "entropy": 5.551695823669434,
      "epoch": 2.1116572148708252,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045514494358863707,
      "loss": 5.1555,
      "mean_token_accuracy": 0.19007237404584884,
      "num_tokens": 46377534.0,
      "step": 25135
    },
    {
      "entropy": 5.628895902633667,
      "epoch": 2.1120772946859905,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004551269998142363,
      "loss": 5.2762,
      "mean_token_accuracy": 0.18295175731182098,
      "num_tokens": 46386142.0,
      "step": 25140
    },
    {
      "entropy": 5.65904016494751,
      "epoch": 2.1124973745011553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004551090528489786,
      "loss": 5.2293,
      "mean_token_accuracy": 0.1767893299460411,
      "num_tokens": 46395426.0,
      "step": 25145
    },
    {
      "entropy": 5.562749147415161,
      "epoch": 2.11291745431632,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045509110269318173,
      "loss": 5.2251,
      "mean_token_accuracy": 0.18627068251371384,
      "num_tokens": 46404184.0,
      "step": 25150
    },
    {
      "entropy": 5.573929500579834,
      "epoch": 2.113337534131485,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004550731493471637,
      "loss": 5.2533,
      "mean_token_accuracy": 0.1815846085548401,
      "num_tokens": 46412587.0,
      "step": 25155
    },
    {
      "entropy": 5.633985567092895,
      "epoch": 2.11375761394665,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045505519281124256,
      "loss": 5.3256,
      "mean_token_accuracy": 0.17555470913648605,
      "num_tokens": 46421851.0,
      "step": 25160
    },
    {
      "entropy": 5.614722490310669,
      "epoch": 2.114177693761815,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004550372330857364,
      "loss": 5.2641,
      "mean_token_accuracy": 0.17879040390253068,
      "num_tokens": 46430917.0,
      "step": 25165
    },
    {
      "entropy": 5.552657985687256,
      "epoch": 2.1145977735769796,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004550192701709634,
      "loss": 5.2119,
      "mean_token_accuracy": 0.1803215280175209,
      "num_tokens": 46439808.0,
      "step": 25170
    },
    {
      "entropy": 5.523211812973022,
      "epoch": 2.1150178533921444,
      "grad_norm": 1.25,
      "learning_rate": 0.00045500130406724167,
      "loss": 5.2069,
      "mean_token_accuracy": 0.17699734270572662,
      "num_tokens": 46450121.0,
      "step": 25175
    },
    {
      "entropy": 5.627210807800293,
      "epoch": 2.115437933207309,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045498333477488956,
      "loss": 5.3289,
      "mean_token_accuracy": 0.17111555188894273,
      "num_tokens": 46459407.0,
      "step": 25180
    },
    {
      "entropy": 5.626559638977051,
      "epoch": 2.1158580130224744,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004549653622942254,
      "loss": 5.2732,
      "mean_token_accuracy": 0.1795559361577034,
      "num_tokens": 46468830.0,
      "step": 25185
    },
    {
      "entropy": 5.677704906463623,
      "epoch": 2.1162780928376392,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045494738662556737,
      "loss": 5.3702,
      "mean_token_accuracy": 0.17664424926042557,
      "num_tokens": 46477649.0,
      "step": 25190
    },
    {
      "entropy": 5.550295448303222,
      "epoch": 2.116698172652804,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004549294077692342,
      "loss": 5.1803,
      "mean_token_accuracy": 0.18493889421224594,
      "num_tokens": 46486782.0,
      "step": 25195
    },
    {
      "entropy": 5.548875570297241,
      "epoch": 2.117118252467969,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045491142572554413,
      "loss": 5.2058,
      "mean_token_accuracy": 0.17966037392616271,
      "num_tokens": 46496902.0,
      "step": 25200
    },
    {
      "entropy": 5.6073465824127195,
      "epoch": 2.1175383322831336,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004548934404948158,
      "loss": 5.3205,
      "mean_token_accuracy": 0.17011562436819078,
      "num_tokens": 46506839.0,
      "step": 25205
    },
    {
      "entropy": 5.572475290298462,
      "epoch": 2.117958412098299,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004548754520773678,
      "loss": 5.1876,
      "mean_token_accuracy": 0.18221004605293273,
      "num_tokens": 46515338.0,
      "step": 25210
    },
    {
      "entropy": 5.6154787063598635,
      "epoch": 2.1183784919134636,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045485746047351877,
      "loss": 5.2183,
      "mean_token_accuracy": 0.17754581272602082,
      "num_tokens": 46524765.0,
      "step": 25215
    },
    {
      "entropy": 5.50904655456543,
      "epoch": 2.1187985717286284,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004548394656835875,
      "loss": 5.267,
      "mean_token_accuracy": 0.18178560435771943,
      "num_tokens": 46534470.0,
      "step": 25220
    },
    {
      "entropy": 5.546042346954346,
      "epoch": 2.119218651543793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045482146770789265,
      "loss": 5.1341,
      "mean_token_accuracy": 0.18850773125886916,
      "num_tokens": 46543597.0,
      "step": 25225
    },
    {
      "entropy": 5.494284343719483,
      "epoch": 2.1196387313589584,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004548034665467531,
      "loss": 5.1746,
      "mean_token_accuracy": 0.18332998752593993,
      "num_tokens": 46552745.0,
      "step": 25230
    },
    {
      "entropy": 5.558862018585205,
      "epoch": 2.120058811174123,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045478546220048777,
      "loss": 5.2087,
      "mean_token_accuracy": 0.18302785605192184,
      "num_tokens": 46561429.0,
      "step": 25235
    },
    {
      "entropy": 5.595675134658814,
      "epoch": 2.120478890989288,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004547674546694155,
      "loss": 5.2151,
      "mean_token_accuracy": 0.17740505188703537,
      "num_tokens": 46571044.0,
      "step": 25240
    },
    {
      "entropy": 5.578395175933838,
      "epoch": 2.1208989708044528,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004547494439538554,
      "loss": 5.3173,
      "mean_token_accuracy": 0.17900589406490325,
      "num_tokens": 46580165.0,
      "step": 25245
    },
    {
      "entropy": 5.613035202026367,
      "epoch": 2.1213190506196176,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004547314300541264,
      "loss": 5.223,
      "mean_token_accuracy": 0.18304192125797272,
      "num_tokens": 46588188.0,
      "step": 25250
    },
    {
      "entropy": 5.590068101882935,
      "epoch": 2.121739130434783,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004547134129705477,
      "loss": 5.2293,
      "mean_token_accuracy": 0.1776566043496132,
      "num_tokens": 46597067.0,
      "step": 25255
    },
    {
      "entropy": 5.479618930816651,
      "epoch": 2.1221592102499476,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004546953927034385,
      "loss": 5.1646,
      "mean_token_accuracy": 0.18879205733537674,
      "num_tokens": 46606021.0,
      "step": 25260
    },
    {
      "entropy": 5.618114280700683,
      "epoch": 2.1225792900651124,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045467736925311773,
      "loss": 5.2823,
      "mean_token_accuracy": 0.17487065494060516,
      "num_tokens": 46616300.0,
      "step": 25265
    },
    {
      "entropy": 5.494199800491333,
      "epoch": 2.122999369880277,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004546593426199051,
      "loss": 5.1681,
      "mean_token_accuracy": 0.18658257126808167,
      "num_tokens": 46625960.0,
      "step": 25270
    },
    {
      "entropy": 5.600203323364258,
      "epoch": 2.123419449695442,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004546413128041196,
      "loss": 5.3385,
      "mean_token_accuracy": 0.17630091011524202,
      "num_tokens": 46635310.0,
      "step": 25275
    },
    {
      "entropy": 5.62066159248352,
      "epoch": 2.123839529510607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045462327980608084,
      "loss": 5.274,
      "mean_token_accuracy": 0.1786182627081871,
      "num_tokens": 46644771.0,
      "step": 25280
    },
    {
      "entropy": 5.578542470932007,
      "epoch": 2.124259609325772,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045460524362610807,
      "loss": 5.2343,
      "mean_token_accuracy": 0.19069404900074005,
      "num_tokens": 46654315.0,
      "step": 25285
    },
    {
      "entropy": 5.602096891403198,
      "epoch": 2.1246796891409367,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004545872042645209,
      "loss": 5.2373,
      "mean_token_accuracy": 0.18174768835306168,
      "num_tokens": 46663310.0,
      "step": 25290
    },
    {
      "entropy": 5.5504354476928714,
      "epoch": 2.1250997689561015,
      "grad_norm": 1.53125,
      "learning_rate": 0.00045456916172163895,
      "loss": 5.1477,
      "mean_token_accuracy": 0.1825025662779808,
      "num_tokens": 46672996.0,
      "step": 25295
    },
    {
      "entropy": 5.552848148345947,
      "epoch": 2.1255198487712663,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004545511159977817,
      "loss": 5.3163,
      "mean_token_accuracy": 0.17230420112609862,
      "num_tokens": 46682635.0,
      "step": 25300
    },
    {
      "entropy": 5.649492406845093,
      "epoch": 2.1259399285864315,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045453306709326895,
      "loss": 5.3766,
      "mean_token_accuracy": 0.17168773859739303,
      "num_tokens": 46691611.0,
      "step": 25305
    },
    {
      "entropy": 5.5923120975494385,
      "epoch": 2.1263600084015963,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004545150150084203,
      "loss": 5.2064,
      "mean_token_accuracy": 0.17876987010240555,
      "num_tokens": 46700227.0,
      "step": 25310
    },
    {
      "entropy": 5.6015708446502686,
      "epoch": 2.126780088216761,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004544969597435556,
      "loss": 5.3115,
      "mean_token_accuracy": 0.17157361060380935,
      "num_tokens": 46709296.0,
      "step": 25315
    },
    {
      "entropy": 5.59907169342041,
      "epoch": 2.127200168031926,
      "grad_norm": 1.25,
      "learning_rate": 0.0004544789012989947,
      "loss": 5.1729,
      "mean_token_accuracy": 0.18459595143795013,
      "num_tokens": 46717918.0,
      "step": 25320
    },
    {
      "entropy": 5.51675386428833,
      "epoch": 2.127620247847091,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004544608396750575,
      "loss": 5.2147,
      "mean_token_accuracy": 0.181985704600811,
      "num_tokens": 46727654.0,
      "step": 25325
    },
    {
      "entropy": 5.530831909179687,
      "epoch": 2.128040327662256,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004544427748720639,
      "loss": 5.1979,
      "mean_token_accuracy": 0.17948512285947799,
      "num_tokens": 46736771.0,
      "step": 25330
    },
    {
      "entropy": 5.5988757610321045,
      "epoch": 2.1284604074774207,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045442470689033407,
      "loss": 5.3147,
      "mean_token_accuracy": 0.1799807593226433,
      "num_tokens": 46745815.0,
      "step": 25335
    },
    {
      "entropy": 5.570446729660034,
      "epoch": 2.1288804872925855,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004544066357301878,
      "loss": 5.3133,
      "mean_token_accuracy": 0.17561154663562775,
      "num_tokens": 46755819.0,
      "step": 25340
    },
    {
      "entropy": 5.563976001739502,
      "epoch": 2.1293005671077503,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004543885613919454,
      "loss": 5.304,
      "mean_token_accuracy": 0.17914692014455796,
      "num_tokens": 46765647.0,
      "step": 25345
    },
    {
      "entropy": 5.515087556838989,
      "epoch": 2.1297206469229155,
      "grad_norm": 1.296875,
      "learning_rate": 0.000454370483875927,
      "loss": 5.1238,
      "mean_token_accuracy": 0.1842019483447075,
      "num_tokens": 46774355.0,
      "step": 25350
    },
    {
      "entropy": 5.529027843475342,
      "epoch": 2.1301407267380803,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045435240318245285,
      "loss": 5.2077,
      "mean_token_accuracy": 0.18879560083150865,
      "num_tokens": 46782660.0,
      "step": 25355
    },
    {
      "entropy": 5.604471015930176,
      "epoch": 2.130560806553245,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004543343193118433,
      "loss": 5.3068,
      "mean_token_accuracy": 0.17594387233257294,
      "num_tokens": 46791637.0,
      "step": 25360
    },
    {
      "entropy": 5.603600597381591,
      "epoch": 2.13098088636841,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004543162322644185,
      "loss": 5.239,
      "mean_token_accuracy": 0.1825880602002144,
      "num_tokens": 46801022.0,
      "step": 25365
    },
    {
      "entropy": 5.576964616775513,
      "epoch": 2.131400966183575,
      "grad_norm": 1.21875,
      "learning_rate": 0.000454298142040499,
      "loss": 5.2853,
      "mean_token_accuracy": 0.17863944172859192,
      "num_tokens": 46810566.0,
      "step": 25370
    },
    {
      "entropy": 5.5883214473724365,
      "epoch": 2.13182104599874,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045428004864040534,
      "loss": 5.1841,
      "mean_token_accuracy": 0.18201329559087753,
      "num_tokens": 46819611.0,
      "step": 25375
    },
    {
      "entropy": 5.564843320846558,
      "epoch": 2.1322411258139047,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045426195206445784,
      "loss": 5.1799,
      "mean_token_accuracy": 0.18575197011232375,
      "num_tokens": 46829166.0,
      "step": 25380
    },
    {
      "entropy": 5.532839918136597,
      "epoch": 2.1326612056290695,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004542438523129772,
      "loss": 5.1948,
      "mean_token_accuracy": 0.18475055396556855,
      "num_tokens": 46837929.0,
      "step": 25385
    },
    {
      "entropy": 5.549889802932739,
      "epoch": 2.1330812854442343,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045422574938628403,
      "loss": 5.2454,
      "mean_token_accuracy": 0.180537086725235,
      "num_tokens": 46846823.0,
      "step": 25390
    },
    {
      "entropy": 5.575441646575928,
      "epoch": 2.1335013652593995,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000454207643284699,
      "loss": 5.2327,
      "mean_token_accuracy": 0.18402076214551927,
      "num_tokens": 46857117.0,
      "step": 25395
    },
    {
      "entropy": 5.54499077796936,
      "epoch": 2.1339214450745643,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045418953400854276,
      "loss": 5.1642,
      "mean_token_accuracy": 0.18940766602754594,
      "num_tokens": 46865698.0,
      "step": 25400
    },
    {
      "entropy": 5.633910512924194,
      "epoch": 2.134341524889729,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045417142155813623,
      "loss": 5.3369,
      "mean_token_accuracy": 0.17188064455986024,
      "num_tokens": 46874294.0,
      "step": 25405
    },
    {
      "entropy": 5.507958984375,
      "epoch": 2.134761604704894,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004541533059338003,
      "loss": 5.1913,
      "mean_token_accuracy": 0.1830242916941643,
      "num_tokens": 46883267.0,
      "step": 25410
    },
    {
      "entropy": 5.58703236579895,
      "epoch": 2.1351816845200586,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045413518713585574,
      "loss": 5.2734,
      "mean_token_accuracy": 0.1777835562825203,
      "num_tokens": 46892799.0,
      "step": 25415
    },
    {
      "entropy": 5.576012992858887,
      "epoch": 2.135601764335224,
      "grad_norm": 1.375,
      "learning_rate": 0.00045411706516462356,
      "loss": 5.2157,
      "mean_token_accuracy": 0.18584727942943574,
      "num_tokens": 46902817.0,
      "step": 25420
    },
    {
      "entropy": 5.645403480529785,
      "epoch": 2.1360218441503886,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045409894002042484,
      "loss": 5.3355,
      "mean_token_accuracy": 0.1761436179280281,
      "num_tokens": 46911879.0,
      "step": 25425
    },
    {
      "entropy": 5.604062938690186,
      "epoch": 2.1364419239655534,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004540808117035807,
      "loss": 5.2901,
      "mean_token_accuracy": 0.17402357161045073,
      "num_tokens": 46921728.0,
      "step": 25430
    },
    {
      "entropy": 5.539064598083496,
      "epoch": 2.1368620037807182,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004540626802144121,
      "loss": 5.1465,
      "mean_token_accuracy": 0.18887661546468734,
      "num_tokens": 46930882.0,
      "step": 25435
    },
    {
      "entropy": 5.565151929855347,
      "epoch": 2.137282083595883,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004540445455532403,
      "loss": 5.261,
      "mean_token_accuracy": 0.18579446524381638,
      "num_tokens": 46940116.0,
      "step": 25440
    },
    {
      "entropy": 5.60378098487854,
      "epoch": 2.1377021634110482,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004540264077203866,
      "loss": 5.2863,
      "mean_token_accuracy": 0.17504278272390367,
      "num_tokens": 46949906.0,
      "step": 25445
    },
    {
      "entropy": 5.485409736633301,
      "epoch": 2.138122243226213,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004540082667161723,
      "loss": 5.1436,
      "mean_token_accuracy": 0.18526019901037216,
      "num_tokens": 46959176.0,
      "step": 25450
    },
    {
      "entropy": 5.528769254684448,
      "epoch": 2.138542323041378,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004539901225409187,
      "loss": 5.1684,
      "mean_token_accuracy": 0.1913315623998642,
      "num_tokens": 46967912.0,
      "step": 25455
    },
    {
      "entropy": 5.493267774581909,
      "epoch": 2.1389624028565426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045397197519494726,
      "loss": 5.1406,
      "mean_token_accuracy": 0.18913838714361192,
      "num_tokens": 46976436.0,
      "step": 25460
    },
    {
      "entropy": 5.520484256744385,
      "epoch": 2.139382482671708,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045395382467857944,
      "loss": 5.2257,
      "mean_token_accuracy": 0.18157807290554046,
      "num_tokens": 46986249.0,
      "step": 25465
    },
    {
      "entropy": 5.598328590393066,
      "epoch": 2.1398025624868726,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045393567099213673,
      "loss": 5.1808,
      "mean_token_accuracy": 0.19307401925325393,
      "num_tokens": 46995234.0,
      "step": 25470
    },
    {
      "entropy": 5.577827310562133,
      "epoch": 2.1402226423020374,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045391751413594077,
      "loss": 5.3269,
      "mean_token_accuracy": 0.17916010171175004,
      "num_tokens": 47004646.0,
      "step": 25475
    },
    {
      "entropy": 5.501791334152221,
      "epoch": 2.140642722117202,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045389935411031324,
      "loss": 5.1622,
      "mean_token_accuracy": 0.18877279609441758,
      "num_tokens": 47013803.0,
      "step": 25480
    },
    {
      "entropy": 5.560681438446045,
      "epoch": 2.141062801932367,
      "grad_norm": 1.125,
      "learning_rate": 0.00045388119091557567,
      "loss": 5.2503,
      "mean_token_accuracy": 0.1818718820810318,
      "num_tokens": 47022691.0,
      "step": 25485
    },
    {
      "entropy": 5.551713228225708,
      "epoch": 2.141482881747532,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045386302455205,
      "loss": 5.2405,
      "mean_token_accuracy": 0.18485425263643265,
      "num_tokens": 47032262.0,
      "step": 25490
    },
    {
      "entropy": 5.551495695114136,
      "epoch": 2.141902961562697,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045384485502005786,
      "loss": 5.236,
      "mean_token_accuracy": 0.1773254930973053,
      "num_tokens": 47041533.0,
      "step": 25495
    },
    {
      "entropy": 5.618907880783081,
      "epoch": 2.142323041377862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045382668231992127,
      "loss": 5.2651,
      "mean_token_accuracy": 0.18468783497810365,
      "num_tokens": 47050227.0,
      "step": 25500
    },
    {
      "entropy": 5.520976924896241,
      "epoch": 2.1427431211930266,
      "grad_norm": 1.28125,
      "learning_rate": 0.000453808506451962,
      "loss": 5.1349,
      "mean_token_accuracy": 0.18293166309595107,
      "num_tokens": 47059100.0,
      "step": 25505
    },
    {
      "entropy": 5.479643392562866,
      "epoch": 2.1431632010081914,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004537903274165022,
      "loss": 5.1519,
      "mean_token_accuracy": 0.19040325433015823,
      "num_tokens": 47068861.0,
      "step": 25510
    },
    {
      "entropy": 5.506395626068115,
      "epoch": 2.1435832808233566,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004537721452138638,
      "loss": 5.1645,
      "mean_token_accuracy": 0.18873895555734635,
      "num_tokens": 47077453.0,
      "step": 25515
    },
    {
      "entropy": 5.553389596939087,
      "epoch": 2.1440033606385214,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004537539598443689,
      "loss": 5.1706,
      "mean_token_accuracy": 0.18462664484977723,
      "num_tokens": 47086944.0,
      "step": 25520
    },
    {
      "entropy": 5.591655826568603,
      "epoch": 2.144423440453686,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004537357713083396,
      "loss": 5.2151,
      "mean_token_accuracy": 0.17335921078920363,
      "num_tokens": 47097308.0,
      "step": 25525
    },
    {
      "entropy": 5.5366082191467285,
      "epoch": 2.144843520268851,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045371757960609816,
      "loss": 5.2339,
      "mean_token_accuracy": 0.18321507275104523,
      "num_tokens": 47106895.0,
      "step": 25530
    },
    {
      "entropy": 5.519483470916748,
      "epoch": 2.145263600084016,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004536993847379669,
      "loss": 5.1654,
      "mean_token_accuracy": 0.19175878614187242,
      "num_tokens": 47115452.0,
      "step": 25535
    },
    {
      "entropy": 5.5441851139068605,
      "epoch": 2.145683679899181,
      "grad_norm": 1.25,
      "learning_rate": 0.0004536811867042679,
      "loss": 5.3181,
      "mean_token_accuracy": 0.17465200126171113,
      "num_tokens": 47124361.0,
      "step": 25540
    },
    {
      "entropy": 5.597061061859131,
      "epoch": 2.1461037597143457,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045366298550532385,
      "loss": 5.2026,
      "mean_token_accuracy": 0.1918330356478691,
      "num_tokens": 47133256.0,
      "step": 25545
    },
    {
      "entropy": 5.544729423522949,
      "epoch": 2.1465238395295105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000453644781141457,
      "loss": 5.2246,
      "mean_token_accuracy": 0.18513346016407012,
      "num_tokens": 47141716.0,
      "step": 25550
    },
    {
      "entropy": 5.595617771148682,
      "epoch": 2.1469439193446753,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045362657361298973,
      "loss": 5.2129,
      "mean_token_accuracy": 0.17364211082458497,
      "num_tokens": 47150894.0,
      "step": 25555
    },
    {
      "entropy": 5.523682403564453,
      "epoch": 2.1473639991598406,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004536083629202448,
      "loss": 5.1683,
      "mean_token_accuracy": 0.18459151983261107,
      "num_tokens": 47160097.0,
      "step": 25560
    },
    {
      "entropy": 5.505368757247925,
      "epoch": 2.1477840789750053,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004535901490635446,
      "loss": 5.248,
      "mean_token_accuracy": 0.17930907160043716,
      "num_tokens": 47169779.0,
      "step": 25565
    },
    {
      "entropy": 5.526602649688721,
      "epoch": 2.14820415879017,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004535719320432119,
      "loss": 5.1843,
      "mean_token_accuracy": 0.18865631967782975,
      "num_tokens": 47179486.0,
      "step": 25570
    },
    {
      "entropy": 5.632379865646362,
      "epoch": 2.148624238605335,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004535537118595694,
      "loss": 5.2375,
      "mean_token_accuracy": 0.1828949362039566,
      "num_tokens": 47188241.0,
      "step": 25575
    },
    {
      "entropy": 5.5763341903686525,
      "epoch": 2.1490443184204997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045353548851293976,
      "loss": 5.2612,
      "mean_token_accuracy": 0.18426885604858398,
      "num_tokens": 47197320.0,
      "step": 25580
    },
    {
      "entropy": 5.548977661132812,
      "epoch": 2.149464398235665,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004535172620036459,
      "loss": 5.2218,
      "mean_token_accuracy": 0.1816057413816452,
      "num_tokens": 47206514.0,
      "step": 25585
    },
    {
      "entropy": 5.587989330291748,
      "epoch": 2.1498844780508297,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045349903233201066,
      "loss": 5.2146,
      "mean_token_accuracy": 0.1866088330745697,
      "num_tokens": 47215253.0,
      "step": 25590
    },
    {
      "entropy": 5.4367570877075195,
      "epoch": 2.1503045578659945,
      "grad_norm": 1.125,
      "learning_rate": 0.0004534807994983571,
      "loss": 5.1362,
      "mean_token_accuracy": 0.18901285380125046,
      "num_tokens": 47224360.0,
      "step": 25595
    },
    {
      "entropy": 5.563683795928955,
      "epoch": 2.1507246376811593,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045346256350300794,
      "loss": 5.2382,
      "mean_token_accuracy": 0.1844244807958603,
      "num_tokens": 47234289.0,
      "step": 25600
    },
    {
      "entropy": 5.605845928192139,
      "epoch": 2.151144717496324,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004534443243462865,
      "loss": 5.2141,
      "mean_token_accuracy": 0.1865975245833397,
      "num_tokens": 47244393.0,
      "step": 25605
    },
    {
      "entropy": 5.593759059906006,
      "epoch": 2.1515647973114893,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004534260820285156,
      "loss": 5.3356,
      "mean_token_accuracy": 0.1789148986339569,
      "num_tokens": 47253947.0,
      "step": 25610
    },
    {
      "entropy": 5.524381685256958,
      "epoch": 2.151984877126654,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004534078365500186,
      "loss": 5.2669,
      "mean_token_accuracy": 0.18267868757247924,
      "num_tokens": 47263361.0,
      "step": 25615
    },
    {
      "entropy": 5.719612503051758,
      "epoch": 2.152404956941819,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004533895879111186,
      "loss": 5.4697,
      "mean_token_accuracy": 0.17128380984067917,
      "num_tokens": 47274460.0,
      "step": 25620
    },
    {
      "entropy": 5.697061347961426,
      "epoch": 2.1528250367569837,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045337133611213893,
      "loss": 5.3112,
      "mean_token_accuracy": 0.17914173603057862,
      "num_tokens": 47283651.0,
      "step": 25625
    },
    {
      "entropy": 5.639389705657959,
      "epoch": 2.153245116572149,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004533530811534028,
      "loss": 5.1804,
      "mean_token_accuracy": 0.18341459035873414,
      "num_tokens": 47293220.0,
      "step": 25630
    },
    {
      "entropy": 5.459346055984497,
      "epoch": 2.1536651963873137,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045333482303523383,
      "loss": 5.0601,
      "mean_token_accuracy": 0.1888313978910446,
      "num_tokens": 47302468.0,
      "step": 25635
    },
    {
      "entropy": 5.518938064575195,
      "epoch": 2.1540852762024785,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045331656175795525,
      "loss": 5.1816,
      "mean_token_accuracy": 0.1867467999458313,
      "num_tokens": 47311241.0,
      "step": 25640
    },
    {
      "entropy": 5.6093244552612305,
      "epoch": 2.1545053560176433,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004532982973218906,
      "loss": 5.244,
      "mean_token_accuracy": 0.17565635442733765,
      "num_tokens": 47321526.0,
      "step": 25645
    },
    {
      "entropy": 5.6693384647369385,
      "epoch": 2.154925435832808,
      "grad_norm": 1.75,
      "learning_rate": 0.0004532800297273634,
      "loss": 5.2768,
      "mean_token_accuracy": 0.17493624985218048,
      "num_tokens": 47331449.0,
      "step": 25650
    },
    {
      "entropy": 5.568662786483765,
      "epoch": 2.1553455156479733,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004532617589746972,
      "loss": 5.2309,
      "mean_token_accuracy": 0.18395858258008957,
      "num_tokens": 47340836.0,
      "step": 25655
    },
    {
      "entropy": 5.558906936645508,
      "epoch": 2.155765595463138,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004532434850642159,
      "loss": 5.2293,
      "mean_token_accuracy": 0.18105486333370208,
      "num_tokens": 47350735.0,
      "step": 25660
    },
    {
      "entropy": 5.548009061813355,
      "epoch": 2.156185675278303,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045322520799624287,
      "loss": 5.2344,
      "mean_token_accuracy": 0.18420572876930236,
      "num_tokens": 47360089.0,
      "step": 25665
    },
    {
      "entropy": 5.531437349319458,
      "epoch": 2.1566057550934676,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004532069277711022,
      "loss": 5.1377,
      "mean_token_accuracy": 0.18216256201267242,
      "num_tokens": 47369570.0,
      "step": 25670
    },
    {
      "entropy": 5.566018867492676,
      "epoch": 2.157025834908633,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004531886443891175,
      "loss": 5.1862,
      "mean_token_accuracy": 0.17375797033309937,
      "num_tokens": 47379504.0,
      "step": 25675
    },
    {
      "entropy": 5.595839262008667,
      "epoch": 2.1574459147237977,
      "grad_norm": 2.125,
      "learning_rate": 0.0004531703578506127,
      "loss": 5.2664,
      "mean_token_accuracy": 0.17863921374082564,
      "num_tokens": 47388968.0,
      "step": 25680
    },
    {
      "entropy": 5.517437553405761,
      "epoch": 2.1578659945389624,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045315206815591176,
      "loss": 5.1795,
      "mean_token_accuracy": 0.18288652896881102,
      "num_tokens": 47398124.0,
      "step": 25685
    },
    {
      "entropy": 5.491002893447876,
      "epoch": 2.1582860743541272,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045313377530533874,
      "loss": 5.1804,
      "mean_token_accuracy": 0.18371145725250243,
      "num_tokens": 47407082.0,
      "step": 25690
    },
    {
      "entropy": 5.532442617416382,
      "epoch": 2.158706154169292,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045311547929921753,
      "loss": 5.1879,
      "mean_token_accuracy": 0.1853356257081032,
      "num_tokens": 47416369.0,
      "step": 25695
    },
    {
      "entropy": 5.461814498901367,
      "epoch": 2.1591262339844572,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004530971801378723,
      "loss": 5.1472,
      "mean_token_accuracy": 0.18261002004146576,
      "num_tokens": 47424947.0,
      "step": 25700
    },
    {
      "entropy": 5.51474962234497,
      "epoch": 2.159546313799622,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004530788778216274,
      "loss": 5.1901,
      "mean_token_accuracy": 0.18545023053884507,
      "num_tokens": 47433659.0,
      "step": 25705
    },
    {
      "entropy": 5.599772262573242,
      "epoch": 2.159966393614787,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004530605723508067,
      "loss": 5.2866,
      "mean_token_accuracy": 0.17416015714406968,
      "num_tokens": 47444706.0,
      "step": 25710
    },
    {
      "entropy": 5.6379588603973385,
      "epoch": 2.1603864734299516,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004530422637257347,
      "loss": 5.3456,
      "mean_token_accuracy": 0.17926814556121826,
      "num_tokens": 47454200.0,
      "step": 25715
    },
    {
      "entropy": 5.709537601470947,
      "epoch": 2.1608065532451164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004530239519467356,
      "loss": 5.277,
      "mean_token_accuracy": 0.18328871876001357,
      "num_tokens": 47463660.0,
      "step": 25720
    },
    {
      "entropy": 5.536508798599243,
      "epoch": 2.1612266330602816,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045300563701413407,
      "loss": 5.1616,
      "mean_token_accuracy": 0.18135672807693481,
      "num_tokens": 47472014.0,
      "step": 25725
    },
    {
      "entropy": 5.52828049659729,
      "epoch": 2.1616467128754464,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045298731892825416,
      "loss": 5.2341,
      "mean_token_accuracy": 0.18090033233165742,
      "num_tokens": 47481997.0,
      "step": 25730
    },
    {
      "entropy": 5.59212212562561,
      "epoch": 2.162066792690611,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045296899768942054,
      "loss": 5.1998,
      "mean_token_accuracy": 0.17792314738035203,
      "num_tokens": 47490929.0,
      "step": 25735
    },
    {
      "entropy": 5.531614351272583,
      "epoch": 2.162486872505776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045295067329795784,
      "loss": 5.2404,
      "mean_token_accuracy": 0.1819904029369354,
      "num_tokens": 47500238.0,
      "step": 25740
    },
    {
      "entropy": 5.621511173248291,
      "epoch": 2.1629069523209408,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045293234575419054,
      "loss": 5.362,
      "mean_token_accuracy": 0.17397900819778442,
      "num_tokens": 47510017.0,
      "step": 25745
    },
    {
      "entropy": 5.53157639503479,
      "epoch": 2.163327032136106,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004529140150584435,
      "loss": 5.2063,
      "mean_token_accuracy": 0.18551530092954635,
      "num_tokens": 47519552.0,
      "step": 25750
    },
    {
      "entropy": 5.50888466835022,
      "epoch": 2.163747111951271,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004528956812110411,
      "loss": 5.1236,
      "mean_token_accuracy": 0.1888862133026123,
      "num_tokens": 47528045.0,
      "step": 25755
    },
    {
      "entropy": 5.503701019287109,
      "epoch": 2.1641671917664356,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004528773442123084,
      "loss": 5.1775,
      "mean_token_accuracy": 0.18165435492992402,
      "num_tokens": 47537171.0,
      "step": 25760
    },
    {
      "entropy": 5.608512735366821,
      "epoch": 2.1645872715816004,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004528590040625701,
      "loss": 5.3371,
      "mean_token_accuracy": 0.17623130977153778,
      "num_tokens": 47547283.0,
      "step": 25765
    },
    {
      "entropy": 5.589131927490234,
      "epoch": 2.1650073513967656,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004528406607621512,
      "loss": 5.2273,
      "mean_token_accuracy": 0.18407107889652252,
      "num_tokens": 47556042.0,
      "step": 25770
    },
    {
      "entropy": 5.533044862747192,
      "epoch": 2.1654274312119304,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045282231431137654,
      "loss": 5.2449,
      "mean_token_accuracy": 0.18520289659500122,
      "num_tokens": 47564167.0,
      "step": 25775
    },
    {
      "entropy": 5.5821513652801515,
      "epoch": 2.165847511027095,
      "grad_norm": 1.375,
      "learning_rate": 0.0004528039647105712,
      "loss": 5.2259,
      "mean_token_accuracy": 0.18578939735889435,
      "num_tokens": 47573049.0,
      "step": 25780
    },
    {
      "entropy": 5.560171604156494,
      "epoch": 2.16626759084226,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004527856119600601,
      "loss": 5.2128,
      "mean_token_accuracy": 0.17929146885871888,
      "num_tokens": 47581977.0,
      "step": 25785
    },
    {
      "entropy": 5.5724128723144535,
      "epoch": 2.1666876706574247,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045276725606016837,
      "loss": 5.3105,
      "mean_token_accuracy": 0.17640087902545928,
      "num_tokens": 47591302.0,
      "step": 25790
    },
    {
      "entropy": 5.642348146438598,
      "epoch": 2.16710775047259,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004527488970112214,
      "loss": 5.2908,
      "mean_token_accuracy": 0.17374746203422547,
      "num_tokens": 47600935.0,
      "step": 25795
    },
    {
      "entropy": 5.571298027038575,
      "epoch": 2.1675278302877548,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045273053481354413,
      "loss": 5.1772,
      "mean_token_accuracy": 0.18073158860206603,
      "num_tokens": 47609918.0,
      "step": 25800
    },
    {
      "entropy": 5.523872327804566,
      "epoch": 2.1679479101029195,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045271216946746195,
      "loss": 5.2483,
      "mean_token_accuracy": 0.1824260652065277,
      "num_tokens": 47618407.0,
      "step": 25805
    },
    {
      "entropy": 5.575429391860962,
      "epoch": 2.1683679899180843,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045269380097330027,
      "loss": 5.2499,
      "mean_token_accuracy": 0.1761736288666725,
      "num_tokens": 47627900.0,
      "step": 25810
    },
    {
      "entropy": 5.595975542068482,
      "epoch": 2.168788069733249,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004526754293313844,
      "loss": 5.3154,
      "mean_token_accuracy": 0.17856395989656448,
      "num_tokens": 47636740.0,
      "step": 25815
    },
    {
      "entropy": 5.507689762115478,
      "epoch": 2.1692081495484143,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004526570545420397,
      "loss": 5.0982,
      "mean_token_accuracy": 0.1947345942258835,
      "num_tokens": 47646135.0,
      "step": 25820
    },
    {
      "entropy": 5.6392443656921385,
      "epoch": 2.169628229363579,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004526386766055918,
      "loss": 5.3105,
      "mean_token_accuracy": 0.18027545064687728,
      "num_tokens": 47654203.0,
      "step": 25825
    },
    {
      "entropy": 5.554869222640991,
      "epoch": 2.170048309178744,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004526202955223662,
      "loss": 5.2521,
      "mean_token_accuracy": 0.18327800631523133,
      "num_tokens": 47663232.0,
      "step": 25830
    },
    {
      "entropy": 5.614356517791748,
      "epoch": 2.1704683889939087,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045260191129268857,
      "loss": 5.2578,
      "mean_token_accuracy": 0.18316095024347306,
      "num_tokens": 47672121.0,
      "step": 25835
    },
    {
      "entropy": 5.559417390823365,
      "epoch": 2.170888468809074,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045258352391688454,
      "loss": 5.2305,
      "mean_token_accuracy": 0.18086334615945815,
      "num_tokens": 47681512.0,
      "step": 25840
    },
    {
      "entropy": 5.537013912200928,
      "epoch": 2.1713085486242387,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045256513339527984,
      "loss": 5.1409,
      "mean_token_accuracy": 0.1924825817346573,
      "num_tokens": 47690340.0,
      "step": 25845
    },
    {
      "entropy": 5.558413982391357,
      "epoch": 2.1717286284394035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045254673972820015,
      "loss": 5.1556,
      "mean_token_accuracy": 0.19420274049043657,
      "num_tokens": 47698917.0,
      "step": 25850
    },
    {
      "entropy": 5.542934799194336,
      "epoch": 2.1721487082545683,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045252834291597144,
      "loss": 5.2374,
      "mean_token_accuracy": 0.17875467985868454,
      "num_tokens": 47708832.0,
      "step": 25855
    },
    {
      "entropy": 5.50212025642395,
      "epoch": 2.172568788069733,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004525099429589195,
      "loss": 5.2018,
      "mean_token_accuracy": 0.18479922711849212,
      "num_tokens": 47717820.0,
      "step": 25860
    },
    {
      "entropy": 5.519931411743164,
      "epoch": 2.1729888678848983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004524915398573704,
      "loss": 5.1573,
      "mean_token_accuracy": 0.1895698755979538,
      "num_tokens": 47726959.0,
      "step": 25865
    },
    {
      "entropy": 5.552200984954834,
      "epoch": 2.173408947700063,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045247313361165007,
      "loss": 5.2015,
      "mean_token_accuracy": 0.18706825673580169,
      "num_tokens": 47736100.0,
      "step": 25870
    },
    {
      "entropy": 5.520269393920898,
      "epoch": 2.173829027515228,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004524547242220845,
      "loss": 5.2103,
      "mean_token_accuracy": 0.18202619403600692,
      "num_tokens": 47745356.0,
      "step": 25875
    },
    {
      "entropy": 5.5808587074279785,
      "epoch": 2.1742491073303927,
      "grad_norm": 1.25,
      "learning_rate": 0.0004524363116889998,
      "loss": 5.3036,
      "mean_token_accuracy": 0.1755009725689888,
      "num_tokens": 47755109.0,
      "step": 25880
    },
    {
      "entropy": 5.688870000839233,
      "epoch": 2.1746691871455575,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004524178960127224,
      "loss": 5.335,
      "mean_token_accuracy": 0.17612747102975845,
      "num_tokens": 47764463.0,
      "step": 25885
    },
    {
      "entropy": 5.5994010925292965,
      "epoch": 2.1750892669607227,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004523994771935782,
      "loss": 5.261,
      "mean_token_accuracy": 0.1823442131280899,
      "num_tokens": 47773860.0,
      "step": 25890
    },
    {
      "entropy": 5.561999797821045,
      "epoch": 2.1755093467758875,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004523810552318937,
      "loss": 5.2479,
      "mean_token_accuracy": 0.17895616888999938,
      "num_tokens": 47782850.0,
      "step": 25895
    },
    {
      "entropy": 5.551925134658814,
      "epoch": 2.1759294265910523,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004523626301279951,
      "loss": 5.2023,
      "mean_token_accuracy": 0.18073381185531617,
      "num_tokens": 47792074.0,
      "step": 25900
    },
    {
      "entropy": 5.488715124130249,
      "epoch": 2.176349506406217,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045234420188220874,
      "loss": 5.0179,
      "mean_token_accuracy": 0.19902902692556382,
      "num_tokens": 47800419.0,
      "step": 25905
    },
    {
      "entropy": 5.409627199172974,
      "epoch": 2.176769586221382,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004523257704948612,
      "loss": 5.1122,
      "mean_token_accuracy": 0.18975480347871782,
      "num_tokens": 47809124.0,
      "step": 25910
    },
    {
      "entropy": 5.452526664733886,
      "epoch": 2.177189666036547,
      "grad_norm": 1.4375,
      "learning_rate": 0.000452307335966279,
      "loss": 5.1623,
      "mean_token_accuracy": 0.18304960131645204,
      "num_tokens": 47818242.0,
      "step": 25915
    },
    {
      "entropy": 5.650811862945557,
      "epoch": 2.177609745851712,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045228889829678864,
      "loss": 5.2061,
      "mean_token_accuracy": 0.17597699016332627,
      "num_tokens": 47826905.0,
      "step": 25920
    },
    {
      "entropy": 5.477949619293213,
      "epoch": 2.1780298256668766,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045227045748671675,
      "loss": 5.188,
      "mean_token_accuracy": 0.19186608046293258,
      "num_tokens": 47835584.0,
      "step": 25925
    },
    {
      "entropy": 5.446269989013672,
      "epoch": 2.1784499054820414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045225201353638994,
      "loss": 5.168,
      "mean_token_accuracy": 0.19047574549913407,
      "num_tokens": 47844230.0,
      "step": 25930
    },
    {
      "entropy": 5.550783777236939,
      "epoch": 2.1788699852972067,
      "grad_norm": 1.265625,
      "learning_rate": 0.000452233566446135,
      "loss": 5.0979,
      "mean_token_accuracy": 0.1873260870575905,
      "num_tokens": 47852560.0,
      "step": 25935
    },
    {
      "entropy": 5.595924758911133,
      "epoch": 2.1792900651123714,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004522151162162787,
      "loss": 5.2009,
      "mean_token_accuracy": 0.18841852843761445,
      "num_tokens": 47861003.0,
      "step": 25940
    },
    {
      "entropy": 5.547629928588867,
      "epoch": 2.1797101449275362,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004521966628471478,
      "loss": 5.2222,
      "mean_token_accuracy": 0.1770896315574646,
      "num_tokens": 47869959.0,
      "step": 25945
    },
    {
      "entropy": 5.543789672851562,
      "epoch": 2.180130224742701,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004521782063390694,
      "loss": 5.2371,
      "mean_token_accuracy": 0.1805410712957382,
      "num_tokens": 47879330.0,
      "step": 25950
    },
    {
      "entropy": 5.535733890533447,
      "epoch": 2.180550304557866,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004521597466923702,
      "loss": 5.1904,
      "mean_token_accuracy": 0.1808372288942337,
      "num_tokens": 47888046.0,
      "step": 25955
    },
    {
      "entropy": 5.601400327682495,
      "epoch": 2.180970384373031,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045214128390737735,
      "loss": 5.2564,
      "mean_token_accuracy": 0.17988570183515548,
      "num_tokens": 47898107.0,
      "step": 25960
    },
    {
      "entropy": 5.572787237167359,
      "epoch": 2.181390464188196,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004521228179844178,
      "loss": 5.2397,
      "mean_token_accuracy": 0.1808492973446846,
      "num_tokens": 47907417.0,
      "step": 25965
    },
    {
      "entropy": 5.609670162200928,
      "epoch": 2.1818105440033606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004521043489238188,
      "loss": 5.3527,
      "mean_token_accuracy": 0.1714984133839607,
      "num_tokens": 47915727.0,
      "step": 25970
    },
    {
      "entropy": 5.594399118423462,
      "epoch": 2.1822306238185254,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045208587672590744,
      "loss": 5.2095,
      "mean_token_accuracy": 0.18584977239370346,
      "num_tokens": 47924718.0,
      "step": 25975
    },
    {
      "entropy": 5.5011961460113525,
      "epoch": 2.1826507036336906,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000452067401391011,
      "loss": 5.159,
      "mean_token_accuracy": 0.19530830681324005,
      "num_tokens": 47933869.0,
      "step": 25980
    },
    {
      "entropy": 5.482358312606811,
      "epoch": 2.1830707834488554,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045204892291945666,
      "loss": 5.1971,
      "mean_token_accuracy": 0.1872905120253563,
      "num_tokens": 47942723.0,
      "step": 25985
    },
    {
      "entropy": 5.498285293579102,
      "epoch": 2.18349086326402,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004520304413115719,
      "loss": 5.1235,
      "mean_token_accuracy": 0.18266915082931517,
      "num_tokens": 47950862.0,
      "step": 25990
    },
    {
      "entropy": 5.581514453887939,
      "epoch": 2.183910943079185,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045201195656768394,
      "loss": 5.2918,
      "mean_token_accuracy": 0.17694679647684097,
      "num_tokens": 47960207.0,
      "step": 25995
    },
    {
      "entropy": 5.608252429962159,
      "epoch": 2.1843310228943498,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045199346868812036,
      "loss": 5.3219,
      "mean_token_accuracy": 0.17655700147151948,
      "num_tokens": 47970022.0,
      "step": 26000
    },
    {
      "entropy": 5.539549541473389,
      "epoch": 2.184751102709515,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004519749776732086,
      "loss": 5.1087,
      "mean_token_accuracy": 0.18980390429496766,
      "num_tokens": 47978524.0,
      "step": 26005
    },
    {
      "entropy": 5.506338500976563,
      "epoch": 2.18517118252468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004519564835232762,
      "loss": 5.1307,
      "mean_token_accuracy": 0.18752649873495103,
      "num_tokens": 47988004.0,
      "step": 26010
    },
    {
      "entropy": 5.504634284973145,
      "epoch": 2.1855912623398446,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045193798623865093,
      "loss": 5.2204,
      "mean_token_accuracy": 0.1832644075155258,
      "num_tokens": 47996706.0,
      "step": 26015
    },
    {
      "entropy": 5.560032749176026,
      "epoch": 2.1860113421550094,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004519194858196602,
      "loss": 5.1903,
      "mean_token_accuracy": 0.18179081082344056,
      "num_tokens": 48005305.0,
      "step": 26020
    },
    {
      "entropy": 5.520339584350586,
      "epoch": 2.186431421970174,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045190098226663194,
      "loss": 5.1401,
      "mean_token_accuracy": 0.1881472572684288,
      "num_tokens": 48014349.0,
      "step": 26025
    },
    {
      "entropy": 5.488553857803344,
      "epoch": 2.1868515017853394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004518824755798939,
      "loss": 5.2412,
      "mean_token_accuracy": 0.1822286382317543,
      "num_tokens": 48023630.0,
      "step": 26030
    },
    {
      "entropy": 5.497898435592651,
      "epoch": 2.187271581600504,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004518639657597739,
      "loss": 5.1264,
      "mean_token_accuracy": 0.18377941250801086,
      "num_tokens": 48032869.0,
      "step": 26035
    },
    {
      "entropy": 5.640784072875976,
      "epoch": 2.187691661415669,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045184545280659976,
      "loss": 5.2907,
      "mean_token_accuracy": 0.1814137727022171,
      "num_tokens": 48041834.0,
      "step": 26040
    },
    {
      "entropy": 5.516853189468383,
      "epoch": 2.1881117412308337,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004518269367206996,
      "loss": 5.1216,
      "mean_token_accuracy": 0.18992259502410888,
      "num_tokens": 48051114.0,
      "step": 26045
    },
    {
      "entropy": 5.538951683044433,
      "epoch": 2.1885318210459985,
      "grad_norm": 1.125,
      "learning_rate": 0.0004518084175024012,
      "loss": 5.2729,
      "mean_token_accuracy": 0.18366690278053283,
      "num_tokens": 48059788.0,
      "step": 26050
    },
    {
      "entropy": 5.5674560546875,
      "epoch": 2.1889519008611638,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004517898951520327,
      "loss": 5.1737,
      "mean_token_accuracy": 0.18681625574827193,
      "num_tokens": 48068419.0,
      "step": 26055
    },
    {
      "entropy": 5.503195095062256,
      "epoch": 2.1893719806763285,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045177136966992234,
      "loss": 5.1637,
      "mean_token_accuracy": 0.1856912210583687,
      "num_tokens": 48077115.0,
      "step": 26060
    },
    {
      "entropy": 5.517162084579468,
      "epoch": 2.1897920604914933,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045175284105639806,
      "loss": 5.1929,
      "mean_token_accuracy": 0.18420469909906387,
      "num_tokens": 48085744.0,
      "step": 26065
    },
    {
      "entropy": 5.524509286880493,
      "epoch": 2.190212140306658,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045173430931178834,
      "loss": 5.2492,
      "mean_token_accuracy": 0.18381085395812988,
      "num_tokens": 48094829.0,
      "step": 26070
    },
    {
      "entropy": 5.612383127212524,
      "epoch": 2.1906322201218233,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004517157744364213,
      "loss": 5.2241,
      "mean_token_accuracy": 0.1775051549077034,
      "num_tokens": 48104086.0,
      "step": 26075
    },
    {
      "entropy": 5.549058103561402,
      "epoch": 2.191052299936988,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004516972364306253,
      "loss": 5.2305,
      "mean_token_accuracy": 0.18897917717695237,
      "num_tokens": 48112268.0,
      "step": 26080
    },
    {
      "entropy": 5.548488569259644,
      "epoch": 2.191472379752153,
      "grad_norm": 1.25,
      "learning_rate": 0.0004516786952947287,
      "loss": 5.142,
      "mean_token_accuracy": 0.19093283116817475,
      "num_tokens": 48121157.0,
      "step": 26085
    },
    {
      "entropy": 5.61608624458313,
      "epoch": 2.1918924595673177,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004516601510290601,
      "loss": 5.2857,
      "mean_token_accuracy": 0.17918362319469452,
      "num_tokens": 48131093.0,
      "step": 26090
    },
    {
      "entropy": 5.60462212562561,
      "epoch": 2.1923125393824825,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004516416036339478,
      "loss": 5.2464,
      "mean_token_accuracy": 0.17758110165596008,
      "num_tokens": 48140972.0,
      "step": 26095
    },
    {
      "entropy": 5.547426652908325,
      "epoch": 2.1927326191976477,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045162305310972046,
      "loss": 5.1616,
      "mean_token_accuracy": 0.18535294383764267,
      "num_tokens": 48150947.0,
      "step": 26100
    },
    {
      "entropy": 5.576661729812622,
      "epoch": 2.1931526990128125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045160449945670666,
      "loss": 5.2789,
      "mean_token_accuracy": 0.17353548258543014,
      "num_tokens": 48159859.0,
      "step": 26105
    },
    {
      "entropy": 5.531643009185791,
      "epoch": 2.1935727788279773,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004515859426752352,
      "loss": 5.1886,
      "mean_token_accuracy": 0.18886552155017852,
      "num_tokens": 48168779.0,
      "step": 26110
    },
    {
      "entropy": 5.510588788986206,
      "epoch": 2.193992858643142,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004515673827656345,
      "loss": 5.1575,
      "mean_token_accuracy": 0.19068916738033295,
      "num_tokens": 48178044.0,
      "step": 26115
    },
    {
      "entropy": 5.502061176300049,
      "epoch": 2.194412938458307,
      "grad_norm": 1.515625,
      "learning_rate": 0.00045154881972823377,
      "loss": 5.1311,
      "mean_token_accuracy": 0.1851045936346054,
      "num_tokens": 48186929.0,
      "step": 26120
    },
    {
      "entropy": 5.585215663909912,
      "epoch": 2.194833018273472,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004515302535633614,
      "loss": 5.2304,
      "mean_token_accuracy": 0.18430652022361754,
      "num_tokens": 48196677.0,
      "step": 26125
    },
    {
      "entropy": 5.6187464714050295,
      "epoch": 2.195253098088637,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004515116842713466,
      "loss": 5.3253,
      "mean_token_accuracy": 0.1817328676581383,
      "num_tokens": 48206317.0,
      "step": 26130
    },
    {
      "entropy": 5.5347710132598875,
      "epoch": 2.1956731779038017,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045149311185251816,
      "loss": 5.2397,
      "mean_token_accuracy": 0.18049706667661666,
      "num_tokens": 48215985.0,
      "step": 26135
    },
    {
      "entropy": 5.538452672958374,
      "epoch": 2.1960932577189665,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045147453630720513,
      "loss": 5.173,
      "mean_token_accuracy": 0.17959247529506683,
      "num_tokens": 48224939.0,
      "step": 26140
    },
    {
      "entropy": 5.556299543380737,
      "epoch": 2.1965133375341317,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004514559576357366,
      "loss": 5.1982,
      "mean_token_accuracy": 0.183385369181633,
      "num_tokens": 48234476.0,
      "step": 26145
    },
    {
      "entropy": 5.518967199325561,
      "epoch": 2.1969334173492965,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004514373758384416,
      "loss": 5.2139,
      "mean_token_accuracy": 0.18907600045204162,
      "num_tokens": 48243620.0,
      "step": 26150
    },
    {
      "entropy": 5.601477479934692,
      "epoch": 2.1973534971644613,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045141879091564936,
      "loss": 5.2923,
      "mean_token_accuracy": 0.17571036219596864,
      "num_tokens": 48253227.0,
      "step": 26155
    },
    {
      "entropy": 5.573794794082642,
      "epoch": 2.197773576979626,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004514002028676891,
      "loss": 5.1145,
      "mean_token_accuracy": 0.18485604971647263,
      "num_tokens": 48262304.0,
      "step": 26160
    },
    {
      "entropy": 5.617762422561645,
      "epoch": 2.198193656794791,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045138161169489,
      "loss": 5.2586,
      "mean_token_accuracy": 0.17732227295637132,
      "num_tokens": 48271810.0,
      "step": 26165
    },
    {
      "entropy": 5.463292789459229,
      "epoch": 2.198613736609956,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045136301739758157,
      "loss": 5.177,
      "mean_token_accuracy": 0.18305970579385758,
      "num_tokens": 48282223.0,
      "step": 26170
    },
    {
      "entropy": 5.577669668197632,
      "epoch": 2.199033816425121,
      "grad_norm": 1.734375,
      "learning_rate": 0.00045134441997609297,
      "loss": 5.3138,
      "mean_token_accuracy": 0.17167676538228988,
      "num_tokens": 48293376.0,
      "step": 26175
    },
    {
      "entropy": 5.61583456993103,
      "epoch": 2.1994538962402856,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004513258194307539,
      "loss": 5.2342,
      "mean_token_accuracy": 0.18126412332057953,
      "num_tokens": 48302976.0,
      "step": 26180
    },
    {
      "entropy": 5.615239191055298,
      "epoch": 2.1998739760554504,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004513072157618937,
      "loss": 5.2279,
      "mean_token_accuracy": 0.1861242413520813,
      "num_tokens": 48311145.0,
      "step": 26185
    },
    {
      "entropy": 5.512467288970948,
      "epoch": 2.200294055870615,
      "grad_norm": 1.28125,
      "learning_rate": 0.000451288608969842,
      "loss": 5.2477,
      "mean_token_accuracy": 0.18317418694496154,
      "num_tokens": 48320368.0,
      "step": 26190
    },
    {
      "entropy": 5.569041585922241,
      "epoch": 2.2007141356857804,
      "grad_norm": 1.125,
      "learning_rate": 0.00045126999905492827,
      "loss": 5.2147,
      "mean_token_accuracy": 0.18979903906583787,
      "num_tokens": 48329192.0,
      "step": 26195
    },
    {
      "entropy": 5.608487892150879,
      "epoch": 2.2011342155009452,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045125138601748225,
      "loss": 5.2094,
      "mean_token_accuracy": 0.1871396690607071,
      "num_tokens": 48338134.0,
      "step": 26200
    },
    {
      "entropy": 5.5634369373321535,
      "epoch": 2.20155429531611,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045123276985783377,
      "loss": 5.2399,
      "mean_token_accuracy": 0.17849182784557344,
      "num_tokens": 48347072.0,
      "step": 26205
    },
    {
      "entropy": 5.58306245803833,
      "epoch": 2.201974375131275,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045121415057631256,
      "loss": 5.2371,
      "mean_token_accuracy": 0.17817579358816146,
      "num_tokens": 48356528.0,
      "step": 26210
    },
    {
      "entropy": 5.641669130325317,
      "epoch": 2.2023944549464396,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004511955281732484,
      "loss": 5.3064,
      "mean_token_accuracy": 0.17816840261220931,
      "num_tokens": 48365875.0,
      "step": 26215
    },
    {
      "entropy": 5.551389884948731,
      "epoch": 2.202814534761605,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045117690264897105,
      "loss": 5.171,
      "mean_token_accuracy": 0.18769343048334122,
      "num_tokens": 48375016.0,
      "step": 26220
    },
    {
      "entropy": 5.499376440048218,
      "epoch": 2.2032346145767696,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004511582740038107,
      "loss": 5.272,
      "mean_token_accuracy": 0.17498243153095244,
      "num_tokens": 48384615.0,
      "step": 26225
    },
    {
      "entropy": 5.555689668655395,
      "epoch": 2.2036546943919344,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045113964223809715,
      "loss": 5.2131,
      "mean_token_accuracy": 0.1904737174510956,
      "num_tokens": 48393276.0,
      "step": 26230
    },
    {
      "entropy": 5.608794116973877,
      "epoch": 2.204074774207099,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004511210073521606,
      "loss": 5.163,
      "mean_token_accuracy": 0.1907846823334694,
      "num_tokens": 48402815.0,
      "step": 26235
    },
    {
      "entropy": 5.61847939491272,
      "epoch": 2.2044948540222644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004511023693463311,
      "loss": 5.3406,
      "mean_token_accuracy": 0.1799048438668251,
      "num_tokens": 48413122.0,
      "step": 26240
    },
    {
      "entropy": 5.5677601337432865,
      "epoch": 2.204914933837429,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004510837282209388,
      "loss": 5.2105,
      "mean_token_accuracy": 0.18694012016057968,
      "num_tokens": 48422423.0,
      "step": 26245
    },
    {
      "entropy": 5.579599857330322,
      "epoch": 2.205335013652594,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004510650839763139,
      "loss": 5.2538,
      "mean_token_accuracy": 0.18143879771232604,
      "num_tokens": 48431690.0,
      "step": 26250
    },
    {
      "entropy": 5.529786396026611,
      "epoch": 2.2057550934677588,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004510464366127867,
      "loss": 5.1943,
      "mean_token_accuracy": 0.184973306953907,
      "num_tokens": 48441827.0,
      "step": 26255
    },
    {
      "entropy": 5.6507248878479,
      "epoch": 2.2061751732829236,
      "grad_norm": 1.125,
      "learning_rate": 0.00045102778613068757,
      "loss": 5.2828,
      "mean_token_accuracy": 0.18272757530212402,
      "num_tokens": 48452520.0,
      "step": 26260
    },
    {
      "entropy": 5.5613096237182615,
      "epoch": 2.206595253098089,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004510091325303468,
      "loss": 5.1772,
      "mean_token_accuracy": 0.18480051308870316,
      "num_tokens": 48461756.0,
      "step": 26265
    },
    {
      "entropy": 5.519424295425415,
      "epoch": 2.2070153329132536,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045099047581209493,
      "loss": 5.2289,
      "mean_token_accuracy": 0.18311526626348495,
      "num_tokens": 48471415.0,
      "step": 26270
    },
    {
      "entropy": 5.5037031173706055,
      "epoch": 2.2074354127284184,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004509718159762623,
      "loss": 5.0173,
      "mean_token_accuracy": 0.19962505847215653,
      "num_tokens": 48480296.0,
      "step": 26275
    },
    {
      "entropy": 5.470512962341308,
      "epoch": 2.207855492543583,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045095315302317966,
      "loss": 5.2132,
      "mean_token_accuracy": 0.1824524313211441,
      "num_tokens": 48489176.0,
      "step": 26280
    },
    {
      "entropy": 5.517434310913086,
      "epoch": 2.2082755723587484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004509344869531775,
      "loss": 5.2247,
      "mean_token_accuracy": 0.18331936448812486,
      "num_tokens": 48499274.0,
      "step": 26285
    },
    {
      "entropy": 5.653523349761963,
      "epoch": 2.208695652173913,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045091581776658646,
      "loss": 5.2998,
      "mean_token_accuracy": 0.17889634519815445,
      "num_tokens": 48508515.0,
      "step": 26290
    },
    {
      "entropy": 5.606757545471192,
      "epoch": 2.209115731989078,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004508971454637373,
      "loss": 5.2756,
      "mean_token_accuracy": 0.17664328664541246,
      "num_tokens": 48517794.0,
      "step": 26295
    },
    {
      "entropy": 5.560481548309326,
      "epoch": 2.2095358118042427,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004508784700449608,
      "loss": 5.1823,
      "mean_token_accuracy": 0.18505071401596068,
      "num_tokens": 48526185.0,
      "step": 26300
    },
    {
      "entropy": 5.570043897628784,
      "epoch": 2.2099558916194075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004508597915105878,
      "loss": 5.2371,
      "mean_token_accuracy": 0.18625471740961075,
      "num_tokens": 48535562.0,
      "step": 26305
    },
    {
      "entropy": 5.4967893123626705,
      "epoch": 2.2103759714345728,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004508411098609491,
      "loss": 5.1475,
      "mean_token_accuracy": 0.1868896469473839,
      "num_tokens": 48545571.0,
      "step": 26310
    },
    {
      "entropy": 5.544439888000488,
      "epoch": 2.2107960512497375,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045082242509637566,
      "loss": 5.1667,
      "mean_token_accuracy": 0.1839189574122429,
      "num_tokens": 48554793.0,
      "step": 26315
    },
    {
      "entropy": 5.6095305442810055,
      "epoch": 2.2112161310649023,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045080373721719853,
      "loss": 5.2852,
      "mean_token_accuracy": 0.1774291217327118,
      "num_tokens": 48563346.0,
      "step": 26320
    },
    {
      "entropy": 5.564439582824707,
      "epoch": 2.211636210880067,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004507850462237488,
      "loss": 5.2277,
      "mean_token_accuracy": 0.1771346628665924,
      "num_tokens": 48572806.0,
      "step": 26325
    },
    {
      "entropy": 5.450683212280273,
      "epoch": 2.212056290695232,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045076635211635735,
      "loss": 5.173,
      "mean_token_accuracy": 0.1804874286055565,
      "num_tokens": 48583278.0,
      "step": 26330
    },
    {
      "entropy": 5.660644388198852,
      "epoch": 2.212476370510397,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045074765489535554,
      "loss": 5.2544,
      "mean_token_accuracy": 0.17726797610521317,
      "num_tokens": 48592824.0,
      "step": 26335
    },
    {
      "entropy": 5.58365044593811,
      "epoch": 2.212896450325562,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045072895456107456,
      "loss": 5.2204,
      "mean_token_accuracy": 0.181408828496933,
      "num_tokens": 48602353.0,
      "step": 26340
    },
    {
      "entropy": 5.656280374526977,
      "epoch": 2.2133165301407267,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045071025111384554,
      "loss": 5.379,
      "mean_token_accuracy": 0.16524552255868913,
      "num_tokens": 48611346.0,
      "step": 26345
    },
    {
      "entropy": 5.570623445510864,
      "epoch": 2.2137366099558915,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000450691544554,
      "loss": 5.1734,
      "mean_token_accuracy": 0.1919173553586006,
      "num_tokens": 48619882.0,
      "step": 26350
    },
    {
      "entropy": 5.493794679641724,
      "epoch": 2.2141566897710563,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004506728348818691,
      "loss": 5.1546,
      "mean_token_accuracy": 0.18940701484680175,
      "num_tokens": 48629003.0,
      "step": 26355
    },
    {
      "entropy": 5.573445844650268,
      "epoch": 2.2145767695862215,
      "grad_norm": 1.25,
      "learning_rate": 0.00045065412209778456,
      "loss": 5.2086,
      "mean_token_accuracy": 0.18517985492944716,
      "num_tokens": 48637858.0,
      "step": 26360
    },
    {
      "entropy": 5.620487642288208,
      "epoch": 2.2149968494013863,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004506354062020776,
      "loss": 5.3069,
      "mean_token_accuracy": 0.1786564841866493,
      "num_tokens": 48647935.0,
      "step": 26365
    },
    {
      "entropy": 5.5941376209259035,
      "epoch": 2.215416929216551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004506166871950798,
      "loss": 5.2137,
      "mean_token_accuracy": 0.19013828486204148,
      "num_tokens": 48657335.0,
      "step": 26370
    },
    {
      "entropy": 5.548377227783203,
      "epoch": 2.215837009031716,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004505979650771229,
      "loss": 5.2167,
      "mean_token_accuracy": 0.1876661255955696,
      "num_tokens": 48665665.0,
      "step": 26375
    },
    {
      "entropy": 5.642461061477661,
      "epoch": 2.216257088846881,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045057923984853843,
      "loss": 5.3032,
      "mean_token_accuracy": 0.17923520207405091,
      "num_tokens": 48674429.0,
      "step": 26380
    },
    {
      "entropy": 5.554673099517823,
      "epoch": 2.216677168662046,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045056051150965817,
      "loss": 5.1387,
      "mean_token_accuracy": 0.1877530872821808,
      "num_tokens": 48683550.0,
      "step": 26385
    },
    {
      "entropy": 5.564275646209717,
      "epoch": 2.2170972484772107,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045054178006081374,
      "loss": 5.2523,
      "mean_token_accuracy": 0.182179793715477,
      "num_tokens": 48692812.0,
      "step": 26390
    },
    {
      "entropy": 5.585380411148071,
      "epoch": 2.2175173282923755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004505230455023372,
      "loss": 5.186,
      "mean_token_accuracy": 0.18376641273498534,
      "num_tokens": 48700921.0,
      "step": 26395
    },
    {
      "entropy": 5.5918559551239015,
      "epoch": 2.2179374081075403,
      "grad_norm": 1.25,
      "learning_rate": 0.00045050430783456026,
      "loss": 5.2678,
      "mean_token_accuracy": 0.18038954734802246,
      "num_tokens": 48710092.0,
      "step": 26400
    },
    {
      "entropy": 5.488486051559448,
      "epoch": 2.2183574879227055,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004504855670578148,
      "loss": 5.1362,
      "mean_token_accuracy": 0.19088755548000336,
      "num_tokens": 48719660.0,
      "step": 26405
    },
    {
      "entropy": 5.6217738628387455,
      "epoch": 2.2187775677378703,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000450466823172433,
      "loss": 5.3052,
      "mean_token_accuracy": 0.17178284525871276,
      "num_tokens": 48729042.0,
      "step": 26410
    },
    {
      "entropy": 5.535201120376587,
      "epoch": 2.219197647553035,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004504480761787467,
      "loss": 5.0902,
      "mean_token_accuracy": 0.1857855200767517,
      "num_tokens": 48737550.0,
      "step": 26415
    },
    {
      "entropy": 5.49686222076416,
      "epoch": 2.2196177273682,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045042932607708804,
      "loss": 5.2545,
      "mean_token_accuracy": 0.18522861748933792,
      "num_tokens": 48745708.0,
      "step": 26420
    },
    {
      "entropy": 5.570236253738403,
      "epoch": 2.220037807183365,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004504105728677893,
      "loss": 5.2295,
      "mean_token_accuracy": 0.17426010519266127,
      "num_tokens": 48755319.0,
      "step": 26425
    },
    {
      "entropy": 5.668315982818603,
      "epoch": 2.22045788699853,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004503918165511825,
      "loss": 5.2939,
      "mean_token_accuracy": 0.1767928719520569,
      "num_tokens": 48764818.0,
      "step": 26430
    },
    {
      "entropy": 5.5070513725280765,
      "epoch": 2.2208779668136946,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004503730571276,
      "loss": 5.2051,
      "mean_token_accuracy": 0.182926943898201,
      "num_tokens": 48773790.0,
      "step": 26435
    },
    {
      "entropy": 5.504943609237671,
      "epoch": 2.2212980466288594,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004503542945973741,
      "loss": 5.2573,
      "mean_token_accuracy": 0.17833145558834076,
      "num_tokens": 48782987.0,
      "step": 26440
    },
    {
      "entropy": 5.623423147201538,
      "epoch": 2.221718126444024,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004503355289608372,
      "loss": 5.2552,
      "mean_token_accuracy": 0.1795510545372963,
      "num_tokens": 48792375.0,
      "step": 26445
    },
    {
      "entropy": 5.598920154571533,
      "epoch": 2.2221382062591895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004503167602183217,
      "loss": 5.2178,
      "mean_token_accuracy": 0.18116666227579117,
      "num_tokens": 48801972.0,
      "step": 26450
    },
    {
      "entropy": 5.537636137008667,
      "epoch": 2.2225582860743542,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004502979883701601,
      "loss": 5.2539,
      "mean_token_accuracy": 0.18317264169454575,
      "num_tokens": 48810510.0,
      "step": 26455
    },
    {
      "entropy": 5.592058229446411,
      "epoch": 2.222978365889519,
      "grad_norm": 1.546875,
      "learning_rate": 0.00045027921341668476,
      "loss": 5.2269,
      "mean_token_accuracy": 0.18022556155920028,
      "num_tokens": 48819528.0,
      "step": 26460
    },
    {
      "entropy": 5.540153503417969,
      "epoch": 2.223398445704684,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004502604353582285,
      "loss": 5.2158,
      "mean_token_accuracy": 0.1828978031873703,
      "num_tokens": 48828523.0,
      "step": 26465
    },
    {
      "entropy": 5.565620708465576,
      "epoch": 2.2238185255198486,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045024165419512384,
      "loss": 5.2257,
      "mean_token_accuracy": 0.1844302386045456,
      "num_tokens": 48838813.0,
      "step": 26470
    },
    {
      "entropy": 5.625452995300293,
      "epoch": 2.224238605335014,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004502228699277035,
      "loss": 5.232,
      "mean_token_accuracy": 0.17425791472196578,
      "num_tokens": 48847694.0,
      "step": 26475
    },
    {
      "entropy": 5.628651809692383,
      "epoch": 2.2246586851501786,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004502040825563003,
      "loss": 5.288,
      "mean_token_accuracy": 0.17827118039131165,
      "num_tokens": 48856927.0,
      "step": 26480
    },
    {
      "entropy": 5.526731157302857,
      "epoch": 2.2250787649653434,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045018529208124697,
      "loss": 5.2193,
      "mean_token_accuracy": 0.1791377916932106,
      "num_tokens": 48865971.0,
      "step": 26485
    },
    {
      "entropy": 5.66391773223877,
      "epoch": 2.225498844780508,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045016649850287646,
      "loss": 5.3104,
      "mean_token_accuracy": 0.18059692531824112,
      "num_tokens": 48874398.0,
      "step": 26490
    },
    {
      "entropy": 5.628163576126099,
      "epoch": 2.225918924595673,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045014770182152155,
      "loss": 5.1925,
      "mean_token_accuracy": 0.17827367186546325,
      "num_tokens": 48884414.0,
      "step": 26495
    },
    {
      "entropy": 5.506485939025879,
      "epoch": 2.226339004410838,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004501289020375153,
      "loss": 5.2031,
      "mean_token_accuracy": 0.18162247240543367,
      "num_tokens": 48893289.0,
      "step": 26500
    },
    {
      "entropy": 5.509644842147827,
      "epoch": 2.226759084226003,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004501100991511907,
      "loss": 5.225,
      "mean_token_accuracy": 0.1816165715456009,
      "num_tokens": 48902532.0,
      "step": 26505
    },
    {
      "entropy": 5.619868421554566,
      "epoch": 2.227179164041168,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004500912931628809,
      "loss": 5.214,
      "mean_token_accuracy": 0.18145921379327773,
      "num_tokens": 48910222.0,
      "step": 26510
    },
    {
      "entropy": 5.487632513046265,
      "epoch": 2.2275992438563326,
      "grad_norm": 1.171875,
      "learning_rate": 0.000450072484072919,
      "loss": 5.1824,
      "mean_token_accuracy": 0.19102548956871032,
      "num_tokens": 48919204.0,
      "step": 26515
    },
    {
      "entropy": 5.5337141990661625,
      "epoch": 2.2280193236714974,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004500536718816382,
      "loss": 5.1301,
      "mean_token_accuracy": 0.17968171387910842,
      "num_tokens": 48929041.0,
      "step": 26520
    },
    {
      "entropy": 5.540727043151856,
      "epoch": 2.2284394034866626,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004500348565893718,
      "loss": 5.2355,
      "mean_token_accuracy": 0.18427301943302155,
      "num_tokens": 48938610.0,
      "step": 26525
    },
    {
      "entropy": 5.616687488555908,
      "epoch": 2.2288594833018274,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045001603819645286,
      "loss": 5.2574,
      "mean_token_accuracy": 0.1826654076576233,
      "num_tokens": 48946859.0,
      "step": 26530
    },
    {
      "entropy": 5.6231849193573,
      "epoch": 2.229279563116992,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004499972167032151,
      "loss": 5.2967,
      "mean_token_accuracy": 0.1787433609366417,
      "num_tokens": 48955379.0,
      "step": 26535
    },
    {
      "entropy": 5.537955713272095,
      "epoch": 2.229699642932157,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004499783921099917,
      "loss": 5.1506,
      "mean_token_accuracy": 0.1901013270020485,
      "num_tokens": 48963970.0,
      "step": 26540
    },
    {
      "entropy": 5.507649660110474,
      "epoch": 2.230119722747322,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004499595644171162,
      "loss": 5.0819,
      "mean_token_accuracy": 0.18671495765447615,
      "num_tokens": 48972999.0,
      "step": 26545
    },
    {
      "entropy": 5.604022789001465,
      "epoch": 2.230539802562487,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044994073362492206,
      "loss": 5.2581,
      "mean_token_accuracy": 0.1858968883752823,
      "num_tokens": 48982173.0,
      "step": 26550
    },
    {
      "entropy": 5.653421640396118,
      "epoch": 2.2309598823776517,
      "grad_norm": 1.125,
      "learning_rate": 0.0004499218997337429,
      "loss": 5.2436,
      "mean_token_accuracy": 0.17890587449073792,
      "num_tokens": 48991484.0,
      "step": 26555
    },
    {
      "entropy": 5.6845954895019535,
      "epoch": 2.2313799621928165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004499030627439124,
      "loss": 5.2953,
      "mean_token_accuracy": 0.17637033462524415,
      "num_tokens": 49001749.0,
      "step": 26560
    },
    {
      "entropy": 5.532224512100219,
      "epoch": 2.2318000420079813,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044988422265576416,
      "loss": 5.2168,
      "mean_token_accuracy": 0.18262371867895127,
      "num_tokens": 49011649.0,
      "step": 26565
    },
    {
      "entropy": 5.518923997879028,
      "epoch": 2.2322201218231466,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044986537946963207,
      "loss": 5.1576,
      "mean_token_accuracy": 0.1895251139998436,
      "num_tokens": 49020290.0,
      "step": 26570
    },
    {
      "entropy": 5.544116544723511,
      "epoch": 2.2326402016383113,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044984653318584975,
      "loss": 5.1399,
      "mean_token_accuracy": 0.1874341368675232,
      "num_tokens": 49029496.0,
      "step": 26575
    },
    {
      "entropy": 5.545217895507813,
      "epoch": 2.233060281453476,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044982768380475113,
      "loss": 5.1982,
      "mean_token_accuracy": 0.1847977429628372,
      "num_tokens": 49038402.0,
      "step": 26580
    },
    {
      "entropy": 5.558094310760498,
      "epoch": 2.233480361268641,
      "grad_norm": 1.125,
      "learning_rate": 0.0004498088313266701,
      "loss": 5.2591,
      "mean_token_accuracy": 0.18075227588415146,
      "num_tokens": 49047568.0,
      "step": 26585
    },
    {
      "entropy": 5.614464569091797,
      "epoch": 2.233900441083806,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044978997575194064,
      "loss": 5.1757,
      "mean_token_accuracy": 0.1941094607114792,
      "num_tokens": 49056136.0,
      "step": 26590
    },
    {
      "entropy": 5.526753091812134,
      "epoch": 2.234320520898971,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004497711170808968,
      "loss": 5.1758,
      "mean_token_accuracy": 0.1860203504562378,
      "num_tokens": 49065307.0,
      "step": 26595
    },
    {
      "entropy": 5.690349388122558,
      "epoch": 2.2347406007141357,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044975225531387257,
      "loss": 5.3358,
      "mean_token_accuracy": 0.17914211004972458,
      "num_tokens": 49074619.0,
      "step": 26600
    },
    {
      "entropy": 5.73836760520935,
      "epoch": 2.2351606805293005,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004497333904512022,
      "loss": 5.4095,
      "mean_token_accuracy": 0.17640625089406967,
      "num_tokens": 49084932.0,
      "step": 26605
    },
    {
      "entropy": 5.562035942077637,
      "epoch": 2.2355807603444653,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004497145224932197,
      "loss": 5.2258,
      "mean_token_accuracy": 0.1814653292298317,
      "num_tokens": 49093203.0,
      "step": 26610
    },
    {
      "entropy": 5.590414810180664,
      "epoch": 2.2360008401596305,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044969565144025944,
      "loss": 5.231,
      "mean_token_accuracy": 0.19023652970790864,
      "num_tokens": 49101334.0,
      "step": 26615
    },
    {
      "entropy": 5.62574577331543,
      "epoch": 2.2364209199747953,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044967677729265573,
      "loss": 5.206,
      "mean_token_accuracy": 0.18439803123474122,
      "num_tokens": 49110179.0,
      "step": 26620
    },
    {
      "entropy": 5.506131029129028,
      "epoch": 2.23684099978996,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044965790005074277,
      "loss": 5.1557,
      "mean_token_accuracy": 0.19026787281036378,
      "num_tokens": 49119544.0,
      "step": 26625
    },
    {
      "entropy": 5.3976466178894045,
      "epoch": 2.237261079605125,
      "grad_norm": 1.125,
      "learning_rate": 0.00044963901971485507,
      "loss": 5.0283,
      "mean_token_accuracy": 0.20053786784410477,
      "num_tokens": 49128857.0,
      "step": 26630
    },
    {
      "entropy": 5.626853656768799,
      "epoch": 2.2376811594202897,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000449620136285327,
      "loss": 5.2907,
      "mean_token_accuracy": 0.17269188314676284,
      "num_tokens": 49138268.0,
      "step": 26635
    },
    {
      "entropy": 5.586330270767212,
      "epoch": 2.238101239235455,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044960124976249323,
      "loss": 5.2136,
      "mean_token_accuracy": 0.18517946898937226,
      "num_tokens": 49148436.0,
      "step": 26640
    },
    {
      "entropy": 5.557907772064209,
      "epoch": 2.2385213190506197,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004495823601466882,
      "loss": 5.2363,
      "mean_token_accuracy": 0.18687652349472045,
      "num_tokens": 49158477.0,
      "step": 26645
    },
    {
      "entropy": 5.568363380432129,
      "epoch": 2.2389413988657845,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004495634674382466,
      "loss": 5.1803,
      "mean_token_accuracy": 0.18877385407686234,
      "num_tokens": 49168589.0,
      "step": 26650
    },
    {
      "entropy": 5.592052268981933,
      "epoch": 2.2393614786809493,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000449544571637503,
      "loss": 5.2692,
      "mean_token_accuracy": 0.18027844727039338,
      "num_tokens": 49177865.0,
      "step": 26655
    },
    {
      "entropy": 5.5814310073852536,
      "epoch": 2.239781558496114,
      "grad_norm": 2.140625,
      "learning_rate": 0.00044952567274479216,
      "loss": 5.2435,
      "mean_token_accuracy": 0.18021032512187957,
      "num_tokens": 49187427.0,
      "step": 26660
    },
    {
      "entropy": 5.5100226402282715,
      "epoch": 2.2402016383112793,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004495067707604489,
      "loss": 5.2423,
      "mean_token_accuracy": 0.1882011190056801,
      "num_tokens": 49196833.0,
      "step": 26665
    },
    {
      "entropy": 5.574241876602173,
      "epoch": 2.240621718126444,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044948786568480806,
      "loss": 5.0623,
      "mean_token_accuracy": 0.1929984360933304,
      "num_tokens": 49205424.0,
      "step": 26670
    },
    {
      "entropy": 5.513074541091919,
      "epoch": 2.241041797941609,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044946895751820455,
      "loss": 5.1795,
      "mean_token_accuracy": 0.1837304338812828,
      "num_tokens": 49213934.0,
      "step": 26675
    },
    {
      "entropy": 5.545872068405151,
      "epoch": 2.2414618777567736,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044945004626097335,
      "loss": 5.2278,
      "mean_token_accuracy": 0.18263258039951324,
      "num_tokens": 49223875.0,
      "step": 26680
    },
    {
      "entropy": 5.512023448944092,
      "epoch": 2.241881957571939,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004494311319134493,
      "loss": 5.13,
      "mean_token_accuracy": 0.19264503121376036,
      "num_tokens": 49233298.0,
      "step": 26685
    },
    {
      "entropy": 5.517332887649536,
      "epoch": 2.2423020373871037,
      "grad_norm": 1.703125,
      "learning_rate": 0.00044941221447596764,
      "loss": 5.2382,
      "mean_token_accuracy": 0.17857345640659333,
      "num_tokens": 49242399.0,
      "step": 26690
    },
    {
      "entropy": 5.656974935531617,
      "epoch": 2.2427221172022684,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004493932939488634,
      "loss": 5.2857,
      "mean_token_accuracy": 0.1735707014799118,
      "num_tokens": 49251500.0,
      "step": 26695
    },
    {
      "entropy": 5.6401763439178465,
      "epoch": 2.2431421970174332,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004493743703324717,
      "loss": 5.1994,
      "mean_token_accuracy": 0.1872829094529152,
      "num_tokens": 49260525.0,
      "step": 26700
    },
    {
      "entropy": 5.540627813339233,
      "epoch": 2.243562276832598,
      "grad_norm": 1.25,
      "learning_rate": 0.0004493554436271279,
      "loss": 5.1967,
      "mean_token_accuracy": 0.18589021861553193,
      "num_tokens": 49271049.0,
      "step": 26705
    },
    {
      "entropy": 5.463886785507202,
      "epoch": 2.2439823566477632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044933651383316715,
      "loss": 5.1518,
      "mean_token_accuracy": 0.19345300048589706,
      "num_tokens": 49280707.0,
      "step": 26710
    },
    {
      "entropy": 5.611087656021118,
      "epoch": 2.244402436462928,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004493175809509249,
      "loss": 5.2801,
      "mean_token_accuracy": 0.17849674373865126,
      "num_tokens": 49291047.0,
      "step": 26715
    },
    {
      "entropy": 5.546320962905884,
      "epoch": 2.244822516278093,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044929864498073635,
      "loss": 5.1528,
      "mean_token_accuracy": 0.18586063683032988,
      "num_tokens": 49300085.0,
      "step": 26720
    },
    {
      "entropy": 5.474385404586792,
      "epoch": 2.2452425960932576,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004492797059229371,
      "loss": 5.0689,
      "mean_token_accuracy": 0.1942398115992546,
      "num_tokens": 49308876.0,
      "step": 26725
    },
    {
      "entropy": 5.532810544967651,
      "epoch": 2.245662675908423,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044926076377786256,
      "loss": 5.1391,
      "mean_token_accuracy": 0.18166615813970566,
      "num_tokens": 49317422.0,
      "step": 26730
    },
    {
      "entropy": 5.528246927261352,
      "epoch": 2.2460827557235876,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004492418185458484,
      "loss": 5.1593,
      "mean_token_accuracy": 0.1845986947417259,
      "num_tokens": 49325895.0,
      "step": 26735
    },
    {
      "entropy": 5.457337093353272,
      "epoch": 2.2465028355387524,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044922287022723016,
      "loss": 5.1384,
      "mean_token_accuracy": 0.18213678449392318,
      "num_tokens": 49335998.0,
      "step": 26740
    },
    {
      "entropy": 5.546372652053833,
      "epoch": 2.246922915353917,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004492039188223434,
      "loss": 5.2311,
      "mean_token_accuracy": 0.17811405658721924,
      "num_tokens": 49345630.0,
      "step": 26745
    },
    {
      "entropy": 5.472971725463867,
      "epoch": 2.247342995169082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044918496433152396,
      "loss": 5.0235,
      "mean_token_accuracy": 0.1881635457277298,
      "num_tokens": 49354818.0,
      "step": 26750
    },
    {
      "entropy": 5.527751111984253,
      "epoch": 2.247763074984247,
      "grad_norm": 1.25,
      "learning_rate": 0.0004491660067551076,
      "loss": 5.2278,
      "mean_token_accuracy": 0.1861103042960167,
      "num_tokens": 49364604.0,
      "step": 26755
    },
    {
      "entropy": 5.465829467773437,
      "epoch": 2.248183154799412,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044914704609343007,
      "loss": 5.0722,
      "mean_token_accuracy": 0.18719889968633652,
      "num_tokens": 49373362.0,
      "step": 26760
    },
    {
      "entropy": 5.490401887893677,
      "epoch": 2.248603234614577,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044912808234682736,
      "loss": 5.0704,
      "mean_token_accuracy": 0.1889788419008255,
      "num_tokens": 49381670.0,
      "step": 26765
    },
    {
      "entropy": 5.541893720626831,
      "epoch": 2.2490233144297416,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004491091155156352,
      "loss": 5.2115,
      "mean_token_accuracy": 0.18806126415729524,
      "num_tokens": 49390554.0,
      "step": 26770
    },
    {
      "entropy": 5.538183689117432,
      "epoch": 2.2494433942449064,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044909014560018986,
      "loss": 5.1703,
      "mean_token_accuracy": 0.18382145017385482,
      "num_tokens": 49400571.0,
      "step": 26775
    },
    {
      "entropy": 5.535412883758545,
      "epoch": 2.2498634740600716,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044907117260082713,
      "loss": 5.1398,
      "mean_token_accuracy": 0.1844318613409996,
      "num_tokens": 49409476.0,
      "step": 26780
    },
    {
      "entropy": 5.601237201690674,
      "epoch": 2.2502835538752364,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044905219651788326,
      "loss": 5.1943,
      "mean_token_accuracy": 0.1895414635539055,
      "num_tokens": 49418289.0,
      "step": 26785
    },
    {
      "entropy": 5.4910400390625,
      "epoch": 2.250703633690401,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044903321735169426,
      "loss": 5.107,
      "mean_token_accuracy": 0.18795407861471175,
      "num_tokens": 49427145.0,
      "step": 26790
    },
    {
      "entropy": 5.473565769195557,
      "epoch": 2.251123713505566,
      "grad_norm": 1.25,
      "learning_rate": 0.00044901423510259654,
      "loss": 5.1016,
      "mean_token_accuracy": 0.1913008525967598,
      "num_tokens": 49436018.0,
      "step": 26795
    },
    {
      "entropy": 5.510963582992554,
      "epoch": 2.2515437933207307,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044899524977092627,
      "loss": 5.1863,
      "mean_token_accuracy": 0.18121721744537353,
      "num_tokens": 49445288.0,
      "step": 26800
    },
    {
      "entropy": 5.561471748352051,
      "epoch": 2.251963873135896,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004489762613570196,
      "loss": 5.1673,
      "mean_token_accuracy": 0.1905657708644867,
      "num_tokens": 49454287.0,
      "step": 26805
    },
    {
      "entropy": 5.617612743377686,
      "epoch": 2.2523839529510608,
      "grad_norm": 1.125,
      "learning_rate": 0.0004489572698612131,
      "loss": 5.1817,
      "mean_token_accuracy": 0.18043486028909683,
      "num_tokens": 49462788.0,
      "step": 26810
    },
    {
      "entropy": 5.504171991348267,
      "epoch": 2.2528040327662255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004489382752838432,
      "loss": 5.3208,
      "mean_token_accuracy": 0.18149765580892563,
      "num_tokens": 49471517.0,
      "step": 26815
    },
    {
      "entropy": 5.565081548690796,
      "epoch": 2.2532241125813903,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004489192776252463,
      "loss": 5.1902,
      "mean_token_accuracy": 0.18276180028915406,
      "num_tokens": 49479651.0,
      "step": 26820
    },
    {
      "entropy": 5.542251205444336,
      "epoch": 2.253644192396555,
      "grad_norm": 1.125,
      "learning_rate": 0.00044890027688575896,
      "loss": 5.1468,
      "mean_token_accuracy": 0.1860174223780632,
      "num_tokens": 49488218.0,
      "step": 26825
    },
    {
      "entropy": 5.514791250228882,
      "epoch": 2.2540642722117203,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044888127306571776,
      "loss": 5.1605,
      "mean_token_accuracy": 0.19827504754066466,
      "num_tokens": 49498143.0,
      "step": 26830
    },
    {
      "entropy": 5.539866542816162,
      "epoch": 2.254484352026885,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044886226616545926,
      "loss": 5.1462,
      "mean_token_accuracy": 0.18986725211143493,
      "num_tokens": 49507923.0,
      "step": 26835
    },
    {
      "entropy": 5.5649971008300785,
      "epoch": 2.25490443184205,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044884325618532034,
      "loss": 5.2332,
      "mean_token_accuracy": 0.17870157212018967,
      "num_tokens": 49516872.0,
      "step": 26840
    },
    {
      "entropy": 5.523595190048217,
      "epoch": 2.2553245116572147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004488242431256376,
      "loss": 5.094,
      "mean_token_accuracy": 0.19251408576965331,
      "num_tokens": 49525786.0,
      "step": 26845
    },
    {
      "entropy": 5.553067064285278,
      "epoch": 2.25574459147238,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044880522698674796,
      "loss": 5.207,
      "mean_token_accuracy": 0.18804218769073486,
      "num_tokens": 49534028.0,
      "step": 26850
    },
    {
      "entropy": 5.570934009552002,
      "epoch": 2.2561646712875447,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044878620776898817,
      "loss": 5.2923,
      "mean_token_accuracy": 0.17990431189537048,
      "num_tokens": 49543105.0,
      "step": 26855
    },
    {
      "entropy": 5.575034141540527,
      "epoch": 2.2565847511027095,
      "grad_norm": 1.25,
      "learning_rate": 0.0004487671854726952,
      "loss": 5.2513,
      "mean_token_accuracy": 0.17906778156757355,
      "num_tokens": 49552842.0,
      "step": 26860
    },
    {
      "entropy": 5.558677339553833,
      "epoch": 2.2570048309178743,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044874816009820597,
      "loss": 5.1335,
      "mean_token_accuracy": 0.1863091066479683,
      "num_tokens": 49561952.0,
      "step": 26865
    },
    {
      "entropy": 5.539755010604859,
      "epoch": 2.2574249107330395,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004487291316458576,
      "loss": 5.2439,
      "mean_token_accuracy": 0.18432301580905913,
      "num_tokens": 49571435.0,
      "step": 26870
    },
    {
      "entropy": 5.554397678375244,
      "epoch": 2.2578449905482043,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004487101001159871,
      "loss": 5.2239,
      "mean_token_accuracy": 0.17967016696929933,
      "num_tokens": 49580205.0,
      "step": 26875
    },
    {
      "entropy": 5.541103744506836,
      "epoch": 2.258265070363369,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004486910655089317,
      "loss": 5.1489,
      "mean_token_accuracy": 0.18564743399620057,
      "num_tokens": 49588776.0,
      "step": 26880
    },
    {
      "entropy": 5.603924798965454,
      "epoch": 2.258685150178534,
      "grad_norm": 1.484375,
      "learning_rate": 0.00044867202782502847,
      "loss": 5.2624,
      "mean_token_accuracy": 0.1837603121995926,
      "num_tokens": 49597684.0,
      "step": 26885
    },
    {
      "entropy": 5.606839990615844,
      "epoch": 2.2591052299936987,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004486529870646147,
      "loss": 5.2275,
      "mean_token_accuracy": 0.18288439214229585,
      "num_tokens": 49606984.0,
      "step": 26890
    },
    {
      "entropy": 5.516634607315064,
      "epoch": 2.259525309808864,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044863394322802757,
      "loss": 5.1434,
      "mean_token_accuracy": 0.18772388249635696,
      "num_tokens": 49615908.0,
      "step": 26895
    },
    {
      "entropy": 5.569117212295533,
      "epoch": 2.2599453896240287,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004486148963156047,
      "loss": 5.2331,
      "mean_token_accuracy": 0.18966741263866424,
      "num_tokens": 49624308.0,
      "step": 26900
    },
    {
      "entropy": 5.539221096038818,
      "epoch": 2.2603654694391935,
      "grad_norm": 1.25,
      "learning_rate": 0.0004485958463276832,
      "loss": 5.1364,
      "mean_token_accuracy": 0.181783065199852,
      "num_tokens": 49634229.0,
      "step": 26905
    },
    {
      "entropy": 5.4645123958587645,
      "epoch": 2.2607855492543583,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044857679326460077,
      "loss": 5.1717,
      "mean_token_accuracy": 0.18668661564588546,
      "num_tokens": 49643129.0,
      "step": 26910
    },
    {
      "entropy": 5.578624296188354,
      "epoch": 2.261205629069523,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004485577371266948,
      "loss": 5.2186,
      "mean_token_accuracy": 0.1793380230665207,
      "num_tokens": 49652494.0,
      "step": 26915
    },
    {
      "entropy": 5.54378604888916,
      "epoch": 2.2616257088846883,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044853867791430293,
      "loss": 5.1274,
      "mean_token_accuracy": 0.19577271044254302,
      "num_tokens": 49662339.0,
      "step": 26920
    },
    {
      "entropy": 5.472779226303101,
      "epoch": 2.262045788699853,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044851961562776267,
      "loss": 5.0855,
      "mean_token_accuracy": 0.19263802468776703,
      "num_tokens": 49671582.0,
      "step": 26925
    },
    {
      "entropy": 5.4504491806030275,
      "epoch": 2.262465868515018,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004485005502674118,
      "loss": 5.1779,
      "mean_token_accuracy": 0.18331872224807738,
      "num_tokens": 49680391.0,
      "step": 26930
    },
    {
      "entropy": 5.590726566314697,
      "epoch": 2.2628859483301826,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044848148183358803,
      "loss": 5.258,
      "mean_token_accuracy": 0.18223726451396943,
      "num_tokens": 49689701.0,
      "step": 26935
    },
    {
      "entropy": 5.621439695358276,
      "epoch": 2.2633060281453474,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004484624103266291,
      "loss": 5.214,
      "mean_token_accuracy": 0.18594740629196166,
      "num_tokens": 49699498.0,
      "step": 26940
    },
    {
      "entropy": 5.548307228088379,
      "epoch": 2.2637261079605127,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004484433357468729,
      "loss": 5.1494,
      "mean_token_accuracy": 0.18524881452322006,
      "num_tokens": 49708416.0,
      "step": 26945
    },
    {
      "entropy": 5.5698929786682125,
      "epoch": 2.2641461877756774,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004484242580946574,
      "loss": 5.2907,
      "mean_token_accuracy": 0.18228030651807786,
      "num_tokens": 49717483.0,
      "step": 26950
    },
    {
      "entropy": 5.5982325077056885,
      "epoch": 2.2645662675908422,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004484051773703204,
      "loss": 5.3608,
      "mean_token_accuracy": 0.17621270716190338,
      "num_tokens": 49727073.0,
      "step": 26955
    },
    {
      "entropy": 5.563378763198853,
      "epoch": 2.264986347406007,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044838609357420004,
      "loss": 5.1754,
      "mean_token_accuracy": 0.1842700108885765,
      "num_tokens": 49735523.0,
      "step": 26960
    },
    {
      "entropy": 5.524639415740967,
      "epoch": 2.265406427221172,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004483670067066342,
      "loss": 5.1957,
      "mean_token_accuracy": 0.1822855591773987,
      "num_tokens": 49745347.0,
      "step": 26965
    },
    {
      "entropy": 5.565494203567505,
      "epoch": 2.265826507036337,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044834791676796117,
      "loss": 5.2426,
      "mean_token_accuracy": 0.18487610965967177,
      "num_tokens": 49755076.0,
      "step": 26970
    },
    {
      "entropy": 5.587360572814942,
      "epoch": 2.266246586851502,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044832882375851903,
      "loss": 5.2061,
      "mean_token_accuracy": 0.18480460047721864,
      "num_tokens": 49763580.0,
      "step": 26975
    },
    {
      "entropy": 5.533010196685791,
      "epoch": 2.2666666666666666,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044830972767864605,
      "loss": 5.2288,
      "mean_token_accuracy": 0.1806843101978302,
      "num_tokens": 49773167.0,
      "step": 26980
    },
    {
      "entropy": 5.4855812072753904,
      "epoch": 2.2670867464818314,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004482906285286804,
      "loss": 5.1624,
      "mean_token_accuracy": 0.1852910816669464,
      "num_tokens": 49782273.0,
      "step": 26985
    },
    {
      "entropy": 5.562114572525024,
      "epoch": 2.267506826296996,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044827152630896064,
      "loss": 5.2074,
      "mean_token_accuracy": 0.18233684599399566,
      "num_tokens": 49791426.0,
      "step": 26990
    },
    {
      "entropy": 5.661167430877685,
      "epoch": 2.2679269061121614,
      "grad_norm": 1.25,
      "learning_rate": 0.00044825242101982496,
      "loss": 5.2443,
      "mean_token_accuracy": 0.17909808456897736,
      "num_tokens": 49799643.0,
      "step": 26995
    },
    {
      "entropy": 5.553542804718018,
      "epoch": 2.268346985927326,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004482333126616118,
      "loss": 5.2339,
      "mean_token_accuracy": 0.1813248485326767,
      "num_tokens": 49809401.0,
      "step": 27000
    },
    {
      "epoch": 2.268346985927326,
      "eval_entropy": 5.29914059646123,
      "eval_loss": 5.307155132293701,
      "eval_mean_token_accuracy": 0.18789035074207058,
      "eval_num_tokens": 49809401.0,
      "eval_runtime": 27.3996,
      "eval_samples_per_second": 1363.744,
      "eval_steps_per_second": 170.477,
      "step": 27000
    },
    {
      "entropy": 5.480902576446534,
      "epoch": 2.268767065742491,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044821420123465974,
      "loss": 5.1698,
      "mean_token_accuracy": 0.18958532810211182,
      "num_tokens": 49818395.0,
      "step": 27005
    },
    {
      "entropy": 5.612220001220703,
      "epoch": 2.2691871455576558,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044819508673930733,
      "loss": 5.2884,
      "mean_token_accuracy": 0.17911423444747926,
      "num_tokens": 49828057.0,
      "step": 27010
    },
    {
      "entropy": 5.549051427841187,
      "epoch": 2.269607225372821,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044817596917589306,
      "loss": 5.1524,
      "mean_token_accuracy": 0.1815236732363701,
      "num_tokens": 49837581.0,
      "step": 27015
    },
    {
      "entropy": 5.549178838729858,
      "epoch": 2.270027305187986,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044815684854475574,
      "loss": 5.1814,
      "mean_token_accuracy": 0.1922007218003273,
      "num_tokens": 49846462.0,
      "step": 27020
    },
    {
      "entropy": 5.431716108322144,
      "epoch": 2.2704473850031506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000448137724846234,
      "loss": 5.1233,
      "mean_token_accuracy": 0.18601950109004975,
      "num_tokens": 49855185.0,
      "step": 27025
    },
    {
      "entropy": 5.648405408859253,
      "epoch": 2.2708674648183154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044811859808066665,
      "loss": 5.2916,
      "mean_token_accuracy": 0.18535915911197662,
      "num_tokens": 49865358.0,
      "step": 27030
    },
    {
      "entropy": 5.6609211444854735,
      "epoch": 2.2712875446334806,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044809946824839235,
      "loss": 5.2034,
      "mean_token_accuracy": 0.18694685399532318,
      "num_tokens": 49873850.0,
      "step": 27035
    },
    {
      "entropy": 5.527661848068237,
      "epoch": 2.2717076244486454,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004480803353497503,
      "loss": 5.1506,
      "mean_token_accuracy": 0.18766432404518127,
      "num_tokens": 49882925.0,
      "step": 27040
    },
    {
      "entropy": 5.424364042282105,
      "epoch": 2.27212770426381,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00044806119938507906,
      "loss": 5.0888,
      "mean_token_accuracy": 0.1945415049791336,
      "num_tokens": 49891620.0,
      "step": 27045
    },
    {
      "entropy": 5.529623794555664,
      "epoch": 2.272547784078975,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004480420603547179,
      "loss": 5.2828,
      "mean_token_accuracy": 0.18426181226968766,
      "num_tokens": 49901654.0,
      "step": 27050
    },
    {
      "entropy": 5.550444078445435,
      "epoch": 2.2729678638941397,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044802291825900563,
      "loss": 5.2431,
      "mean_token_accuracy": 0.18601664751768113,
      "num_tokens": 49910437.0,
      "step": 27055
    },
    {
      "entropy": 5.620884132385254,
      "epoch": 2.273387943709305,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044800377309828156,
      "loss": 5.2731,
      "mean_token_accuracy": 0.1786436140537262,
      "num_tokens": 49919309.0,
      "step": 27060
    },
    {
      "entropy": 5.567262935638428,
      "epoch": 2.2738080235244698,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044798462487288473,
      "loss": 5.1839,
      "mean_token_accuracy": 0.1898314133286476,
      "num_tokens": 49927729.0,
      "step": 27065
    },
    {
      "entropy": 5.473944234848022,
      "epoch": 2.2742281033396345,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004479654735831543,
      "loss": 5.1587,
      "mean_token_accuracy": 0.19413935244083405,
      "num_tokens": 49937365.0,
      "step": 27070
    },
    {
      "entropy": 5.543932723999023,
      "epoch": 2.2746481831547993,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004479463192294295,
      "loss": 5.2324,
      "mean_token_accuracy": 0.18583708256483078,
      "num_tokens": 49946084.0,
      "step": 27075
    },
    {
      "entropy": 5.533625078201294,
      "epoch": 2.275068262969964,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004479271618120498,
      "loss": 5.1123,
      "mean_token_accuracy": 0.1862211659550667,
      "num_tokens": 49955841.0,
      "step": 27080
    },
    {
      "entropy": 5.567842960357666,
      "epoch": 2.2754883427851293,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004479080013313545,
      "loss": 5.321,
      "mean_token_accuracy": 0.17727283835411073,
      "num_tokens": 49965127.0,
      "step": 27085
    },
    {
      "entropy": 5.547437143325806,
      "epoch": 2.275908422600294,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004478888377876829,
      "loss": 5.2113,
      "mean_token_accuracy": 0.18301722556352615,
      "num_tokens": 49974099.0,
      "step": 27090
    },
    {
      "entropy": 5.6180031299591064,
      "epoch": 2.276328502415459,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044786967118137463,
      "loss": 5.3022,
      "mean_token_accuracy": 0.17837065160274507,
      "num_tokens": 49983667.0,
      "step": 27095
    },
    {
      "entropy": 5.46148567199707,
      "epoch": 2.2767485822306237,
      "grad_norm": 1.25,
      "learning_rate": 0.0004478505015127691,
      "loss": 5.1454,
      "mean_token_accuracy": 0.19103490710258483,
      "num_tokens": 49993278.0,
      "step": 27100
    },
    {
      "entropy": 5.545971536636353,
      "epoch": 2.2771686620457885,
      "grad_norm": 1.28125,
      "learning_rate": 0.000447831328782206,
      "loss": 5.1396,
      "mean_token_accuracy": 0.18627176582813262,
      "num_tokens": 50002679.0,
      "step": 27105
    },
    {
      "entropy": 5.6212574481964115,
      "epoch": 2.2775887418609537,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044781215299002485,
      "loss": 5.2372,
      "mean_token_accuracy": 0.18435110449790953,
      "num_tokens": 50012019.0,
      "step": 27110
    },
    {
      "entropy": 5.488028812408447,
      "epoch": 2.2780088216761185,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044779297413656544,
      "loss": 5.2559,
      "mean_token_accuracy": 0.18455987870693208,
      "num_tokens": 50021024.0,
      "step": 27115
    },
    {
      "entropy": 5.592011594772339,
      "epoch": 2.2784289014912833,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004477737922221675,
      "loss": 5.231,
      "mean_token_accuracy": 0.17774474769830703,
      "num_tokens": 50029919.0,
      "step": 27120
    },
    {
      "entropy": 5.559778022766113,
      "epoch": 2.278848981306448,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004477546072471707,
      "loss": 5.0945,
      "mean_token_accuracy": 0.1956455424427986,
      "num_tokens": 50038779.0,
      "step": 27125
    },
    {
      "entropy": 5.526449346542359,
      "epoch": 2.279269061121613,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000447735419211915,
      "loss": 5.1686,
      "mean_token_accuracy": 0.186903777718544,
      "num_tokens": 50047107.0,
      "step": 27130
    },
    {
      "entropy": 5.54357647895813,
      "epoch": 2.279689140936778,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004477162281167404,
      "loss": 5.1565,
      "mean_token_accuracy": 0.18744371384382247,
      "num_tokens": 50055268.0,
      "step": 27135
    },
    {
      "entropy": 5.554645967483521,
      "epoch": 2.280109220751943,
      "grad_norm": 1.25,
      "learning_rate": 0.00044769703396198664,
      "loss": 5.2219,
      "mean_token_accuracy": 0.17888370901346207,
      "num_tokens": 50064356.0,
      "step": 27140
    },
    {
      "entropy": 5.587028932571411,
      "epoch": 2.2805293005671077,
      "grad_norm": 1.328125,
      "learning_rate": 0.000447677836747994,
      "loss": 5.1738,
      "mean_token_accuracy": 0.18374885767698287,
      "num_tokens": 50072968.0,
      "step": 27145
    },
    {
      "entropy": 5.500021076202392,
      "epoch": 2.2809493803822725,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004476586364751022,
      "loss": 5.1306,
      "mean_token_accuracy": 0.18980663269758224,
      "num_tokens": 50081600.0,
      "step": 27150
    },
    {
      "entropy": 5.560178804397583,
      "epoch": 2.2813694601974377,
      "grad_norm": 1.421875,
      "learning_rate": 0.00044763943314365164,
      "loss": 5.2707,
      "mean_token_accuracy": 0.17931890040636062,
      "num_tokens": 50091406.0,
      "step": 27155
    },
    {
      "entropy": 5.553939723968506,
      "epoch": 2.2817895400126025,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004476202267539824,
      "loss": 5.1898,
      "mean_token_accuracy": 0.18090270459651947,
      "num_tokens": 50100508.0,
      "step": 27160
    },
    {
      "entropy": 5.66865086555481,
      "epoch": 2.2822096198277673,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004476010173064348,
      "loss": 5.3429,
      "mean_token_accuracy": 0.17665794789791106,
      "num_tokens": 50110441.0,
      "step": 27165
    },
    {
      "entropy": 5.564552879333496,
      "epoch": 2.282629699642932,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044758180480134895,
      "loss": 5.1362,
      "mean_token_accuracy": 0.18999483287334443,
      "num_tokens": 50119896.0,
      "step": 27170
    },
    {
      "entropy": 5.477430629730224,
      "epoch": 2.2830497794580973,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044756258923906545,
      "loss": 5.1489,
      "mean_token_accuracy": 0.1883263662457466,
      "num_tokens": 50128988.0,
      "step": 27175
    },
    {
      "entropy": 5.527046346664429,
      "epoch": 2.283469859273262,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044754337061992446,
      "loss": 5.2026,
      "mean_token_accuracy": 0.1826902449131012,
      "num_tokens": 50138431.0,
      "step": 27180
    },
    {
      "entropy": 5.564195156097412,
      "epoch": 2.283889939088427,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044752414894426646,
      "loss": 5.1728,
      "mean_token_accuracy": 0.17930901050567627,
      "num_tokens": 50148149.0,
      "step": 27185
    },
    {
      "entropy": 5.64191837310791,
      "epoch": 2.2843100189035916,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000447504924212432,
      "loss": 5.3349,
      "mean_token_accuracy": 0.17403687685728073,
      "num_tokens": 50157982.0,
      "step": 27190
    },
    {
      "entropy": 5.571288967132569,
      "epoch": 2.2847300987187564,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004474856964247617,
      "loss": 5.1968,
      "mean_token_accuracy": 0.18573080897331237,
      "num_tokens": 50167297.0,
      "step": 27195
    },
    {
      "entropy": 5.396124458312988,
      "epoch": 2.2851501785339217,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044746646558159605,
      "loss": 5.0924,
      "mean_token_accuracy": 0.19680628627538682,
      "num_tokens": 50176562.0,
      "step": 27200
    },
    {
      "entropy": 5.4948814868927,
      "epoch": 2.2855702583490864,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004474472316832758,
      "loss": 5.2188,
      "mean_token_accuracy": 0.1876385673880577,
      "num_tokens": 50186404.0,
      "step": 27205
    },
    {
      "entropy": 5.508620119094848,
      "epoch": 2.2859903381642512,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004474279947301416,
      "loss": 5.1356,
      "mean_token_accuracy": 0.18553956151008605,
      "num_tokens": 50196038.0,
      "step": 27210
    },
    {
      "entropy": 5.670940113067627,
      "epoch": 2.286410417979416,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004474087547225343,
      "loss": 5.3378,
      "mean_token_accuracy": 0.17816144824028016,
      "num_tokens": 50205612.0,
      "step": 27215
    },
    {
      "entropy": 5.584204864501953,
      "epoch": 2.286830497794581,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004473895116607947,
      "loss": 5.2004,
      "mean_token_accuracy": 0.18115537762641906,
      "num_tokens": 50214212.0,
      "step": 27220
    },
    {
      "entropy": 5.57800407409668,
      "epoch": 2.287250577609746,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004473702655452636,
      "loss": 5.2528,
      "mean_token_accuracy": 0.18222227841615676,
      "num_tokens": 50223420.0,
      "step": 27225
    },
    {
      "entropy": 5.56588830947876,
      "epoch": 2.287670657424911,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004473510163762821,
      "loss": 5.2153,
      "mean_token_accuracy": 0.18336946219205857,
      "num_tokens": 50232308.0,
      "step": 27230
    },
    {
      "entropy": 5.564267158508301,
      "epoch": 2.2880907372400756,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000447331764154191,
      "loss": 5.1881,
      "mean_token_accuracy": 0.1849137008190155,
      "num_tokens": 50240611.0,
      "step": 27235
    },
    {
      "entropy": 5.5914946556091305,
      "epoch": 2.2885108170552404,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004473125088793315,
      "loss": 5.2317,
      "mean_token_accuracy": 0.18207349628210068,
      "num_tokens": 50249928.0,
      "step": 27240
    },
    {
      "entropy": 5.541820478439331,
      "epoch": 2.288930896870405,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044729325055204464,
      "loss": 5.225,
      "mean_token_accuracy": 0.17706434577703475,
      "num_tokens": 50259465.0,
      "step": 27245
    },
    {
      "entropy": 5.599516534805298,
      "epoch": 2.2893509766855704,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004472739891726715,
      "loss": 5.3147,
      "mean_token_accuracy": 0.17940813899040223,
      "num_tokens": 50269734.0,
      "step": 27250
    },
    {
      "entropy": 5.559195232391358,
      "epoch": 2.289771056500735,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004472547247415534,
      "loss": 5.1266,
      "mean_token_accuracy": 0.18790208846330642,
      "num_tokens": 50279016.0,
      "step": 27255
    },
    {
      "entropy": 5.504245328903198,
      "epoch": 2.2901911363159,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044723545725903146,
      "loss": 5.205,
      "mean_token_accuracy": 0.18973490595817566,
      "num_tokens": 50288039.0,
      "step": 27260
    },
    {
      "entropy": 5.510759353637695,
      "epoch": 2.2906112161310648,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004472161867254472,
      "loss": 5.2195,
      "mean_token_accuracy": 0.1859665498137474,
      "num_tokens": 50297495.0,
      "step": 27265
    },
    {
      "entropy": 5.604281282424926,
      "epoch": 2.2910312959462296,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004471969131411418,
      "loss": 5.2048,
      "mean_token_accuracy": 0.18373763412237168,
      "num_tokens": 50306103.0,
      "step": 27270
    },
    {
      "entropy": 5.508769845962524,
      "epoch": 2.291451375761395,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004471776365064568,
      "loss": 5.2025,
      "mean_token_accuracy": 0.19409721344709396,
      "num_tokens": 50315736.0,
      "step": 27275
    },
    {
      "entropy": 5.636666631698608,
      "epoch": 2.2918714555765596,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044715835682173364,
      "loss": 5.3753,
      "mean_token_accuracy": 0.1804428666830063,
      "num_tokens": 50324742.0,
      "step": 27280
    },
    {
      "entropy": 5.601318597793579,
      "epoch": 2.2922915353917244,
      "grad_norm": 1.375,
      "learning_rate": 0.00044713907408731375,
      "loss": 5.2916,
      "mean_token_accuracy": 0.17474572658538817,
      "num_tokens": 50334322.0,
      "step": 27285
    },
    {
      "entropy": 5.579921865463257,
      "epoch": 2.292711615206889,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004471197883035388,
      "loss": 5.1663,
      "mean_token_accuracy": 0.19652070254087448,
      "num_tokens": 50343587.0,
      "step": 27290
    },
    {
      "entropy": 5.521090078353882,
      "epoch": 2.2931316950220544,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044710049947075054,
      "loss": 5.2286,
      "mean_token_accuracy": 0.1837606355547905,
      "num_tokens": 50352978.0,
      "step": 27295
    },
    {
      "entropy": 5.531412410736084,
      "epoch": 2.293551774837219,
      "grad_norm": 1.125,
      "learning_rate": 0.0004470812075892905,
      "loss": 5.2259,
      "mean_token_accuracy": 0.1796173185110092,
      "num_tokens": 50361528.0,
      "step": 27300
    },
    {
      "entropy": 5.576286363601684,
      "epoch": 2.293971854652384,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044706191265950044,
      "loss": 5.1824,
      "mean_token_accuracy": 0.17727885842323304,
      "num_tokens": 50371362.0,
      "step": 27305
    },
    {
      "entropy": 5.563633394241333,
      "epoch": 2.2943919344675487,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044704261468172217,
      "loss": 5.2232,
      "mean_token_accuracy": 0.18285203129053115,
      "num_tokens": 50380976.0,
      "step": 27310
    },
    {
      "entropy": 5.528816604614258,
      "epoch": 2.2948120142827135,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004470233136562977,
      "loss": 5.1962,
      "mean_token_accuracy": 0.1825564593076706,
      "num_tokens": 50390135.0,
      "step": 27315
    },
    {
      "entropy": 5.501251602172852,
      "epoch": 2.2952320940978788,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044700400958356867,
      "loss": 5.1467,
      "mean_token_accuracy": 0.1900958776473999,
      "num_tokens": 50398979.0,
      "step": 27320
    },
    {
      "entropy": 5.493986988067627,
      "epoch": 2.2956521739130435,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004469847024638773,
      "loss": 5.2301,
      "mean_token_accuracy": 0.17458786368370055,
      "num_tokens": 50407901.0,
      "step": 27325
    },
    {
      "entropy": 5.58450984954834,
      "epoch": 2.2960722537282083,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044696539229756543,
      "loss": 5.2043,
      "mean_token_accuracy": 0.18488730639219284,
      "num_tokens": 50416085.0,
      "step": 27330
    },
    {
      "entropy": 5.491139316558838,
      "epoch": 2.296492333543373,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044694607908497515,
      "loss": 5.1325,
      "mean_token_accuracy": 0.19827849864959718,
      "num_tokens": 50425697.0,
      "step": 27335
    },
    {
      "entropy": 5.5153947353363035,
      "epoch": 2.2969124133585384,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004469267628264486,
      "loss": 5.1503,
      "mean_token_accuracy": 0.18757028728723527,
      "num_tokens": 50434831.0,
      "step": 27340
    },
    {
      "entropy": 5.536505937576294,
      "epoch": 2.297332493173703,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004469074435223281,
      "loss": 5.2003,
      "mean_token_accuracy": 0.18889368176460267,
      "num_tokens": 50443217.0,
      "step": 27345
    },
    {
      "entropy": 5.575624179840088,
      "epoch": 2.297752572988868,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004468881211729556,
      "loss": 5.2229,
      "mean_token_accuracy": 0.1839095190167427,
      "num_tokens": 50452523.0,
      "step": 27350
    },
    {
      "entropy": 5.537959384918213,
      "epoch": 2.2981726528040327,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044686879577867376,
      "loss": 5.1616,
      "mean_token_accuracy": 0.1878587618470192,
      "num_tokens": 50460793.0,
      "step": 27355
    },
    {
      "entropy": 5.6117335796356205,
      "epoch": 2.2985927326191975,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004468494673398245,
      "loss": 5.2778,
      "mean_token_accuracy": 0.18601578176021577,
      "num_tokens": 50469930.0,
      "step": 27360
    },
    {
      "entropy": 5.553171062469483,
      "epoch": 2.2990128124343627,
      "grad_norm": 1.125,
      "learning_rate": 0.0004468301358567505,
      "loss": 5.1931,
      "mean_token_accuracy": 0.1820788323879242,
      "num_tokens": 50478711.0,
      "step": 27365
    },
    {
      "entropy": 5.51889853477478,
      "epoch": 2.2994328922495275,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004468108013297941,
      "loss": 5.1354,
      "mean_token_accuracy": 0.19366977512836456,
      "num_tokens": 50488753.0,
      "step": 27370
    },
    {
      "entropy": 5.565305948257446,
      "epoch": 2.2998529720646923,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004467914637592978,
      "loss": 5.21,
      "mean_token_accuracy": 0.18851572275161743,
      "num_tokens": 50497956.0,
      "step": 27375
    },
    {
      "entropy": 5.556532478332519,
      "epoch": 2.300273051879857,
      "grad_norm": 1.125,
      "learning_rate": 0.0004467721231456042,
      "loss": 5.2159,
      "mean_token_accuracy": 0.17997783422470093,
      "num_tokens": 50507606.0,
      "step": 27380
    },
    {
      "entropy": 5.550961112976074,
      "epoch": 2.300693131695022,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004467527794890559,
      "loss": 5.2461,
      "mean_token_accuracy": 0.18171168118715286,
      "num_tokens": 50517320.0,
      "step": 27385
    },
    {
      "entropy": 5.508461332321167,
      "epoch": 2.301113211510187,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004467334327899955,
      "loss": 5.191,
      "mean_token_accuracy": 0.18277752101421357,
      "num_tokens": 50525806.0,
      "step": 27390
    },
    {
      "entropy": 5.639804553985596,
      "epoch": 2.301533291325352,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004467140830487658,
      "loss": 5.2713,
      "mean_token_accuracy": 0.17378807663917542,
      "num_tokens": 50534548.0,
      "step": 27395
    },
    {
      "entropy": 5.567071008682251,
      "epoch": 2.3019533711405167,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044669473026570964,
      "loss": 5.2083,
      "mean_token_accuracy": 0.18211232572793962,
      "num_tokens": 50543628.0,
      "step": 27400
    },
    {
      "entropy": 5.54638352394104,
      "epoch": 2.3023734509556815,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044667537444116964,
      "loss": 5.3393,
      "mean_token_accuracy": 0.17462950348854064,
      "num_tokens": 50553433.0,
      "step": 27405
    },
    {
      "entropy": 5.575443506240845,
      "epoch": 2.3027935307708463,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044665601557548873,
      "loss": 5.154,
      "mean_token_accuracy": 0.18064171969890594,
      "num_tokens": 50562736.0,
      "step": 27410
    },
    {
      "entropy": 5.541045188903809,
      "epoch": 2.3032136105860115,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044663665366900996,
      "loss": 5.1217,
      "mean_token_accuracy": 0.18464245200157164,
      "num_tokens": 50571492.0,
      "step": 27415
    },
    {
      "entropy": 5.421808433532715,
      "epoch": 2.3036336904011763,
      "grad_norm": 1.578125,
      "learning_rate": 0.00044661728872207626,
      "loss": 5.0696,
      "mean_token_accuracy": 0.18837596029043197,
      "num_tokens": 50580525.0,
      "step": 27420
    },
    {
      "entropy": 5.672775840759277,
      "epoch": 2.304053770216341,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044659792073503067,
      "loss": 5.44,
      "mean_token_accuracy": 0.17098142504692077,
      "num_tokens": 50590350.0,
      "step": 27425
    },
    {
      "entropy": 5.52259373664856,
      "epoch": 2.304473850031506,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004465785497082162,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19573718607425689,
      "num_tokens": 50599017.0,
      "step": 27430
    },
    {
      "entropy": 5.469167947769165,
      "epoch": 2.3048939298466706,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004465591756419762,
      "loss": 5.0792,
      "mean_token_accuracy": 0.1958442896604538,
      "num_tokens": 50607807.0,
      "step": 27435
    },
    {
      "entropy": 5.5011887550354,
      "epoch": 2.305314009661836,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004465397985366536,
      "loss": 5.0758,
      "mean_token_accuracy": 0.19792120307683944,
      "num_tokens": 50615710.0,
      "step": 27440
    },
    {
      "entropy": 5.475981616973877,
      "epoch": 2.3057340894770006,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044652041839259184,
      "loss": 5.2464,
      "mean_token_accuracy": 0.17687831670045853,
      "num_tokens": 50624726.0,
      "step": 27445
    },
    {
      "entropy": 5.585633707046509,
      "epoch": 2.3061541692921654,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044650103521013426,
      "loss": 5.2427,
      "mean_token_accuracy": 0.18421274572610855,
      "num_tokens": 50634424.0,
      "step": 27450
    },
    {
      "entropy": 5.630665683746338,
      "epoch": 2.30657424910733,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004464816489896241,
      "loss": 5.1889,
      "mean_token_accuracy": 0.18351797610521317,
      "num_tokens": 50643809.0,
      "step": 27455
    },
    {
      "entropy": 5.530776691436768,
      "epoch": 2.3069943289224955,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004464622597314048,
      "loss": 5.1165,
      "mean_token_accuracy": 0.19472636729478837,
      "num_tokens": 50652039.0,
      "step": 27460
    },
    {
      "entropy": 5.5179047107696535,
      "epoch": 2.3074144087376602,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004464428674358199,
      "loss": 5.1512,
      "mean_token_accuracy": 0.1864219769835472,
      "num_tokens": 50660755.0,
      "step": 27465
    },
    {
      "entropy": 5.528079462051392,
      "epoch": 2.307834488552825,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004464234721032129,
      "loss": 5.1839,
      "mean_token_accuracy": 0.18442041724920272,
      "num_tokens": 50668867.0,
      "step": 27470
    },
    {
      "entropy": 5.540988731384277,
      "epoch": 2.30825456836799,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004464040737339274,
      "loss": 5.2326,
      "mean_token_accuracy": 0.17864029854536057,
      "num_tokens": 50678506.0,
      "step": 27475
    },
    {
      "entropy": 5.586277675628662,
      "epoch": 2.308674648183155,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004463846723283069,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18509656488895415,
      "num_tokens": 50687037.0,
      "step": 27480
    },
    {
      "entropy": 5.47848072052002,
      "epoch": 2.30909472799832,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004463652678866952,
      "loss": 5.1857,
      "mean_token_accuracy": 0.18890054523944855,
      "num_tokens": 50696438.0,
      "step": 27485
    },
    {
      "entropy": 5.507760286331177,
      "epoch": 2.3095148078134846,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044634586040943604,
      "loss": 5.1365,
      "mean_token_accuracy": 0.18696421831846238,
      "num_tokens": 50705563.0,
      "step": 27490
    },
    {
      "entropy": 5.512574052810669,
      "epoch": 2.3099348876286494,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044632644989687316,
      "loss": 5.1106,
      "mean_token_accuracy": 0.1857517659664154,
      "num_tokens": 50714772.0,
      "step": 27495
    },
    {
      "entropy": 5.546938228607178,
      "epoch": 2.310354967443814,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044630703634935046,
      "loss": 5.1821,
      "mean_token_accuracy": 0.18759274929761888,
      "num_tokens": 50724043.0,
      "step": 27500
    },
    {
      "entropy": 5.597328329086304,
      "epoch": 2.3107750472589794,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044628761976721175,
      "loss": 5.1393,
      "mean_token_accuracy": 0.1854368805885315,
      "num_tokens": 50732840.0,
      "step": 27505
    },
    {
      "entropy": 5.505421304702759,
      "epoch": 2.311195127074144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004462682001508012,
      "loss": 5.1512,
      "mean_token_accuracy": 0.18841395229101182,
      "num_tokens": 50741898.0,
      "step": 27510
    },
    {
      "entropy": 5.437177038192749,
      "epoch": 2.311615206889309,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004462487775004626,
      "loss": 5.1753,
      "mean_token_accuracy": 0.18436780869960784,
      "num_tokens": 50750558.0,
      "step": 27515
    },
    {
      "entropy": 5.596718788146973,
      "epoch": 2.312035286704474,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044622935181654007,
      "loss": 5.2283,
      "mean_token_accuracy": 0.1839207887649536,
      "num_tokens": 50759969.0,
      "step": 27520
    },
    {
      "entropy": 5.5265837669372555,
      "epoch": 2.3124553665196386,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044620992309937774,
      "loss": 5.1659,
      "mean_token_accuracy": 0.18564980179071428,
      "num_tokens": 50768644.0,
      "step": 27525
    },
    {
      "entropy": 5.499352073669433,
      "epoch": 2.312875446334804,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004461904913493198,
      "loss": 5.1772,
      "mean_token_accuracy": 0.1794091984629631,
      "num_tokens": 50778407.0,
      "step": 27530
    },
    {
      "entropy": 5.563594341278076,
      "epoch": 2.3132955261499686,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004461710565667105,
      "loss": 5.2229,
      "mean_token_accuracy": 0.18269515186548232,
      "num_tokens": 50787748.0,
      "step": 27535
    },
    {
      "entropy": 5.536131811141968,
      "epoch": 2.3137156059651334,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000446151618751894,
      "loss": 5.1098,
      "mean_token_accuracy": 0.18848665058612823,
      "num_tokens": 50795809.0,
      "step": 27540
    },
    {
      "entropy": 5.542494440078736,
      "epoch": 2.314135685780298,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004461321779052148,
      "loss": 5.2593,
      "mean_token_accuracy": 0.1777089610695839,
      "num_tokens": 50805370.0,
      "step": 27545
    },
    {
      "entropy": 5.547614431381225,
      "epoch": 2.314555765595463,
      "grad_norm": 1.125,
      "learning_rate": 0.0004461127340270171,
      "loss": 5.2184,
      "mean_token_accuracy": 0.18198493272066116,
      "num_tokens": 50814369.0,
      "step": 27550
    },
    {
      "entropy": 5.6037688732147215,
      "epoch": 2.314975845410628,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004460932871176454,
      "loss": 5.2498,
      "mean_token_accuracy": 0.18600796163082123,
      "num_tokens": 50823537.0,
      "step": 27555
    },
    {
      "entropy": 5.510580110549927,
      "epoch": 2.315395925225793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004460738371774443,
      "loss": 5.1585,
      "mean_token_accuracy": 0.18497988134622573,
      "num_tokens": 50833210.0,
      "step": 27560
    },
    {
      "entropy": 5.540327787399292,
      "epoch": 2.3158160050409577,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004460543842067582,
      "loss": 5.2264,
      "mean_token_accuracy": 0.1790455937385559,
      "num_tokens": 50842085.0,
      "step": 27565
    },
    {
      "entropy": 5.5265645503997805,
      "epoch": 2.3162360848561225,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004460349282059318,
      "loss": 5.2149,
      "mean_token_accuracy": 0.18707195967435836,
      "num_tokens": 50851387.0,
      "step": 27570
    },
    {
      "entropy": 5.573991584777832,
      "epoch": 2.3166561646712873,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004460154691753098,
      "loss": 5.2638,
      "mean_token_accuracy": 0.18863690048456191,
      "num_tokens": 50861093.0,
      "step": 27575
    },
    {
      "entropy": 5.638003158569336,
      "epoch": 2.3170762444864526,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044599600711523667,
      "loss": 5.2649,
      "mean_token_accuracy": 0.18081469237804412,
      "num_tokens": 50870337.0,
      "step": 27580
    },
    {
      "entropy": 5.554800319671631,
      "epoch": 2.3174963243016173,
      "grad_norm": 1.125,
      "learning_rate": 0.0004459765420260574,
      "loss": 5.2779,
      "mean_token_accuracy": 0.18491301387548448,
      "num_tokens": 50879514.0,
      "step": 27585
    },
    {
      "entropy": 5.577145528793335,
      "epoch": 2.317916404116782,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004459570739081167,
      "loss": 5.2744,
      "mean_token_accuracy": 0.17939685583114623,
      "num_tokens": 50889000.0,
      "step": 27590
    },
    {
      "entropy": 5.598965787887574,
      "epoch": 2.318336483931947,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044593760276175954,
      "loss": 5.2266,
      "mean_token_accuracy": 0.18562968373298644,
      "num_tokens": 50898325.0,
      "step": 27595
    },
    {
      "entropy": 5.579704093933105,
      "epoch": 2.318756563747112,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044591812858733073,
      "loss": 5.1491,
      "mean_token_accuracy": 0.18935055285692215,
      "num_tokens": 50907625.0,
      "step": 27600
    },
    {
      "entropy": 5.544568967819214,
      "epoch": 2.319176643562277,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004458986513851752,
      "loss": 5.2357,
      "mean_token_accuracy": 0.18269033432006837,
      "num_tokens": 50917308.0,
      "step": 27605
    },
    {
      "entropy": 5.614783191680909,
      "epoch": 2.3195967233774417,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044587917115563816,
      "loss": 5.3272,
      "mean_token_accuracy": 0.18427644670009613,
      "num_tokens": 50926668.0,
      "step": 27610
    },
    {
      "entropy": 5.591466236114502,
      "epoch": 2.3200168031926065,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044585968789906454,
      "loss": 5.2014,
      "mean_token_accuracy": 0.1814092591404915,
      "num_tokens": 50936070.0,
      "step": 27615
    },
    {
      "entropy": 5.544443511962891,
      "epoch": 2.3204368830077713,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004458402016157996,
      "loss": 5.1906,
      "mean_token_accuracy": 0.1859136253595352,
      "num_tokens": 50944656.0,
      "step": 27620
    },
    {
      "entropy": 5.629294443130493,
      "epoch": 2.3208569628229365,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044582071230618835,
      "loss": 5.2587,
      "mean_token_accuracy": 0.18270053267478942,
      "num_tokens": 50953826.0,
      "step": 27625
    },
    {
      "entropy": 5.58927206993103,
      "epoch": 2.3212770426381013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004458012199705762,
      "loss": 5.1262,
      "mean_token_accuracy": 0.19283167123794556,
      "num_tokens": 50962129.0,
      "step": 27630
    },
    {
      "entropy": 5.469469451904297,
      "epoch": 2.321697122453266,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004457817246093084,
      "loss": 5.1107,
      "mean_token_accuracy": 0.18621859401464463,
      "num_tokens": 50971836.0,
      "step": 27635
    },
    {
      "entropy": 5.457266139984131,
      "epoch": 2.322117202268431,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044576222622273026,
      "loss": 5.1715,
      "mean_token_accuracy": 0.1864071175456047,
      "num_tokens": 50981448.0,
      "step": 27640
    },
    {
      "entropy": 5.676044034957886,
      "epoch": 2.322537282083596,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004457427248111872,
      "loss": 5.3862,
      "mean_token_accuracy": 0.17110629975795746,
      "num_tokens": 50992475.0,
      "step": 27645
    },
    {
      "entropy": 5.651588296890258,
      "epoch": 2.322957361898761,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044572322037502474,
      "loss": 5.2964,
      "mean_token_accuracy": 0.1823749765753746,
      "num_tokens": 51000965.0,
      "step": 27650
    },
    {
      "entropy": 5.522226142883301,
      "epoch": 2.3233774417139257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004457037129145883,
      "loss": 5.1958,
      "mean_token_accuracy": 0.1886395961046219,
      "num_tokens": 51010374.0,
      "step": 27655
    },
    {
      "entropy": 5.481893301010132,
      "epoch": 2.3237975215290905,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004456842024302235,
      "loss": 5.1248,
      "mean_token_accuracy": 0.1834106206893921,
      "num_tokens": 51018714.0,
      "step": 27660
    },
    {
      "entropy": 5.527593278884888,
      "epoch": 2.3242176013442553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004456646889222759,
      "loss": 5.208,
      "mean_token_accuracy": 0.1899716168642044,
      "num_tokens": 51028097.0,
      "step": 27665
    },
    {
      "entropy": 5.5290687084198,
      "epoch": 2.3246376811594205,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004456451723910913,
      "loss": 5.1464,
      "mean_token_accuracy": 0.19114057570695878,
      "num_tokens": 51036270.0,
      "step": 27670
    },
    {
      "entropy": 5.6204911231994625,
      "epoch": 2.3250577609745853,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004456256528370153,
      "loss": 5.2551,
      "mean_token_accuracy": 0.17827585488557815,
      "num_tokens": 51046235.0,
      "step": 27675
    },
    {
      "entropy": 5.535403871536255,
      "epoch": 2.32547784078975,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044560613026039376,
      "loss": 5.1675,
      "mean_token_accuracy": 0.18923941999673843,
      "num_tokens": 51054996.0,
      "step": 27680
    },
    {
      "entropy": 5.419713878631592,
      "epoch": 2.325897920604915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044558660466157237,
      "loss": 5.0466,
      "mean_token_accuracy": 0.18850823789834975,
      "num_tokens": 51063248.0,
      "step": 27685
    },
    {
      "entropy": 5.610247230529785,
      "epoch": 2.3263180004200796,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044556707604089717,
      "loss": 5.3208,
      "mean_token_accuracy": 0.1735820487141609,
      "num_tokens": 51073242.0,
      "step": 27690
    },
    {
      "entropy": 5.537463760375976,
      "epoch": 2.326738080235245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000445547544398714,
      "loss": 5.1872,
      "mean_token_accuracy": 0.18799320608377457,
      "num_tokens": 51082731.0,
      "step": 27695
    },
    {
      "entropy": 5.5952561378479,
      "epoch": 2.3271581600504097,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000445528009735369,
      "loss": 5.3028,
      "mean_token_accuracy": 0.17866193652153015,
      "num_tokens": 51091137.0,
      "step": 27700
    },
    {
      "entropy": 5.570521211624145,
      "epoch": 2.3275782398655744,
      "grad_norm": 1.171875,
      "learning_rate": 0.000445508472051208,
      "loss": 5.2188,
      "mean_token_accuracy": 0.18434504717588424,
      "num_tokens": 51099525.0,
      "step": 27705
    },
    {
      "entropy": 5.5861289501190186,
      "epoch": 2.3279983196807392,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044548893134657735,
      "loss": 5.2628,
      "mean_token_accuracy": 0.18061445355415345,
      "num_tokens": 51108858.0,
      "step": 27710
    },
    {
      "entropy": 5.526335763931274,
      "epoch": 2.328418399495904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000445469387621823,
      "loss": 5.1416,
      "mean_token_accuracy": 0.19257837533950806,
      "num_tokens": 51119009.0,
      "step": 27715
    },
    {
      "entropy": 5.512407398223877,
      "epoch": 2.3288384793110692,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044544984087729124,
      "loss": 5.1988,
      "mean_token_accuracy": 0.18439842760562897,
      "num_tokens": 51128601.0,
      "step": 27720
    },
    {
      "entropy": 5.564397192001342,
      "epoch": 2.329258559126234,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044543029111332834,
      "loss": 5.2312,
      "mean_token_accuracy": 0.1785915657877922,
      "num_tokens": 51138053.0,
      "step": 27725
    },
    {
      "entropy": 5.592361164093018,
      "epoch": 2.329678638941399,
      "grad_norm": 1.125,
      "learning_rate": 0.0004454107383302805,
      "loss": 5.2021,
      "mean_token_accuracy": 0.1806567892432213,
      "num_tokens": 51147178.0,
      "step": 27730
    },
    {
      "entropy": 5.542826128005982,
      "epoch": 2.3300987187565636,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044539118252849433,
      "loss": 5.1782,
      "mean_token_accuracy": 0.18411092311143876,
      "num_tokens": 51156320.0,
      "step": 27735
    },
    {
      "entropy": 5.4533734798431395,
      "epoch": 2.3305187985717284,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004453716237083159,
      "loss": 5.1419,
      "mean_token_accuracy": 0.18645168095827103,
      "num_tokens": 51164997.0,
      "step": 27740
    },
    {
      "entropy": 5.53551607131958,
      "epoch": 2.3309388783868936,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004453520618700921,
      "loss": 5.2554,
      "mean_token_accuracy": 0.1868303656578064,
      "num_tokens": 51174102.0,
      "step": 27745
    },
    {
      "entropy": 5.601465129852295,
      "epoch": 2.3313589582020584,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00044533249701416913,
      "loss": 5.2293,
      "mean_token_accuracy": 0.17740438431501387,
      "num_tokens": 51183990.0,
      "step": 27750
    },
    {
      "entropy": 5.531319999694825,
      "epoch": 2.331779038017223,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004453129291408936,
      "loss": 5.1852,
      "mean_token_accuracy": 0.19011679589748381,
      "num_tokens": 51192675.0,
      "step": 27755
    },
    {
      "entropy": 5.500803518295288,
      "epoch": 2.332199117832388,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044529335825061237,
      "loss": 5.1327,
      "mean_token_accuracy": 0.1852489545941353,
      "num_tokens": 51202427.0,
      "step": 27760
    },
    {
      "entropy": 5.483710765838623,
      "epoch": 2.332619197647553,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044527378434367187,
      "loss": 5.1871,
      "mean_token_accuracy": 0.18018860518932342,
      "num_tokens": 51211668.0,
      "step": 27765
    },
    {
      "entropy": 5.552032089233398,
      "epoch": 2.333039277462718,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004452542074204191,
      "loss": 5.1997,
      "mean_token_accuracy": 0.18195951879024505,
      "num_tokens": 51220169.0,
      "step": 27770
    },
    {
      "entropy": 5.5805220127105715,
      "epoch": 2.333459357277883,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044523462748120065,
      "loss": 5.2248,
      "mean_token_accuracy": 0.1886780872941017,
      "num_tokens": 51229333.0,
      "step": 27775
    },
    {
      "entropy": 5.523029327392578,
      "epoch": 2.3338794370930476,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044521504452636334,
      "loss": 5.1857,
      "mean_token_accuracy": 0.18216458857059478,
      "num_tokens": 51238484.0,
      "step": 27780
    },
    {
      "entropy": 5.478063869476318,
      "epoch": 2.334299516908213,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004451954585562543,
      "loss": 5.0583,
      "mean_token_accuracy": 0.1948428764939308,
      "num_tokens": 51247427.0,
      "step": 27785
    },
    {
      "entropy": 5.52767243385315,
      "epoch": 2.3347195967233776,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004451758695712203,
      "loss": 5.2383,
      "mean_token_accuracy": 0.19224076569080353,
      "num_tokens": 51256259.0,
      "step": 27790
    },
    {
      "entropy": 5.550039148330688,
      "epoch": 2.3351396765385424,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004451562775716083,
      "loss": 5.1257,
      "mean_token_accuracy": 0.18602681308984756,
      "num_tokens": 51265072.0,
      "step": 27795
    },
    {
      "entropy": 5.636503648757935,
      "epoch": 2.335559756353707,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004451366825577656,
      "loss": 5.259,
      "mean_token_accuracy": 0.18507232069969176,
      "num_tokens": 51275383.0,
      "step": 27800
    },
    {
      "entropy": 5.468134784698487,
      "epoch": 2.335979836168872,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044511708453003914,
      "loss": 5.1321,
      "mean_token_accuracy": 0.20254699438810347,
      "num_tokens": 51284025.0,
      "step": 27805
    },
    {
      "entropy": 5.474282312393188,
      "epoch": 2.336399915984037,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044509748348877607,
      "loss": 5.2439,
      "mean_token_accuracy": 0.18325816392898558,
      "num_tokens": 51293794.0,
      "step": 27810
    },
    {
      "entropy": 5.635352563858032,
      "epoch": 2.336819995799202,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004450778794343237,
      "loss": 5.1989,
      "mean_token_accuracy": 0.18648047745227814,
      "num_tokens": 51303358.0,
      "step": 27815
    },
    {
      "entropy": 5.651332426071167,
      "epoch": 2.3372400756143668,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044505827236702934,
      "loss": 5.2198,
      "mean_token_accuracy": 0.18259362429380416,
      "num_tokens": 51312270.0,
      "step": 27820
    },
    {
      "entropy": 5.667574548721314,
      "epoch": 2.3376601554295315,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044503866228724016,
      "loss": 5.2563,
      "mean_token_accuracy": 0.17590213418006898,
      "num_tokens": 51322220.0,
      "step": 27825
    },
    {
      "entropy": 5.561887407302857,
      "epoch": 2.3380802352446963,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044501904919530365,
      "loss": 5.1981,
      "mean_token_accuracy": 0.17819261848926543,
      "num_tokens": 51332240.0,
      "step": 27830
    },
    {
      "entropy": 5.531216716766357,
      "epoch": 2.3385003150598616,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004449994330915673,
      "loss": 5.2605,
      "mean_token_accuracy": 0.17659686058759688,
      "num_tokens": 51340939.0,
      "step": 27835
    },
    {
      "entropy": 5.572314023971558,
      "epoch": 2.3389203948750263,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004449798139763785,
      "loss": 5.1381,
      "mean_token_accuracy": 0.18670845925807952,
      "num_tokens": 51349270.0,
      "step": 27840
    },
    {
      "entropy": 5.529236888885498,
      "epoch": 2.339340474690191,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044496019185008475,
      "loss": 5.0881,
      "mean_token_accuracy": 0.19047731012105942,
      "num_tokens": 51358270.0,
      "step": 27845
    },
    {
      "entropy": 5.526680278778076,
      "epoch": 2.339760554505356,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004449405667130337,
      "loss": 5.2056,
      "mean_token_accuracy": 0.18285173028707505,
      "num_tokens": 51367722.0,
      "step": 27850
    },
    {
      "entropy": 5.496097278594971,
      "epoch": 2.3401806343205207,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044492093856557315,
      "loss": 5.2267,
      "mean_token_accuracy": 0.18885283619165422,
      "num_tokens": 51376920.0,
      "step": 27855
    },
    {
      "entropy": 5.710688638687134,
      "epoch": 2.340600714135686,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044490130740805055,
      "loss": 5.4057,
      "mean_token_accuracy": 0.17034368216991425,
      "num_tokens": 51387175.0,
      "step": 27860
    },
    {
      "entropy": 5.593728733062744,
      "epoch": 2.3410207939508507,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004448816732408138,
      "loss": 5.278,
      "mean_token_accuracy": 0.1758460983633995,
      "num_tokens": 51396552.0,
      "step": 27865
    },
    {
      "entropy": 5.525693607330322,
      "epoch": 2.3414408737660155,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004448620360642108,
      "loss": 5.2119,
      "mean_token_accuracy": 0.179982790350914,
      "num_tokens": 51405587.0,
      "step": 27870
    },
    {
      "entropy": 5.613410139083863,
      "epoch": 2.3418609535811803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044484239587858917,
      "loss": 5.2665,
      "mean_token_accuracy": 0.17784098386764527,
      "num_tokens": 51413905.0,
      "step": 27875
    },
    {
      "entropy": 5.571114826202392,
      "epoch": 2.342281033396345,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000444822752684297,
      "loss": 5.1184,
      "mean_token_accuracy": 0.1859305128455162,
      "num_tokens": 51422503.0,
      "step": 27880
    },
    {
      "entropy": 5.457857131958008,
      "epoch": 2.3427011132115103,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004448031064816822,
      "loss": 5.0889,
      "mean_token_accuracy": 0.19235942363739014,
      "num_tokens": 51431577.0,
      "step": 27885
    },
    {
      "entropy": 5.627953243255615,
      "epoch": 2.343121193026675,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044478345727109276,
      "loss": 5.2755,
      "mean_token_accuracy": 0.18066827803850175,
      "num_tokens": 51440971.0,
      "step": 27890
    },
    {
      "entropy": 5.530179071426391,
      "epoch": 2.34354127284184,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004447638050528769,
      "loss": 5.1475,
      "mean_token_accuracy": 0.18772403448820113,
      "num_tokens": 51450590.0,
      "step": 27895
    },
    {
      "entropy": 5.567606735229492,
      "epoch": 2.3439613526570047,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004447441498273826,
      "loss": 5.2216,
      "mean_token_accuracy": 0.17999920696020127,
      "num_tokens": 51459662.0,
      "step": 27900
    },
    {
      "entropy": 5.585417032241821,
      "epoch": 2.34438143247217,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044472449159495807,
      "loss": 5.2243,
      "mean_token_accuracy": 0.19024741351604463,
      "num_tokens": 51469347.0,
      "step": 27905
    },
    {
      "entropy": 5.510441875457763,
      "epoch": 2.3448015122873347,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044470483035595156,
      "loss": 5.2198,
      "mean_token_accuracy": 0.1866058275103569,
      "num_tokens": 51478752.0,
      "step": 27910
    },
    {
      "entropy": 5.427440500259399,
      "epoch": 2.3452215921024995,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004446851661107114,
      "loss": 5.0078,
      "mean_token_accuracy": 0.1965171828866005,
      "num_tokens": 51487558.0,
      "step": 27915
    },
    {
      "entropy": 5.505718755722046,
      "epoch": 2.3456416719176643,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004446654988595859,
      "loss": 5.2345,
      "mean_token_accuracy": 0.1865793913602829,
      "num_tokens": 51495824.0,
      "step": 27920
    },
    {
      "entropy": 5.629418611526489,
      "epoch": 2.346061751732829,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044464582860292344,
      "loss": 5.1898,
      "mean_token_accuracy": 0.18914976716041565,
      "num_tokens": 51505199.0,
      "step": 27925
    },
    {
      "entropy": 5.590628528594971,
      "epoch": 2.3464818315479943,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044462615534107247,
      "loss": 5.1815,
      "mean_token_accuracy": 0.18163477778434753,
      "num_tokens": 51514343.0,
      "step": 27930
    },
    {
      "entropy": 5.560458660125732,
      "epoch": 2.346901911363159,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004446064790743815,
      "loss": 5.2242,
      "mean_token_accuracy": 0.18358898013830185,
      "num_tokens": 51522942.0,
      "step": 27935
    },
    {
      "entropy": 5.514103031158447,
      "epoch": 2.347321991178324,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044458679980319916,
      "loss": 5.1733,
      "mean_token_accuracy": 0.18584103584289552,
      "num_tokens": 51532864.0,
      "step": 27940
    },
    {
      "entropy": 5.643073987960816,
      "epoch": 2.3477420709934886,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044456711752787387,
      "loss": 5.2554,
      "mean_token_accuracy": 0.18763115555047988,
      "num_tokens": 51542862.0,
      "step": 27945
    },
    {
      "entropy": 5.646440601348877,
      "epoch": 2.348162150808654,
      "grad_norm": 1.125,
      "learning_rate": 0.00044454743224875453,
      "loss": 5.2638,
      "mean_token_accuracy": 0.176938197016716,
      "num_tokens": 51552502.0,
      "step": 27950
    },
    {
      "entropy": 5.547908544540405,
      "epoch": 2.3485822306238187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004445277439661896,
      "loss": 5.2352,
      "mean_token_accuracy": 0.182615028321743,
      "num_tokens": 51561535.0,
      "step": 27955
    },
    {
      "entropy": 5.5877673625946045,
      "epoch": 2.3490023104389834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000444508052680528,
      "loss": 5.138,
      "mean_token_accuracy": 0.1898419290781021,
      "num_tokens": 51570413.0,
      "step": 27960
    },
    {
      "entropy": 5.517753219604492,
      "epoch": 2.3494223902541482,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044448835839211865,
      "loss": 5.1755,
      "mean_token_accuracy": 0.18486383855342864,
      "num_tokens": 51579370.0,
      "step": 27965
    },
    {
      "entropy": 5.512256669998169,
      "epoch": 2.349842470069313,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044446866110131025,
      "loss": 5.19,
      "mean_token_accuracy": 0.18771204650402068,
      "num_tokens": 51588944.0,
      "step": 27970
    },
    {
      "entropy": 5.734362459182739,
      "epoch": 2.3502625498844782,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004444489608084517,
      "loss": 5.3645,
      "mean_token_accuracy": 0.1748827025294304,
      "num_tokens": 51599061.0,
      "step": 27975
    },
    {
      "entropy": 5.582852411270141,
      "epoch": 2.350682629699643,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004444292575138922,
      "loss": 5.2481,
      "mean_token_accuracy": 0.18211822658777238,
      "num_tokens": 51608742.0,
      "step": 27980
    },
    {
      "entropy": 5.5694053173065186,
      "epoch": 2.351102709514808,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004444095512179805,
      "loss": 5.2696,
      "mean_token_accuracy": 0.1856290355324745,
      "num_tokens": 51618351.0,
      "step": 27985
    },
    {
      "entropy": 5.605921506881714,
      "epoch": 2.3515227893299726,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004443898419210659,
      "loss": 5.2675,
      "mean_token_accuracy": 0.18403708338737487,
      "num_tokens": 51626424.0,
      "step": 27990
    },
    {
      "entropy": 5.577618169784546,
      "epoch": 2.3519428691451374,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004443701296234975,
      "loss": 5.2344,
      "mean_token_accuracy": 0.18088037073612212,
      "num_tokens": 51635852.0,
      "step": 27995
    },
    {
      "entropy": 5.5869299411773685,
      "epoch": 2.3523629489603026,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044435041432562447,
      "loss": 5.1996,
      "mean_token_accuracy": 0.1851030707359314,
      "num_tokens": 51645130.0,
      "step": 28000
    },
    {
      "entropy": 5.58957085609436,
      "epoch": 2.3527830287754674,
      "grad_norm": 1.125,
      "learning_rate": 0.000444330696027796,
      "loss": 5.3438,
      "mean_token_accuracy": 0.18119870871305466,
      "num_tokens": 51655566.0,
      "step": 28005
    },
    {
      "entropy": 5.531573820114136,
      "epoch": 2.353203108590632,
      "grad_norm": 1.125,
      "learning_rate": 0.0004443109747303615,
      "loss": 5.0458,
      "mean_token_accuracy": 0.19572264403104783,
      "num_tokens": 51664597.0,
      "step": 28010
    },
    {
      "entropy": 5.528761005401611,
      "epoch": 2.353623188405797,
      "grad_norm": 1.125,
      "learning_rate": 0.00044429125043367014,
      "loss": 5.2651,
      "mean_token_accuracy": 0.18029631525278092,
      "num_tokens": 51673829.0,
      "step": 28015
    },
    {
      "entropy": 5.587442350387573,
      "epoch": 2.3540432682209618,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004442715231380716,
      "loss": 5.1314,
      "mean_token_accuracy": 0.19258745461702348,
      "num_tokens": 51682382.0,
      "step": 28020
    },
    {
      "entropy": 5.492032814025879,
      "epoch": 2.354463348036127,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004442517928439151,
      "loss": 5.164,
      "mean_token_accuracy": 0.19057848304510117,
      "num_tokens": 51691979.0,
      "step": 28025
    },
    {
      "entropy": 5.491596603393555,
      "epoch": 2.354883427851292,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004442320595515503,
      "loss": 5.0929,
      "mean_token_accuracy": 0.18550011217594148,
      "num_tokens": 51700058.0,
      "step": 28030
    },
    {
      "entropy": 5.514552354812622,
      "epoch": 2.3553035076664566,
      "grad_norm": 1.125,
      "learning_rate": 0.00044421232326132665,
      "loss": 5.208,
      "mean_token_accuracy": 0.17652394473552704,
      "num_tokens": 51708931.0,
      "step": 28035
    },
    {
      "entropy": 5.599909448623658,
      "epoch": 2.3557235874816214,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004441925839735939,
      "loss": 5.2926,
      "mean_token_accuracy": 0.17692455500364304,
      "num_tokens": 51717616.0,
      "step": 28040
    },
    {
      "entropy": 5.628249597549439,
      "epoch": 2.356143667296786,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004441728416887016,
      "loss": 5.2144,
      "mean_token_accuracy": 0.17858208417892457,
      "num_tokens": 51727074.0,
      "step": 28045
    },
    {
      "entropy": 5.548230504989624,
      "epoch": 2.3565637471119514,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044415309640699955,
      "loss": 5.1921,
      "mean_token_accuracy": 0.1901193603873253,
      "num_tokens": 51735233.0,
      "step": 28050
    },
    {
      "entropy": 5.491315698623657,
      "epoch": 2.356983826927116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004441333481288374,
      "loss": 5.1117,
      "mean_token_accuracy": 0.19141121208667755,
      "num_tokens": 51743958.0,
      "step": 28055
    },
    {
      "entropy": 5.478999137878418,
      "epoch": 2.357403906742281,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044411359685456517,
      "loss": 5.1655,
      "mean_token_accuracy": 0.18701365888118743,
      "num_tokens": 51752755.0,
      "step": 28060
    },
    {
      "entropy": 5.519311237335205,
      "epoch": 2.3578239865574457,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044409384258453264,
      "loss": 5.2141,
      "mean_token_accuracy": 0.1792498230934143,
      "num_tokens": 51762184.0,
      "step": 28065
    },
    {
      "entropy": 5.509305572509765,
      "epoch": 2.358244066372611,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004440740853190897,
      "loss": 5.0874,
      "mean_token_accuracy": 0.18868587911128998,
      "num_tokens": 51771919.0,
      "step": 28070
    },
    {
      "entropy": 5.557280969619751,
      "epoch": 2.3586641461877758,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004440543250585864,
      "loss": 5.1635,
      "mean_token_accuracy": 0.1906384989619255,
      "num_tokens": 51781020.0,
      "step": 28075
    },
    {
      "entropy": 5.614663982391358,
      "epoch": 2.3590842260029405,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044403456180337286,
      "loss": 5.2715,
      "mean_token_accuracy": 0.1789061024785042,
      "num_tokens": 51790169.0,
      "step": 28080
    },
    {
      "entropy": 5.528173017501831,
      "epoch": 2.3595043058181053,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044401479555379917,
      "loss": 5.1602,
      "mean_token_accuracy": 0.1896404966711998,
      "num_tokens": 51799291.0,
      "step": 28085
    },
    {
      "entropy": 5.640162420272827,
      "epoch": 2.3599243856332706,
      "grad_norm": 1.125,
      "learning_rate": 0.0004439950263102152,
      "loss": 5.3056,
      "mean_token_accuracy": 0.18035952001810074,
      "num_tokens": 51808821.0,
      "step": 28090
    },
    {
      "entropy": 5.448464250564575,
      "epoch": 2.3603444654484353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044397525407297153,
      "loss": 5.0824,
      "mean_token_accuracy": 0.19840161502361298,
      "num_tokens": 51817691.0,
      "step": 28095
    },
    {
      "entropy": 5.577154350280762,
      "epoch": 2.3607645452636,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044395547884241814,
      "loss": 5.2248,
      "mean_token_accuracy": 0.17973207384347917,
      "num_tokens": 51827164.0,
      "step": 28100
    },
    {
      "entropy": 5.594835376739502,
      "epoch": 2.361184625078765,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004439357006189055,
      "loss": 5.1972,
      "mean_token_accuracy": 0.1861821323633194,
      "num_tokens": 51836841.0,
      "step": 28105
    },
    {
      "entropy": 5.600129747390747,
      "epoch": 2.3616047048939297,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004439159194027839,
      "loss": 5.2125,
      "mean_token_accuracy": 0.17833598107099533,
      "num_tokens": 51846589.0,
      "step": 28110
    },
    {
      "entropy": 5.530082845687867,
      "epoch": 2.362024784709095,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044389613519440373,
      "loss": 5.1663,
      "mean_token_accuracy": 0.1887248769402504,
      "num_tokens": 51855099.0,
      "step": 28115
    },
    {
      "entropy": 5.493421936035157,
      "epoch": 2.3624448645242597,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044387634799411557,
      "loss": 5.1137,
      "mean_token_accuracy": 0.1931380182504654,
      "num_tokens": 51864377.0,
      "step": 28120
    },
    {
      "entropy": 5.547094011306763,
      "epoch": 2.3628649443394245,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004438565578022698,
      "loss": 5.2578,
      "mean_token_accuracy": 0.18183549195528032,
      "num_tokens": 51874235.0,
      "step": 28125
    },
    {
      "entropy": 5.62247953414917,
      "epoch": 2.3632850241545893,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044383676461921703,
      "loss": 5.3156,
      "mean_token_accuracy": 0.1722257435321808,
      "num_tokens": 51884332.0,
      "step": 28130
    },
    {
      "entropy": 5.571455240249634,
      "epoch": 2.363705103969754,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000443816968445308,
      "loss": 5.176,
      "mean_token_accuracy": 0.18499821275472642,
      "num_tokens": 51893880.0,
      "step": 28135
    },
    {
      "entropy": 5.558259201049805,
      "epoch": 2.3641251837849193,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004437971692808932,
      "loss": 5.2605,
      "mean_token_accuracy": 0.1798187106847763,
      "num_tokens": 51902568.0,
      "step": 28140
    },
    {
      "entropy": 5.50118317604065,
      "epoch": 2.364545263600084,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044377736712632355,
      "loss": 5.1499,
      "mean_token_accuracy": 0.18638604879379272,
      "num_tokens": 51911919.0,
      "step": 28145
    },
    {
      "entropy": 5.56800422668457,
      "epoch": 2.364965343415249,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004437575619819497,
      "loss": 5.198,
      "mean_token_accuracy": 0.1886693000793457,
      "num_tokens": 51921348.0,
      "step": 28150
    },
    {
      "entropy": 5.638720846176147,
      "epoch": 2.3653854232304137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004437377538481225,
      "loss": 5.3128,
      "mean_token_accuracy": 0.18138468116521836,
      "num_tokens": 51930683.0,
      "step": 28155
    },
    {
      "entropy": 5.606554555892944,
      "epoch": 2.3658055030455785,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000443717942725193,
      "loss": 5.2557,
      "mean_token_accuracy": 0.18436689674854279,
      "num_tokens": 51939387.0,
      "step": 28160
    },
    {
      "entropy": 5.5551361560821535,
      "epoch": 2.3662255828607437,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004436981286135119,
      "loss": 5.2477,
      "mean_token_accuracy": 0.18488434106111526,
      "num_tokens": 51949385.0,
      "step": 28165
    },
    {
      "entropy": 5.516095447540283,
      "epoch": 2.3666456626759085,
      "grad_norm": 1.0,
      "learning_rate": 0.0004436783115134304,
      "loss": 5.1987,
      "mean_token_accuracy": 0.19431472569704056,
      "num_tokens": 51958710.0,
      "step": 28170
    },
    {
      "entropy": 5.505274772644043,
      "epoch": 2.3670657424910733,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004436584914252995,
      "loss": 5.0806,
      "mean_token_accuracy": 0.19409053176641464,
      "num_tokens": 51967392.0,
      "step": 28175
    },
    {
      "entropy": 5.48991208076477,
      "epoch": 2.367485822306238,
      "grad_norm": 1.125,
      "learning_rate": 0.00044363866834947017,
      "loss": 5.1797,
      "mean_token_accuracy": 0.1850089907646179,
      "num_tokens": 51976504.0,
      "step": 28180
    },
    {
      "entropy": 5.620902061462402,
      "epoch": 2.367905902121403,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044361884228629367,
      "loss": 5.2396,
      "mean_token_accuracy": 0.18396025747060776,
      "num_tokens": 51985692.0,
      "step": 28185
    },
    {
      "entropy": 5.503563356399536,
      "epoch": 2.368325981936568,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004435990132361212,
      "loss": 5.1165,
      "mean_token_accuracy": 0.1855497658252716,
      "num_tokens": 51994905.0,
      "step": 28190
    },
    {
      "entropy": 5.567619371414184,
      "epoch": 2.368746061751733,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044357918119930414,
      "loss": 5.2325,
      "mean_token_accuracy": 0.18861991465091704,
      "num_tokens": 52003610.0,
      "step": 28195
    },
    {
      "entropy": 5.432132768630981,
      "epoch": 2.3691661415668976,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004435593461761936,
      "loss": 5.0494,
      "mean_token_accuracy": 0.19788996428251265,
      "num_tokens": 52012932.0,
      "step": 28200
    },
    {
      "entropy": 5.569317436218261,
      "epoch": 2.3695862213820624,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004435395081671411,
      "loss": 5.2447,
      "mean_token_accuracy": 0.18033053129911422,
      "num_tokens": 52022343.0,
      "step": 28205
    },
    {
      "entropy": 5.52894082069397,
      "epoch": 2.3700063011972277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044351966717249793,
      "loss": 5.1432,
      "mean_token_accuracy": 0.1869301125407219,
      "num_tokens": 52031382.0,
      "step": 28210
    },
    {
      "entropy": 5.474100494384766,
      "epoch": 2.3704263810123924,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004434998231926156,
      "loss": 5.0823,
      "mean_token_accuracy": 0.19282765537500382,
      "num_tokens": 52039463.0,
      "step": 28215
    },
    {
      "entropy": 5.484936809539795,
      "epoch": 2.3708464608275572,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044347997622784573,
      "loss": 5.1567,
      "mean_token_accuracy": 0.1892690032720566,
      "num_tokens": 52049568.0,
      "step": 28220
    },
    {
      "entropy": 5.535483884811401,
      "epoch": 2.371266540642722,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044346012627853983,
      "loss": 5.2337,
      "mean_token_accuracy": 0.19100067019462585,
      "num_tokens": 52059968.0,
      "step": 28225
    },
    {
      "entropy": 5.429222536087036,
      "epoch": 2.371686620457887,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004434402733450495,
      "loss": 5.1707,
      "mean_token_accuracy": 0.18609771877527237,
      "num_tokens": 52068430.0,
      "step": 28230
    },
    {
      "entropy": 5.499593925476074,
      "epoch": 2.372106700273052,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044342041742772653,
      "loss": 5.1336,
      "mean_token_accuracy": 0.19164807796478273,
      "num_tokens": 52076749.0,
      "step": 28235
    },
    {
      "entropy": 5.487621641159057,
      "epoch": 2.372526780088217,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004434005585269225,
      "loss": 5.1084,
      "mean_token_accuracy": 0.1942272663116455,
      "num_tokens": 52086205.0,
      "step": 28240
    },
    {
      "entropy": 5.535723352432251,
      "epoch": 2.3729468599033816,
      "grad_norm": 1.25,
      "learning_rate": 0.00044338069664298933,
      "loss": 5.156,
      "mean_token_accuracy": 0.19017360955476761,
      "num_tokens": 52096101.0,
      "step": 28245
    },
    {
      "entropy": 5.585770750045777,
      "epoch": 2.3733669397185464,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004433608317762789,
      "loss": 5.3703,
      "mean_token_accuracy": 0.17903836965560913,
      "num_tokens": 52106046.0,
      "step": 28250
    },
    {
      "entropy": 5.505969381332397,
      "epoch": 2.3737870195337116,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044334096392714293,
      "loss": 5.2302,
      "mean_token_accuracy": 0.18316590487957002,
      "num_tokens": 52115599.0,
      "step": 28255
    },
    {
      "entropy": 5.602386093139648,
      "epoch": 2.3742070993488764,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004433210930959335,
      "loss": 5.263,
      "mean_token_accuracy": 0.17689010202884675,
      "num_tokens": 52124746.0,
      "step": 28260
    },
    {
      "entropy": 5.572861909866333,
      "epoch": 2.374627179164041,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004433012192830026,
      "loss": 5.2316,
      "mean_token_accuracy": 0.18358784914016724,
      "num_tokens": 52133995.0,
      "step": 28265
    },
    {
      "entropy": 5.5213298320770265,
      "epoch": 2.375047258979206,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044328134248870223,
      "loss": 5.1509,
      "mean_token_accuracy": 0.188263438642025,
      "num_tokens": 52142897.0,
      "step": 28270
    },
    {
      "entropy": 5.495863151550293,
      "epoch": 2.3754673387943708,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044326146271338456,
      "loss": 5.1439,
      "mean_token_accuracy": 0.18626012802124023,
      "num_tokens": 52151880.0,
      "step": 28275
    },
    {
      "entropy": 5.462082290649414,
      "epoch": 2.375887418609536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004432415799574017,
      "loss": 5.1403,
      "mean_token_accuracy": 0.18841230273246765,
      "num_tokens": 52161143.0,
      "step": 28280
    },
    {
      "entropy": 5.592006969451904,
      "epoch": 2.376307498424701,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004432216942211059,
      "loss": 5.2197,
      "mean_token_accuracy": 0.1787035197019577,
      "num_tokens": 52170906.0,
      "step": 28285
    },
    {
      "entropy": 5.549290418624878,
      "epoch": 2.3767275782398656,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044320180550484935,
      "loss": 5.1674,
      "mean_token_accuracy": 0.1900693655014038,
      "num_tokens": 52179698.0,
      "step": 28290
    },
    {
      "entropy": 5.49228982925415,
      "epoch": 2.3771476580550304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004431819138089845,
      "loss": 5.1346,
      "mean_token_accuracy": 0.19055932611227036,
      "num_tokens": 52187664.0,
      "step": 28295
    },
    {
      "entropy": 5.616365575790406,
      "epoch": 2.377567737870195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004431620191338636,
      "loss": 5.3908,
      "mean_token_accuracy": 0.17513343393802644,
      "num_tokens": 52197604.0,
      "step": 28300
    },
    {
      "entropy": 5.531540060043335,
      "epoch": 2.3779878176853604,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044314212147983915,
      "loss": 5.2406,
      "mean_token_accuracy": 0.18504862636327743,
      "num_tokens": 52206814.0,
      "step": 28305
    },
    {
      "entropy": 5.5253888130187985,
      "epoch": 2.378407897500525,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044312222084726355,
      "loss": 5.2337,
      "mean_token_accuracy": 0.18544088155031205,
      "num_tokens": 52216298.0,
      "step": 28310
    },
    {
      "entropy": 5.5641132354736325,
      "epoch": 2.37882797731569,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004431023172364894,
      "loss": 5.1826,
      "mean_token_accuracy": 0.18768933415412903,
      "num_tokens": 52225433.0,
      "step": 28315
    },
    {
      "entropy": 5.501485157012939,
      "epoch": 2.3792480571308547,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044308241064786925,
      "loss": 5.161,
      "mean_token_accuracy": 0.18972741365432738,
      "num_tokens": 52233519.0,
      "step": 28320
    },
    {
      "entropy": 5.4559102058410645,
      "epoch": 2.3796681369460195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044306250108175577,
      "loss": 5.2181,
      "mean_token_accuracy": 0.18226444274187087,
      "num_tokens": 52242188.0,
      "step": 28325
    },
    {
      "entropy": 5.576568222045898,
      "epoch": 2.3800882167611848,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044304258853850167,
      "loss": 5.1408,
      "mean_token_accuracy": 0.19228406697511674,
      "num_tokens": 52251626.0,
      "step": 28330
    },
    {
      "entropy": 5.578274726867676,
      "epoch": 2.3805082965763495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004430226730184596,
      "loss": 5.2166,
      "mean_token_accuracy": 0.18462162911891938,
      "num_tokens": 52261008.0,
      "step": 28335
    },
    {
      "entropy": 5.562291193008423,
      "epoch": 2.3809283763915143,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044300275452198234,
      "loss": 5.1704,
      "mean_token_accuracy": 0.18668585568666457,
      "num_tokens": 52270255.0,
      "step": 28340
    },
    {
      "entropy": 5.484687232971192,
      "epoch": 2.381348456206679,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00044298283304942287,
      "loss": 5.1759,
      "mean_token_accuracy": 0.18994148671627045,
      "num_tokens": 52279550.0,
      "step": 28345
    },
    {
      "entropy": 5.499300336837768,
      "epoch": 2.381768536021844,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000442962908601134,
      "loss": 5.2,
      "mean_token_accuracy": 0.17576824873685837,
      "num_tokens": 52289458.0,
      "step": 28350
    },
    {
      "entropy": 5.566563701629638,
      "epoch": 2.382188615837009,
      "grad_norm": 1.125,
      "learning_rate": 0.00044294298117746863,
      "loss": 5.2238,
      "mean_token_accuracy": 0.18372552394866942,
      "num_tokens": 52299254.0,
      "step": 28355
    },
    {
      "entropy": 5.628901624679566,
      "epoch": 2.382608695652174,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044292305077877985,
      "loss": 5.2655,
      "mean_token_accuracy": 0.1795145958662033,
      "num_tokens": 52308907.0,
      "step": 28360
    },
    {
      "entropy": 5.637467575073242,
      "epoch": 2.3830287754673387,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004429031174054207,
      "loss": 5.1899,
      "mean_token_accuracy": 0.1850597396492958,
      "num_tokens": 52318092.0,
      "step": 28365
    },
    {
      "entropy": 5.491861867904663,
      "epoch": 2.3834488552825035,
      "grad_norm": 1.25,
      "learning_rate": 0.00044288318105774435,
      "loss": 5.1584,
      "mean_token_accuracy": 0.19270263612270355,
      "num_tokens": 52326797.0,
      "step": 28370
    },
    {
      "entropy": 5.48238787651062,
      "epoch": 2.3838689350976687,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044286324173610384,
      "loss": 5.1234,
      "mean_token_accuracy": 0.19223956316709517,
      "num_tokens": 52335730.0,
      "step": 28375
    },
    {
      "entropy": 5.4915753364562985,
      "epoch": 2.3842890149128335,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004428432994408524,
      "loss": 5.1059,
      "mean_token_accuracy": 0.19602700769901277,
      "num_tokens": 52344700.0,
      "step": 28380
    },
    {
      "entropy": 5.520236301422119,
      "epoch": 2.3847090947279983,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004428233541723434,
      "loss": 5.1561,
      "mean_token_accuracy": 0.19310613572597504,
      "num_tokens": 52354120.0,
      "step": 28385
    },
    {
      "entropy": 5.5985454559326175,
      "epoch": 2.385129174543163,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044280340593093006,
      "loss": 5.1937,
      "mean_token_accuracy": 0.18731513023376464,
      "num_tokens": 52362425.0,
      "step": 28390
    },
    {
      "entropy": 5.559153938293457,
      "epoch": 2.3855492543583283,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004427834547169658,
      "loss": 5.1746,
      "mean_token_accuracy": 0.18834541738033295,
      "num_tokens": 52371988.0,
      "step": 28395
    },
    {
      "entropy": 5.575638055801392,
      "epoch": 2.385969334173493,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004427635005308041,
      "loss": 5.241,
      "mean_token_accuracy": 0.1746767371892929,
      "num_tokens": 52381911.0,
      "step": 28400
    },
    {
      "entropy": 5.533910846710205,
      "epoch": 2.386389413988658,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044274354337279825,
      "loss": 5.112,
      "mean_token_accuracy": 0.19511540979146957,
      "num_tokens": 52391006.0,
      "step": 28405
    },
    {
      "entropy": 5.546326923370361,
      "epoch": 2.3868094938038227,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004427235832433019,
      "loss": 5.2061,
      "mean_token_accuracy": 0.18269854336977004,
      "num_tokens": 52401149.0,
      "step": 28410
    },
    {
      "entropy": 5.548501586914062,
      "epoch": 2.3872295736189875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044270362014266874,
      "loss": 5.187,
      "mean_token_accuracy": 0.1861854001879692,
      "num_tokens": 52409929.0,
      "step": 28415
    },
    {
      "entropy": 5.4906392097473145,
      "epoch": 2.3876496534341527,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044268365407125227,
      "loss": 5.0448,
      "mean_token_accuracy": 0.19479745328426362,
      "num_tokens": 52418900.0,
      "step": 28420
    },
    {
      "entropy": 5.583305692672729,
      "epoch": 2.3880697332493175,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044266368502940623,
      "loss": 5.2647,
      "mean_token_accuracy": 0.17502611577510835,
      "num_tokens": 52428761.0,
      "step": 28425
    },
    {
      "entropy": 5.52843656539917,
      "epoch": 2.3884898130644823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004426437130174843,
      "loss": 5.1512,
      "mean_token_accuracy": 0.19237127304077148,
      "num_tokens": 52437399.0,
      "step": 28430
    },
    {
      "entropy": 5.543826627731323,
      "epoch": 2.388909892879647,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004426237380358404,
      "loss": 5.2095,
      "mean_token_accuracy": 0.1812881797552109,
      "num_tokens": 52446703.0,
      "step": 28435
    },
    {
      "entropy": 5.606802558898925,
      "epoch": 2.389329972694812,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044260376008482816,
      "loss": 5.2387,
      "mean_token_accuracy": 0.19011248797178268,
      "num_tokens": 52456847.0,
      "step": 28440
    },
    {
      "entropy": 5.6582176208496096,
      "epoch": 2.389750052509977,
      "grad_norm": 1.375,
      "learning_rate": 0.00044258377916480177,
      "loss": 5.3225,
      "mean_token_accuracy": 0.1794867381453514,
      "num_tokens": 52466721.0,
      "step": 28445
    },
    {
      "entropy": 5.55763463973999,
      "epoch": 2.390170132325142,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004425637952761149,
      "loss": 5.2515,
      "mean_token_accuracy": 0.18985413163900375,
      "num_tokens": 52475169.0,
      "step": 28450
    },
    {
      "entropy": 5.5992566585540775,
      "epoch": 2.3905902121403066,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044254380841912173,
      "loss": 5.1984,
      "mean_token_accuracy": 0.18893173933029175,
      "num_tokens": 52484031.0,
      "step": 28455
    },
    {
      "entropy": 5.510385036468506,
      "epoch": 2.3910102919554714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004425238185941762,
      "loss": 5.1565,
      "mean_token_accuracy": 0.1877041056752205,
      "num_tokens": 52493511.0,
      "step": 28460
    },
    {
      "entropy": 5.490422534942627,
      "epoch": 2.391430371770636,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004425038258016326,
      "loss": 5.1315,
      "mean_token_accuracy": 0.18644145727157593,
      "num_tokens": 52502595.0,
      "step": 28465
    },
    {
      "entropy": 5.471070671081543,
      "epoch": 2.3918504515858015,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044248383004184485,
      "loss": 5.1046,
      "mean_token_accuracy": 0.19256136566400528,
      "num_tokens": 52511959.0,
      "step": 28470
    },
    {
      "entropy": 5.521812677383423,
      "epoch": 2.3922705314009662,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004424638313151674,
      "loss": 5.1564,
      "mean_token_accuracy": 0.18670637011528016,
      "num_tokens": 52520914.0,
      "step": 28475
    },
    {
      "entropy": 5.511662244796753,
      "epoch": 2.392690611216131,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004424438296219544,
      "loss": 5.1755,
      "mean_token_accuracy": 0.1883690282702446,
      "num_tokens": 52530361.0,
      "step": 28480
    },
    {
      "entropy": 5.470779132843018,
      "epoch": 2.393110691031296,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004424238249625602,
      "loss": 5.1245,
      "mean_token_accuracy": 0.188337542116642,
      "num_tokens": 52538832.0,
      "step": 28485
    },
    {
      "entropy": 5.5308998107910154,
      "epoch": 2.3935307708464606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044240381733733904,
      "loss": 5.1582,
      "mean_token_accuracy": 0.19179647117853166,
      "num_tokens": 52547587.0,
      "step": 28490
    },
    {
      "entropy": 5.52427077293396,
      "epoch": 2.393950850661626,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044238380674664553,
      "loss": 5.1235,
      "mean_token_accuracy": 0.18902744650840758,
      "num_tokens": 52556613.0,
      "step": 28495
    },
    {
      "entropy": 5.537208890914917,
      "epoch": 2.3943709304767906,
      "grad_norm": 1.09375,
      "learning_rate": 0.000442363793190834,
      "loss": 5.2128,
      "mean_token_accuracy": 0.17747605741024017,
      "num_tokens": 52566290.0,
      "step": 28500
    },
    {
      "entropy": 5.582032346725464,
      "epoch": 2.3947910102919554,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044234377667025924,
      "loss": 5.1782,
      "mean_token_accuracy": 0.1815616950392723,
      "num_tokens": 52575029.0,
      "step": 28505
    },
    {
      "entropy": 5.633134222030639,
      "epoch": 2.39521109010712,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044232375718527547,
      "loss": 5.2922,
      "mean_token_accuracy": 0.17977628856897354,
      "num_tokens": 52584285.0,
      "step": 28510
    },
    {
      "entropy": 5.599434328079224,
      "epoch": 2.3956311699222854,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004423037347362375,
      "loss": 5.268,
      "mean_token_accuracy": 0.1765652298927307,
      "num_tokens": 52594011.0,
      "step": 28515
    },
    {
      "entropy": 5.505746269226075,
      "epoch": 2.39605124973745,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004422837093235001,
      "loss": 5.1638,
      "mean_token_accuracy": 0.18815732449293138,
      "num_tokens": 52602482.0,
      "step": 28520
    },
    {
      "entropy": 5.508921957015991,
      "epoch": 2.396471329552615,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044226368094741795,
      "loss": 5.1827,
      "mean_token_accuracy": 0.1880326598882675,
      "num_tokens": 52612439.0,
      "step": 28525
    },
    {
      "entropy": 5.5956474304199215,
      "epoch": 2.39689140936778,
      "grad_norm": 1.25,
      "learning_rate": 0.00044224364960834567,
      "loss": 5.2771,
      "mean_token_accuracy": 0.1823123127222061,
      "num_tokens": 52621784.0,
      "step": 28530
    },
    {
      "entropy": 5.612422084808349,
      "epoch": 2.3973114891829446,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004422236153066384,
      "loss": 5.1549,
      "mean_token_accuracy": 0.19470335096120833,
      "num_tokens": 52631524.0,
      "step": 28535
    },
    {
      "entropy": 5.537948799133301,
      "epoch": 2.39773156899811,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004422035780426508,
      "loss": 5.1961,
      "mean_token_accuracy": 0.18728223741054534,
      "num_tokens": 52641003.0,
      "step": 28540
    },
    {
      "entropy": 5.554596900939941,
      "epoch": 2.3981516488132746,
      "grad_norm": 0.984375,
      "learning_rate": 0.000442183537816738,
      "loss": 5.2125,
      "mean_token_accuracy": 0.18274600952863693,
      "num_tokens": 52651199.0,
      "step": 28545
    },
    {
      "entropy": 5.567118501663208,
      "epoch": 2.3985717286284394,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044216349462925485,
      "loss": 5.2102,
      "mean_token_accuracy": 0.17799440920352935,
      "num_tokens": 52660381.0,
      "step": 28550
    },
    {
      "entropy": 5.4232395648956295,
      "epoch": 2.398991808443604,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004421434484805565,
      "loss": 5.1204,
      "mean_token_accuracy": 0.18729056268930436,
      "num_tokens": 52669961.0,
      "step": 28555
    },
    {
      "entropy": 5.526868152618408,
      "epoch": 2.3994118882587694,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000442123399370998,
      "loss": 5.1611,
      "mean_token_accuracy": 0.18984203040599823,
      "num_tokens": 52679579.0,
      "step": 28560
    },
    {
      "entropy": 5.47414755821228,
      "epoch": 2.399831968073934,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044210334730093454,
      "loss": 5.1089,
      "mean_token_accuracy": 0.19114680141210555,
      "num_tokens": 52688693.0,
      "step": 28565
    },
    {
      "entropy": 5.54118127822876,
      "epoch": 2.400252047889099,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004420832922707213,
      "loss": 5.1873,
      "mean_token_accuracy": 0.18568014204502106,
      "num_tokens": 52697117.0,
      "step": 28570
    },
    {
      "entropy": 5.605531930923462,
      "epoch": 2.4006721277042637,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044206323428071367,
      "loss": 5.3117,
      "mean_token_accuracy": 0.1790427789092064,
      "num_tokens": 52706287.0,
      "step": 28575
    },
    {
      "entropy": 5.56471061706543,
      "epoch": 2.4010922075194285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044204317333126675,
      "loss": 5.2269,
      "mean_token_accuracy": 0.18810694962739943,
      "num_tokens": 52715528.0,
      "step": 28580
    },
    {
      "entropy": 5.468083286285401,
      "epoch": 2.4015122873345938,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044202310942273616,
      "loss": 5.1488,
      "mean_token_accuracy": 0.18463410288095475,
      "num_tokens": 52723910.0,
      "step": 28585
    },
    {
      "entropy": 5.49551043510437,
      "epoch": 2.4019323671497586,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044200304255547705,
      "loss": 5.2239,
      "mean_token_accuracy": 0.18683918714523315,
      "num_tokens": 52733311.0,
      "step": 28590
    },
    {
      "entropy": 5.635931158065796,
      "epoch": 2.4023524469649233,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044198297272984513,
      "loss": 5.1967,
      "mean_token_accuracy": 0.18295684307813645,
      "num_tokens": 52742690.0,
      "step": 28595
    },
    {
      "entropy": 5.496310377120972,
      "epoch": 2.402772526780088,
      "grad_norm": 1.125,
      "learning_rate": 0.0004419628999461958,
      "loss": 5.0705,
      "mean_token_accuracy": 0.19864794462919236,
      "num_tokens": 52751517.0,
      "step": 28600
    },
    {
      "entropy": 5.481616687774658,
      "epoch": 2.403192606595253,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004419428242048847,
      "loss": 5.1433,
      "mean_token_accuracy": 0.18853924721479415,
      "num_tokens": 52760086.0,
      "step": 28605
    },
    {
      "entropy": 5.5477231502532955,
      "epoch": 2.403612686410418,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044192274550626737,
      "loss": 5.2598,
      "mean_token_accuracy": 0.17673608362674714,
      "num_tokens": 52769438.0,
      "step": 28610
    },
    {
      "entropy": 5.511176919937133,
      "epoch": 2.404032766225583,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004419026638506996,
      "loss": 5.1274,
      "mean_token_accuracy": 0.19004578292369842,
      "num_tokens": 52777779.0,
      "step": 28615
    },
    {
      "entropy": 5.575839376449585,
      "epoch": 2.4044528460407477,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044188257923853707,
      "loss": 5.218,
      "mean_token_accuracy": 0.18863237649202347,
      "num_tokens": 52787289.0,
      "step": 28620
    },
    {
      "entropy": 5.577554416656494,
      "epoch": 2.4048729258559125,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044186249167013565,
      "loss": 5.1709,
      "mean_token_accuracy": 0.18503408879041672,
      "num_tokens": 52796382.0,
      "step": 28625
    },
    {
      "entropy": 5.545231294631958,
      "epoch": 2.4052930056710773,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044184240114585104,
      "loss": 5.1715,
      "mean_token_accuracy": 0.18595199286937714,
      "num_tokens": 52805303.0,
      "step": 28630
    },
    {
      "entropy": 5.513259792327881,
      "epoch": 2.4057130854862425,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004418223076660392,
      "loss": 5.1866,
      "mean_token_accuracy": 0.18402250111103058,
      "num_tokens": 52814289.0,
      "step": 28635
    },
    {
      "entropy": 5.51842098236084,
      "epoch": 2.4061331653014073,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044180221123105607,
      "loss": 5.1211,
      "mean_token_accuracy": 0.19301627725362777,
      "num_tokens": 52823231.0,
      "step": 28640
    },
    {
      "entropy": 5.513109540939331,
      "epoch": 2.406553245116572,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044178211184125766,
      "loss": 5.2067,
      "mean_token_accuracy": 0.18538688570261003,
      "num_tokens": 52832380.0,
      "step": 28645
    },
    {
      "entropy": 5.55211877822876,
      "epoch": 2.406973324931737,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004417620094970001,
      "loss": 5.1511,
      "mean_token_accuracy": 0.20163844525814056,
      "num_tokens": 52840846.0,
      "step": 28650
    },
    {
      "entropy": 5.601649522781372,
      "epoch": 2.4073934047469017,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004417419041986393,
      "loss": 5.1571,
      "mean_token_accuracy": 0.1857021138072014,
      "num_tokens": 52850350.0,
      "step": 28655
    },
    {
      "entropy": 5.529404878616333,
      "epoch": 2.407813484562067,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004417217959465316,
      "loss": 5.0799,
      "mean_token_accuracy": 0.1870213821530342,
      "num_tokens": 52858825.0,
      "step": 28660
    },
    {
      "entropy": 5.54740047454834,
      "epoch": 2.4082335643772317,
      "grad_norm": 1.125,
      "learning_rate": 0.0004417016847410331,
      "loss": 5.1503,
      "mean_token_accuracy": 0.18167153298854827,
      "num_tokens": 52867685.0,
      "step": 28665
    },
    {
      "entropy": 5.516757440567017,
      "epoch": 2.4086536441923965,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044168157058250006,
      "loss": 5.1849,
      "mean_token_accuracy": 0.18449119627475738,
      "num_tokens": 52877474.0,
      "step": 28670
    },
    {
      "entropy": 5.4694085121154785,
      "epoch": 2.4090737240075613,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044166145347128893,
      "loss": 5.072,
      "mean_token_accuracy": 0.18739832043647767,
      "num_tokens": 52886059.0,
      "step": 28675
    },
    {
      "entropy": 5.5143438339233395,
      "epoch": 2.4094938038227265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004416413334077559,
      "loss": 5.1797,
      "mean_token_accuracy": 0.184693743288517,
      "num_tokens": 52894958.0,
      "step": 28680
    },
    {
      "entropy": 5.5229264259338375,
      "epoch": 2.4099138836378913,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044162121039225744,
      "loss": 5.1579,
      "mean_token_accuracy": 0.1862056002020836,
      "num_tokens": 52903467.0,
      "step": 28685
    },
    {
      "entropy": 5.543050813674927,
      "epoch": 2.410333963453056,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044160108442515005,
      "loss": 5.1433,
      "mean_token_accuracy": 0.1892212837934494,
      "num_tokens": 52912383.0,
      "step": 28690
    },
    {
      "entropy": 5.487414073944092,
      "epoch": 2.410754043268221,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004415809555067903,
      "loss": 5.1254,
      "mean_token_accuracy": 0.1939299777150154,
      "num_tokens": 52921588.0,
      "step": 28695
    },
    {
      "entropy": 5.565346145629883,
      "epoch": 2.411174123083386,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044156082363753466,
      "loss": 5.1854,
      "mean_token_accuracy": 0.17888289093971252,
      "num_tokens": 52931007.0,
      "step": 28700
    },
    {
      "entropy": 5.545330286026001,
      "epoch": 2.411594202898551,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004415406888177398,
      "loss": 5.2133,
      "mean_token_accuracy": 0.18854071348905563,
      "num_tokens": 52939828.0,
      "step": 28705
    },
    {
      "entropy": 5.564816856384278,
      "epoch": 2.4120142827137157,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004415205510477623,
      "loss": 5.1137,
      "mean_token_accuracy": 0.19134149998426436,
      "num_tokens": 52949010.0,
      "step": 28710
    },
    {
      "entropy": 5.605798959732056,
      "epoch": 2.4124343625288804,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044150041032795905,
      "loss": 5.1909,
      "mean_token_accuracy": 0.19000301659107208,
      "num_tokens": 52957805.0,
      "step": 28715
    },
    {
      "entropy": 5.558403730392456,
      "epoch": 2.4128544423440452,
      "grad_norm": 1.125,
      "learning_rate": 0.00044148026665868684,
      "loss": 5.2245,
      "mean_token_accuracy": 0.1890776500105858,
      "num_tokens": 52967115.0,
      "step": 28720
    },
    {
      "entropy": 5.580177688598633,
      "epoch": 2.4132745221592105,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004414601200403023,
      "loss": 5.1895,
      "mean_token_accuracy": 0.18433496057987214,
      "num_tokens": 52975677.0,
      "step": 28725
    },
    {
      "entropy": 5.564523363113404,
      "epoch": 2.4136946019743752,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044143997047316257,
      "loss": 5.1448,
      "mean_token_accuracy": 0.18864301443099976,
      "num_tokens": 52983961.0,
      "step": 28730
    },
    {
      "entropy": 5.450777578353882,
      "epoch": 2.41411468178954,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004414198179576244,
      "loss": 5.2157,
      "mean_token_accuracy": 0.18837990313768388,
      "num_tokens": 52994147.0,
      "step": 28735
    },
    {
      "entropy": 5.569072294235229,
      "epoch": 2.414534761604705,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044139966249404485,
      "loss": 5.1924,
      "mean_token_accuracy": 0.18067477196455,
      "num_tokens": 53002901.0,
      "step": 28740
    },
    {
      "entropy": 5.522045135498047,
      "epoch": 2.4149548414198696,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000441379504082781,
      "loss": 5.2895,
      "mean_token_accuracy": 0.1766059011220932,
      "num_tokens": 53012634.0,
      "step": 28745
    },
    {
      "entropy": 5.649054718017578,
      "epoch": 2.415374921235035,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044135934272418985,
      "loss": 5.297,
      "mean_token_accuracy": 0.17570248991250992,
      "num_tokens": 53022290.0,
      "step": 28750
    },
    {
      "entropy": 5.667205667495727,
      "epoch": 2.4157950010501996,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004413391784186287,
      "loss": 5.2386,
      "mean_token_accuracy": 0.18844350874423982,
      "num_tokens": 53031317.0,
      "step": 28755
    },
    {
      "entropy": 5.58490252494812,
      "epoch": 2.4162150808653644,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004413190111664545,
      "loss": 5.2374,
      "mean_token_accuracy": 0.17927589863538743,
      "num_tokens": 53040540.0,
      "step": 28760
    },
    {
      "entropy": 5.565689134597778,
      "epoch": 2.416635160680529,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004412988409680247,
      "loss": 5.2483,
      "mean_token_accuracy": 0.1843319684267044,
      "num_tokens": 53050170.0,
      "step": 28765
    },
    {
      "entropy": 5.6016278743743895,
      "epoch": 2.417055240495694,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004412786678236967,
      "loss": 5.1863,
      "mean_token_accuracy": 0.18619930148124694,
      "num_tokens": 53059011.0,
      "step": 28770
    },
    {
      "entropy": 5.581191492080689,
      "epoch": 2.417475320310859,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044125849173382763,
      "loss": 5.1935,
      "mean_token_accuracy": 0.18331741243600846,
      "num_tokens": 53068179.0,
      "step": 28775
    },
    {
      "entropy": 5.463512325286866,
      "epoch": 2.417895400126024,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000441238312698775,
      "loss": 5.1968,
      "mean_token_accuracy": 0.18803776800632477,
      "num_tokens": 53077165.0,
      "step": 28780
    },
    {
      "entropy": 5.422934675216675,
      "epoch": 2.418315479941189,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004412181307188962,
      "loss": 5.0755,
      "mean_token_accuracy": 0.19276057481765746,
      "num_tokens": 53086364.0,
      "step": 28785
    },
    {
      "entropy": 5.616539096832275,
      "epoch": 2.4187355597563536,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004411979457945489,
      "loss": 5.2578,
      "mean_token_accuracy": 0.18165620565414428,
      "num_tokens": 53097519.0,
      "step": 28790
    },
    {
      "entropy": 5.593088817596436,
      "epoch": 2.4191556395715184,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044117775792609055,
      "loss": 5.1989,
      "mean_token_accuracy": 0.19215254187583924,
      "num_tokens": 53106108.0,
      "step": 28795
    },
    {
      "entropy": 5.545949506759643,
      "epoch": 2.4195757193866836,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044115756711387877,
      "loss": 5.2215,
      "mean_token_accuracy": 0.17903274893760682,
      "num_tokens": 53115048.0,
      "step": 28800
    },
    {
      "entropy": 5.556657838821411,
      "epoch": 2.4199957992018484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004411373733582712,
      "loss": 5.2183,
      "mean_token_accuracy": 0.1853382870554924,
      "num_tokens": 53124313.0,
      "step": 28805
    },
    {
      "entropy": 5.56322660446167,
      "epoch": 2.420415879017013,
      "grad_norm": 1.125,
      "learning_rate": 0.0004411171766596257,
      "loss": 5.229,
      "mean_token_accuracy": 0.18843216747045516,
      "num_tokens": 53132957.0,
      "step": 28810
    },
    {
      "entropy": 5.585582733154297,
      "epoch": 2.420835958832178,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044109697701829985,
      "loss": 5.276,
      "mean_token_accuracy": 0.1778305470943451,
      "num_tokens": 53142083.0,
      "step": 28815
    },
    {
      "entropy": 5.600375652313232,
      "epoch": 2.421256038647343,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044107677443465165,
      "loss": 5.2273,
      "mean_token_accuracy": 0.18351965099573136,
      "num_tokens": 53152230.0,
      "step": 28820
    },
    {
      "entropy": 5.597615909576416,
      "epoch": 2.421676118462508,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044105656890903896,
      "loss": 5.2317,
      "mean_token_accuracy": 0.1862358644604683,
      "num_tokens": 53161831.0,
      "step": 28825
    },
    {
      "entropy": 5.5413079261779785,
      "epoch": 2.4220961982776728,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044103636044181947,
      "loss": 5.1509,
      "mean_token_accuracy": 0.19017695039510726,
      "num_tokens": 53171506.0,
      "step": 28830
    },
    {
      "entropy": 5.457686901092529,
      "epoch": 2.4225162780928375,
      "grad_norm": 1.125,
      "learning_rate": 0.00044101614903335154,
      "loss": 5.056,
      "mean_token_accuracy": 0.19090916216373444,
      "num_tokens": 53180795.0,
      "step": 28835
    },
    {
      "entropy": 5.47523865699768,
      "epoch": 2.4229363579080028,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000440995934683993,
      "loss": 5.1076,
      "mean_token_accuracy": 0.19101362228393554,
      "num_tokens": 53189788.0,
      "step": 28840
    },
    {
      "entropy": 5.501348686218262,
      "epoch": 2.4233564377231676,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004409757173941019,
      "loss": 5.154,
      "mean_token_accuracy": 0.19154275059700013,
      "num_tokens": 53199798.0,
      "step": 28845
    },
    {
      "entropy": 5.556332302093506,
      "epoch": 2.4237765175383323,
      "grad_norm": 1.125,
      "learning_rate": 0.0004409554971640363,
      "loss": 5.1106,
      "mean_token_accuracy": 0.1929522454738617,
      "num_tokens": 53208744.0,
      "step": 28850
    },
    {
      "entropy": 5.5548523426055905,
      "epoch": 2.424196597353497,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004409352739941547,
      "loss": 5.2706,
      "mean_token_accuracy": 0.18923532217741013,
      "num_tokens": 53217705.0,
      "step": 28855
    },
    {
      "entropy": 5.598286056518555,
      "epoch": 2.424616677168662,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044091504788481515,
      "loss": 5.2118,
      "mean_token_accuracy": 0.17807336747646332,
      "num_tokens": 53226888.0,
      "step": 28860
    },
    {
      "entropy": 5.526840543746948,
      "epoch": 2.425036756983827,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044089481883637587,
      "loss": 5.1304,
      "mean_token_accuracy": 0.18674751967191697,
      "num_tokens": 53236651.0,
      "step": 28865
    },
    {
      "entropy": 5.496712064743042,
      "epoch": 2.425456836798992,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004408745868491954,
      "loss": 5.1908,
      "mean_token_accuracy": 0.18027398362755775,
      "num_tokens": 53245488.0,
      "step": 28870
    },
    {
      "entropy": 5.560531949996948,
      "epoch": 2.4258769166141567,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004408543519236319,
      "loss": 5.1313,
      "mean_token_accuracy": 0.1918896958231926,
      "num_tokens": 53254244.0,
      "step": 28875
    },
    {
      "entropy": 5.597659730911255,
      "epoch": 2.4262969964293215,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004408341140600441,
      "loss": 5.2273,
      "mean_token_accuracy": 0.18483968824148178,
      "num_tokens": 53263784.0,
      "step": 28880
    },
    {
      "entropy": 5.558574104309082,
      "epoch": 2.4267170762444863,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044081387325879033,
      "loss": 5.1885,
      "mean_token_accuracy": 0.18149202615022658,
      "num_tokens": 53273442.0,
      "step": 28885
    },
    {
      "entropy": 5.5171184062957765,
      "epoch": 2.4271371560596515,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044079362952022916,
      "loss": 5.1947,
      "mean_token_accuracy": 0.1868927761912346,
      "num_tokens": 53282774.0,
      "step": 28890
    },
    {
      "entropy": 5.577679777145386,
      "epoch": 2.4275572358748163,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044077338284471924,
      "loss": 5.2589,
      "mean_token_accuracy": 0.1768503576517105,
      "num_tokens": 53292346.0,
      "step": 28895
    },
    {
      "entropy": 5.648600053787232,
      "epoch": 2.427977315689981,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004407531332326191,
      "loss": 5.2419,
      "mean_token_accuracy": 0.1804145410656929,
      "num_tokens": 53301628.0,
      "step": 28900
    },
    {
      "entropy": 5.5598962783813475,
      "epoch": 2.428397395505146,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044073288068428764,
      "loss": 5.2253,
      "mean_token_accuracy": 0.19050951898097992,
      "num_tokens": 53311514.0,
      "step": 28905
    },
    {
      "entropy": 5.604902458190918,
      "epoch": 2.4288174753203107,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004407126252000835,
      "loss": 5.2671,
      "mean_token_accuracy": 0.18018261194229127,
      "num_tokens": 53320652.0,
      "step": 28910
    },
    {
      "entropy": 5.598904085159302,
      "epoch": 2.429237555135476,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044069236678036553,
      "loss": 5.2607,
      "mean_token_accuracy": 0.18349702805280685,
      "num_tokens": 53329989.0,
      "step": 28915
    },
    {
      "entropy": 5.599282121658325,
      "epoch": 2.4296576349506407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004406721054254927,
      "loss": 5.3191,
      "mean_token_accuracy": 0.1804218992590904,
      "num_tokens": 53339017.0,
      "step": 28920
    },
    {
      "entropy": 5.552645826339722,
      "epoch": 2.4300777147658055,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004406518411358238,
      "loss": 5.1239,
      "mean_token_accuracy": 0.18791444152593612,
      "num_tokens": 53348629.0,
      "step": 28925
    },
    {
      "entropy": 5.450558280944824,
      "epoch": 2.4304977945809703,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044063157391171777,
      "loss": 5.1238,
      "mean_token_accuracy": 0.1917681261897087,
      "num_tokens": 53356997.0,
      "step": 28930
    },
    {
      "entropy": 5.559162950515747,
      "epoch": 2.430917874396135,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044061130375353365,
      "loss": 5.1818,
      "mean_token_accuracy": 0.18487947136163713,
      "num_tokens": 53366688.0,
      "step": 28935
    },
    {
      "entropy": 5.488865900039673,
      "epoch": 2.4313379542113003,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004405910306616307,
      "loss": 5.1071,
      "mean_token_accuracy": 0.19780228435993194,
      "num_tokens": 53375851.0,
      "step": 28940
    },
    {
      "entropy": 5.639904165267945,
      "epoch": 2.431758034026465,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004405707546363678,
      "loss": 5.3588,
      "mean_token_accuracy": 0.17668951898813248,
      "num_tokens": 53385848.0,
      "step": 28945
    },
    {
      "entropy": 5.526924133300781,
      "epoch": 2.43217811384163,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004405504756781043,
      "loss": 5.1091,
      "mean_token_accuracy": 0.18610826432704924,
      "num_tokens": 53395472.0,
      "step": 28950
    },
    {
      "entropy": 5.513401031494141,
      "epoch": 2.4325981936567946,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044053019378719935,
      "loss": 5.1499,
      "mean_token_accuracy": 0.19761219769716262,
      "num_tokens": 53404571.0,
      "step": 28955
    },
    {
      "entropy": 5.522523355484009,
      "epoch": 2.4330182734719594,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004405099089640122,
      "loss": 5.2529,
      "mean_token_accuracy": 0.18409932404756546,
      "num_tokens": 53413991.0,
      "step": 28960
    },
    {
      "entropy": 5.602576732635498,
      "epoch": 2.4334383532871247,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044048962120890237,
      "loss": 5.2228,
      "mean_token_accuracy": 0.1858634665608406,
      "num_tokens": 53423116.0,
      "step": 28965
    },
    {
      "entropy": 5.524973821640015,
      "epoch": 2.4338584331022894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044046933052222897,
      "loss": 5.2033,
      "mean_token_accuracy": 0.18117294311523438,
      "num_tokens": 53432750.0,
      "step": 28970
    },
    {
      "entropy": 5.552004671096801,
      "epoch": 2.4342785129174542,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044044903690435163,
      "loss": 5.2691,
      "mean_token_accuracy": 0.18134916126728057,
      "num_tokens": 53442847.0,
      "step": 28975
    },
    {
      "entropy": 5.582878923416137,
      "epoch": 2.434698592732619,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044042874035562986,
      "loss": 5.2285,
      "mean_token_accuracy": 0.18341782838106155,
      "num_tokens": 53452521.0,
      "step": 28980
    },
    {
      "entropy": 5.480812740325928,
      "epoch": 2.4351186725477842,
      "grad_norm": 1.125,
      "learning_rate": 0.0004404084408764231,
      "loss": 5.1038,
      "mean_token_accuracy": 0.1896662026643753,
      "num_tokens": 53461557.0,
      "step": 28985
    },
    {
      "entropy": 5.482645797729492,
      "epoch": 2.435538752362949,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000440388138467091,
      "loss": 5.1851,
      "mean_token_accuracy": 0.1861457422375679,
      "num_tokens": 53470639.0,
      "step": 28990
    },
    {
      "entropy": 5.559153747558594,
      "epoch": 2.435958832178114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044036783312799313,
      "loss": 5.1716,
      "mean_token_accuracy": 0.18894788175821303,
      "num_tokens": 53480160.0,
      "step": 28995
    },
    {
      "entropy": 5.6010338306427006,
      "epoch": 2.4363789119932786,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044034752485948926,
      "loss": 5.2593,
      "mean_token_accuracy": 0.1800095334649086,
      "num_tokens": 53488853.0,
      "step": 29000
    },
    {
      "entropy": 5.596753358840942,
      "epoch": 2.436798991808444,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004403272136619391,
      "loss": 5.2681,
      "mean_token_accuracy": 0.18534953743219376,
      "num_tokens": 53498629.0,
      "step": 29005
    },
    {
      "entropy": 5.536453771591186,
      "epoch": 2.4372190716236086,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044030689953570255,
      "loss": 5.1298,
      "mean_token_accuracy": 0.18883957117795944,
      "num_tokens": 53506708.0,
      "step": 29010
    },
    {
      "entropy": 5.567599964141846,
      "epoch": 2.4376391514387734,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044028658248113935,
      "loss": 5.1707,
      "mean_token_accuracy": 0.18326132446527482,
      "num_tokens": 53515890.0,
      "step": 29015
    },
    {
      "entropy": 5.587323999404907,
      "epoch": 2.438059231253938,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004402662624986095,
      "loss": 5.2929,
      "mean_token_accuracy": 0.18508219718933105,
      "num_tokens": 53525987.0,
      "step": 29020
    },
    {
      "entropy": 5.505487537384033,
      "epoch": 2.438479311069103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044024593958847286,
      "loss": 5.0859,
      "mean_token_accuracy": 0.18622301518917084,
      "num_tokens": 53535757.0,
      "step": 29025
    },
    {
      "entropy": 5.467153644561767,
      "epoch": 2.438899390884268,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044022561375108954,
      "loss": 5.2067,
      "mean_token_accuracy": 0.18843583911657333,
      "num_tokens": 53544946.0,
      "step": 29030
    },
    {
      "entropy": 5.490068483352661,
      "epoch": 2.439319470699433,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004402052849868195,
      "loss": 5.1045,
      "mean_token_accuracy": 0.18236029744148255,
      "num_tokens": 53554594.0,
      "step": 29035
    },
    {
      "entropy": 5.447255373001099,
      "epoch": 2.439739550514598,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044018495329602294,
      "loss": 5.0744,
      "mean_token_accuracy": 0.19647325426340104,
      "num_tokens": 53563074.0,
      "step": 29040
    },
    {
      "entropy": 5.582076072692871,
      "epoch": 2.4401596303297626,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044016461867906,
      "loss": 5.3071,
      "mean_token_accuracy": 0.178119258582592,
      "num_tokens": 53572552.0,
      "step": 29045
    },
    {
      "entropy": 5.479238986968994,
      "epoch": 2.4405797101449274,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004401442811362908,
      "loss": 5.0771,
      "mean_token_accuracy": 0.19191126227378846,
      "num_tokens": 53582049.0,
      "step": 29050
    },
    {
      "entropy": 5.561147928237915,
      "epoch": 2.4409997899600926,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004401239406680758,
      "loss": 5.2631,
      "mean_token_accuracy": 0.1819860726594925,
      "num_tokens": 53590796.0,
      "step": 29055
    },
    {
      "entropy": 5.508671665191651,
      "epoch": 2.4414198697752574,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044010359727477517,
      "loss": 5.1778,
      "mean_token_accuracy": 0.17894264310598373,
      "num_tokens": 53600221.0,
      "step": 29060
    },
    {
      "entropy": 5.590003108978271,
      "epoch": 2.441839949590422,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004400832509567493,
      "loss": 5.2484,
      "mean_token_accuracy": 0.1888725310564041,
      "num_tokens": 53610601.0,
      "step": 29065
    },
    {
      "entropy": 5.626228952407837,
      "epoch": 2.442260029405587,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004400629017143587,
      "loss": 5.2373,
      "mean_token_accuracy": 0.18696290701627732,
      "num_tokens": 53619563.0,
      "step": 29070
    },
    {
      "entropy": 5.4859401226043705,
      "epoch": 2.4426801092207517,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044004254954796386,
      "loss": 5.1174,
      "mean_token_accuracy": 0.18729193657636642,
      "num_tokens": 53628061.0,
      "step": 29075
    },
    {
      "entropy": 5.499954605102539,
      "epoch": 2.443100189035917,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004400221944579251,
      "loss": 5.2022,
      "mean_token_accuracy": 0.1899036258459091,
      "num_tokens": 53636874.0,
      "step": 29080
    },
    {
      "entropy": 5.541916084289551,
      "epoch": 2.4435202688510818,
      "grad_norm": 1.25,
      "learning_rate": 0.00044000183644460323,
      "loss": 5.2324,
      "mean_token_accuracy": 0.18110139816999435,
      "num_tokens": 53646956.0,
      "step": 29085
    },
    {
      "entropy": 5.612124109268189,
      "epoch": 2.4439403486662465,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004399814755083587,
      "loss": 5.2408,
      "mean_token_accuracy": 0.18037301003932954,
      "num_tokens": 53655878.0,
      "step": 29090
    },
    {
      "entropy": 5.542354965209961,
      "epoch": 2.4443604284814113,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004399611116495524,
      "loss": 5.2217,
      "mean_token_accuracy": 0.184790900349617,
      "num_tokens": 53664852.0,
      "step": 29095
    },
    {
      "entropy": 5.552360868453979,
      "epoch": 2.444780508296576,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004399407448685449,
      "loss": 5.203,
      "mean_token_accuracy": 0.184094075858593,
      "num_tokens": 53674031.0,
      "step": 29100
    },
    {
      "entropy": 5.551133012771606,
      "epoch": 2.4452005881117413,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043992037516569696,
      "loss": 5.1598,
      "mean_token_accuracy": 0.1862758606672287,
      "num_tokens": 53683626.0,
      "step": 29105
    },
    {
      "entropy": 5.491340160369873,
      "epoch": 2.445620667926906,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004399000025413696,
      "loss": 5.2107,
      "mean_token_accuracy": 0.18200964033603667,
      "num_tokens": 53693280.0,
      "step": 29110
    },
    {
      "entropy": 5.577686595916748,
      "epoch": 2.446040747742071,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004398796269959235,
      "loss": 5.2306,
      "mean_token_accuracy": 0.17795751690864564,
      "num_tokens": 53703057.0,
      "step": 29115
    },
    {
      "entropy": 5.559172773361206,
      "epoch": 2.4464608275572357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004398592485297198,
      "loss": 5.1241,
      "mean_token_accuracy": 0.18990721553564072,
      "num_tokens": 53712488.0,
      "step": 29120
    },
    {
      "entropy": 5.498211622238159,
      "epoch": 2.446880907372401,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043983886714311937,
      "loss": 5.2112,
      "mean_token_accuracy": 0.19492910355329512,
      "num_tokens": 53721770.0,
      "step": 29125
    },
    {
      "entropy": 5.551382637023925,
      "epoch": 2.4473009871875657,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004398184828364833,
      "loss": 5.2056,
      "mean_token_accuracy": 0.19021396189928055,
      "num_tokens": 53731074.0,
      "step": 29130
    },
    {
      "entropy": 5.618428373336792,
      "epoch": 2.4477210670027305,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004397980956101727,
      "loss": 5.2313,
      "mean_token_accuracy": 0.19047319740056992,
      "num_tokens": 53740108.0,
      "step": 29135
    },
    {
      "entropy": 5.532301187515259,
      "epoch": 2.4481411468178953,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043977770546454857,
      "loss": 5.1692,
      "mean_token_accuracy": 0.18058692812919616,
      "num_tokens": 53748949.0,
      "step": 29140
    },
    {
      "entropy": 5.532143449783325,
      "epoch": 2.4485612266330605,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043975731239997236,
      "loss": 5.1951,
      "mean_token_accuracy": 0.18589722216129304,
      "num_tokens": 53758098.0,
      "step": 29145
    },
    {
      "entropy": 5.531027793884277,
      "epoch": 2.4489813064482253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043973691641680516,
      "loss": 5.2196,
      "mean_token_accuracy": 0.1826300010085106,
      "num_tokens": 53766833.0,
      "step": 29150
    },
    {
      "entropy": 5.537526655197143,
      "epoch": 2.44940138626339,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004397165175154083,
      "loss": 5.2122,
      "mean_token_accuracy": 0.17827162742614747,
      "num_tokens": 53777292.0,
      "step": 29155
    },
    {
      "entropy": 5.559734106063843,
      "epoch": 2.449821466078555,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004396961156961432,
      "loss": 5.23,
      "mean_token_accuracy": 0.1782834753394127,
      "num_tokens": 53787401.0,
      "step": 29160
    },
    {
      "entropy": 5.560726308822632,
      "epoch": 2.4502415458937197,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004396757109593711,
      "loss": 5.1955,
      "mean_token_accuracy": 0.18336710035800935,
      "num_tokens": 53796429.0,
      "step": 29165
    },
    {
      "entropy": 5.5555120468139645,
      "epoch": 2.450661625708885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043965530330545373,
      "loss": 5.1439,
      "mean_token_accuracy": 0.1878969982266426,
      "num_tokens": 53805107.0,
      "step": 29170
    },
    {
      "entropy": 5.5764671802520756,
      "epoch": 2.4510817055240497,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043963489273475233,
      "loss": 5.2311,
      "mean_token_accuracy": 0.1879859372973442,
      "num_tokens": 53815244.0,
      "step": 29175
    },
    {
      "entropy": 5.504023551940918,
      "epoch": 2.4515017853392145,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004396144792476286,
      "loss": 5.1916,
      "mean_token_accuracy": 0.17959463745355606,
      "num_tokens": 53824278.0,
      "step": 29180
    },
    {
      "entropy": 5.530637407302857,
      "epoch": 2.4519218651543793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004395940628444442,
      "loss": 5.0391,
      "mean_token_accuracy": 0.197456656396389,
      "num_tokens": 53833048.0,
      "step": 29185
    },
    {
      "entropy": 5.554180908203125,
      "epoch": 2.452341944969544,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004395736435255607,
      "loss": 5.2316,
      "mean_token_accuracy": 0.18251992017030716,
      "num_tokens": 53842680.0,
      "step": 29190
    },
    {
      "entropy": 5.519055652618408,
      "epoch": 2.4527620247847093,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004395532212913398,
      "loss": 5.2547,
      "mean_token_accuracy": 0.18099162727594376,
      "num_tokens": 53852501.0,
      "step": 29195
    },
    {
      "entropy": 5.647069072723388,
      "epoch": 2.453182104599874,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043953279614214344,
      "loss": 5.2595,
      "mean_token_accuracy": 0.1819934532046318,
      "num_tokens": 53861957.0,
      "step": 29200
    },
    {
      "entropy": 5.551301670074463,
      "epoch": 2.453602184415039,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043951236807833314,
      "loss": 5.1957,
      "mean_token_accuracy": 0.1843494951725006,
      "num_tokens": 53872273.0,
      "step": 29205
    },
    {
      "entropy": 5.502133321762085,
      "epoch": 2.4540222642302036,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004394919371002711,
      "loss": 5.1306,
      "mean_token_accuracy": 0.18901265561580657,
      "num_tokens": 53881878.0,
      "step": 29210
    },
    {
      "entropy": 5.5266834735870365,
      "epoch": 2.4544423440453684,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043947150320831907,
      "loss": 5.2004,
      "mean_token_accuracy": 0.19439601749181748,
      "num_tokens": 53890836.0,
      "step": 29215
    },
    {
      "entropy": 5.482867002487183,
      "epoch": 2.4548624238605337,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043945106640283895,
      "loss": 5.1381,
      "mean_token_accuracy": 0.19602901935577394,
      "num_tokens": 53899883.0,
      "step": 29220
    },
    {
      "entropy": 5.48344407081604,
      "epoch": 2.4552825036756984,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043943062668419297,
      "loss": 5.072,
      "mean_token_accuracy": 0.19011911749839783,
      "num_tokens": 53908062.0,
      "step": 29225
    },
    {
      "entropy": 5.536656522750855,
      "epoch": 2.4557025834908632,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004394101840527431,
      "loss": 5.2658,
      "mean_token_accuracy": 0.18140083700418472,
      "num_tokens": 53917079.0,
      "step": 29230
    },
    {
      "entropy": 5.669253301620484,
      "epoch": 2.456122663306028,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043938973850885145,
      "loss": 5.1667,
      "mean_token_accuracy": 0.1949290469288826,
      "num_tokens": 53927282.0,
      "step": 29235
    },
    {
      "entropy": 5.587669563293457,
      "epoch": 2.456542743121193,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043936929005288015,
      "loss": 5.2099,
      "mean_token_accuracy": 0.19670029431581498,
      "num_tokens": 53936731.0,
      "step": 29240
    },
    {
      "entropy": 5.424340009689331,
      "epoch": 2.456962822936358,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043934883868519166,
      "loss": 5.1332,
      "mean_token_accuracy": 0.19721993207931518,
      "num_tokens": 53944893.0,
      "step": 29245
    },
    {
      "entropy": 5.545687675476074,
      "epoch": 2.457382902751523,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043932838440614803,
      "loss": 5.2347,
      "mean_token_accuracy": 0.1850252315402031,
      "num_tokens": 53954625.0,
      "step": 29250
    },
    {
      "entropy": 5.60219259262085,
      "epoch": 2.4578029825666876,
      "grad_norm": 1.25,
      "learning_rate": 0.0004393079272161118,
      "loss": 5.2165,
      "mean_token_accuracy": 0.1861887037754059,
      "num_tokens": 53963206.0,
      "step": 29255
    },
    {
      "entropy": 5.467782306671142,
      "epoch": 2.4582230623818524,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004392874671154451,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19192075878381729,
      "num_tokens": 53972256.0,
      "step": 29260
    },
    {
      "entropy": 5.583495569229126,
      "epoch": 2.458643142197017,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043926700410451054,
      "loss": 5.2039,
      "mean_token_accuracy": 0.18597334921360015,
      "num_tokens": 53981253.0,
      "step": 29265
    },
    {
      "entropy": 5.548670291900635,
      "epoch": 2.4590632220121824,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043924653818367066,
      "loss": 5.2817,
      "mean_token_accuracy": 0.17894334495067596,
      "num_tokens": 53991065.0,
      "step": 29270
    },
    {
      "entropy": 5.495116901397705,
      "epoch": 2.459483301827347,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043922606935328784,
      "loss": 5.1208,
      "mean_token_accuracy": 0.18340182304382324,
      "num_tokens": 54000768.0,
      "step": 29275
    },
    {
      "entropy": 5.6307289600372314,
      "epoch": 2.459903381642512,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004392055976137248,
      "loss": 5.3313,
      "mean_token_accuracy": 0.17475153356790543,
      "num_tokens": 54011332.0,
      "step": 29280
    },
    {
      "entropy": 5.653599596023559,
      "epoch": 2.4603234614576768,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043918512296534426,
      "loss": 5.2262,
      "mean_token_accuracy": 0.18628350496292115,
      "num_tokens": 54020627.0,
      "step": 29285
    },
    {
      "entropy": 5.609393119812012,
      "epoch": 2.460743541272842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004391646454085087,
      "loss": 5.2355,
      "mean_token_accuracy": 0.18092881441116332,
      "num_tokens": 54029564.0,
      "step": 29290
    },
    {
      "entropy": 5.545396995544434,
      "epoch": 2.461163621088007,
      "grad_norm": 1.09375,
      "learning_rate": 0.000439144164943581,
      "loss": 5.2464,
      "mean_token_accuracy": 0.18168343752622604,
      "num_tokens": 54038638.0,
      "step": 29295
    },
    {
      "entropy": 5.47169713973999,
      "epoch": 2.4615837009031716,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004391236815709239,
      "loss": 5.069,
      "mean_token_accuracy": 0.1955637663602829,
      "num_tokens": 54047007.0,
      "step": 29300
    },
    {
      "entropy": 5.612274837493897,
      "epoch": 2.4620037807183364,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004391031952909003,
      "loss": 5.2502,
      "mean_token_accuracy": 0.18322889059782027,
      "num_tokens": 54056505.0,
      "step": 29305
    },
    {
      "entropy": 5.571122884750366,
      "epoch": 2.4624238605335016,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043908270610387316,
      "loss": 5.134,
      "mean_token_accuracy": 0.18651456385850906,
      "num_tokens": 54065250.0,
      "step": 29310
    },
    {
      "entropy": 5.483246660232544,
      "epoch": 2.4628439403486664,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043906221401020533,
      "loss": 5.1713,
      "mean_token_accuracy": 0.18693744838237764,
      "num_tokens": 54073780.0,
      "step": 29315
    },
    {
      "entropy": 5.622607564926147,
      "epoch": 2.463264020163831,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043904171901025985,
      "loss": 5.3185,
      "mean_token_accuracy": 0.1850685402750969,
      "num_tokens": 54083765.0,
      "step": 29320
    },
    {
      "entropy": 5.633510112762451,
      "epoch": 2.463684099978996,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043902122110439987,
      "loss": 5.1937,
      "mean_token_accuracy": 0.18450938016176224,
      "num_tokens": 54093163.0,
      "step": 29325
    },
    {
      "entropy": 5.526914978027344,
      "epoch": 2.4641041797941607,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043900072029298833,
      "loss": 5.2036,
      "mean_token_accuracy": 0.18844327330589294,
      "num_tokens": 54101835.0,
      "step": 29330
    },
    {
      "entropy": 5.531244564056396,
      "epoch": 2.464524259609326,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004389802165763885,
      "loss": 5.2154,
      "mean_token_accuracy": 0.18211322724819184,
      "num_tokens": 54111053.0,
      "step": 29335
    },
    {
      "entropy": 5.578092813491821,
      "epoch": 2.4649443394244908,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004389597099549635,
      "loss": 5.1684,
      "mean_token_accuracy": 0.18510694950819015,
      "num_tokens": 54120098.0,
      "step": 29340
    },
    {
      "entropy": 5.547715473175049,
      "epoch": 2.4653644192396555,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043893920042907677,
      "loss": 5.1768,
      "mean_token_accuracy": 0.1901593953371048,
      "num_tokens": 54129129.0,
      "step": 29345
    },
    {
      "entropy": 5.534316778182983,
      "epoch": 2.4657844990548203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043891868799909146,
      "loss": 5.227,
      "mean_token_accuracy": 0.18439068347215654,
      "num_tokens": 54138459.0,
      "step": 29350
    },
    {
      "entropy": 5.6719828128814695,
      "epoch": 2.466204578869985,
      "grad_norm": 1.03125,
      "learning_rate": 0.000438898172665371,
      "loss": 5.2916,
      "mean_token_accuracy": 0.17418304234743118,
      "num_tokens": 54148046.0,
      "step": 29355
    },
    {
      "entropy": 5.6322455406188965,
      "epoch": 2.4666246586851504,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004388776544282789,
      "loss": 5.267,
      "mean_token_accuracy": 0.18093367666006088,
      "num_tokens": 54157310.0,
      "step": 29360
    },
    {
      "entropy": 5.5285602569580075,
      "epoch": 2.467044738500315,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043885713328817847,
      "loss": 5.0896,
      "mean_token_accuracy": 0.1922574907541275,
      "num_tokens": 54165784.0,
      "step": 29365
    },
    {
      "entropy": 5.469275379180909,
      "epoch": 2.46746481831548,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043883660924543325,
      "loss": 5.1516,
      "mean_token_accuracy": 0.1918512061238289,
      "num_tokens": 54174504.0,
      "step": 29370
    },
    {
      "entropy": 5.541509914398193,
      "epoch": 2.4678848981306447,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043881608230040685,
      "loss": 5.1816,
      "mean_token_accuracy": 0.18761757314205169,
      "num_tokens": 54183534.0,
      "step": 29375
    },
    {
      "entropy": 5.582364702224732,
      "epoch": 2.4683049779458095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000438795552453463,
      "loss": 5.2371,
      "mean_token_accuracy": 0.18538999408483506,
      "num_tokens": 54192101.0,
      "step": 29380
    },
    {
      "entropy": 5.542964315414428,
      "epoch": 2.4687250577609747,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004387750197049652,
      "loss": 5.1509,
      "mean_token_accuracy": 0.1827811136841774,
      "num_tokens": 54200771.0,
      "step": 29385
    },
    {
      "entropy": 5.525882720947266,
      "epoch": 2.4691451375761395,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004387544840552773,
      "loss": 5.1853,
      "mean_token_accuracy": 0.1836738705635071,
      "num_tokens": 54208816.0,
      "step": 29390
    },
    {
      "entropy": 5.568076133728027,
      "epoch": 2.4695652173913043,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043873394550476304,
      "loss": 5.183,
      "mean_token_accuracy": 0.18850998282432557,
      "num_tokens": 54218507.0,
      "step": 29395
    },
    {
      "entropy": 5.5765642642974855,
      "epoch": 2.469985297206469,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004387134040537862,
      "loss": 5.2198,
      "mean_token_accuracy": 0.1935688778758049,
      "num_tokens": 54227611.0,
      "step": 29400
    },
    {
      "entropy": 5.614648818969727,
      "epoch": 2.470405377021634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004386928597027108,
      "loss": 5.2438,
      "mean_token_accuracy": 0.18542157411575316,
      "num_tokens": 54237692.0,
      "step": 29405
    },
    {
      "entropy": 5.484953832626343,
      "epoch": 2.470825456836799,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004386723124519006,
      "loss": 5.11,
      "mean_token_accuracy": 0.1928548738360405,
      "num_tokens": 54247187.0,
      "step": 29410
    },
    {
      "entropy": 5.570834636688232,
      "epoch": 2.471245536651964,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043865176230171966,
      "loss": 5.257,
      "mean_token_accuracy": 0.1806565508246422,
      "num_tokens": 54256129.0,
      "step": 29415
    },
    {
      "entropy": 5.582585239410401,
      "epoch": 2.4716656164671287,
      "grad_norm": 1.078125,
      "learning_rate": 0.000438631209252532,
      "loss": 5.2344,
      "mean_token_accuracy": 0.18575143814086914,
      "num_tokens": 54265369.0,
      "step": 29420
    },
    {
      "entropy": 5.482168817520142,
      "epoch": 2.4720856962822935,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043861065330470187,
      "loss": 5.1444,
      "mean_token_accuracy": 0.18811811804771422,
      "num_tokens": 54274342.0,
      "step": 29425
    },
    {
      "entropy": 5.515036773681641,
      "epoch": 2.4725057760974587,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004385900944585931,
      "loss": 5.1917,
      "mean_token_accuracy": 0.18602475374937058,
      "num_tokens": 54283216.0,
      "step": 29430
    },
    {
      "entropy": 5.513940143585205,
      "epoch": 2.4729258559126235,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043856953271457014,
      "loss": 5.1655,
      "mean_token_accuracy": 0.1937987744808197,
      "num_tokens": 54291595.0,
      "step": 29435
    },
    {
      "entropy": 5.45925464630127,
      "epoch": 2.4733459357277883,
      "grad_norm": 1.125,
      "learning_rate": 0.0004385489680729971,
      "loss": 5.1274,
      "mean_token_accuracy": 0.1910024642944336,
      "num_tokens": 54300498.0,
      "step": 29440
    },
    {
      "entropy": 5.571644449234009,
      "epoch": 2.473766015542953,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004385284005342383,
      "loss": 5.1559,
      "mean_token_accuracy": 0.1896064028143883,
      "num_tokens": 54309929.0,
      "step": 29445
    },
    {
      "entropy": 5.570986795425415,
      "epoch": 2.4741860953581183,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004385078300986581,
      "loss": 5.2224,
      "mean_token_accuracy": 0.18318683058023452,
      "num_tokens": 54320440.0,
      "step": 29450
    },
    {
      "entropy": 5.588951110839844,
      "epoch": 2.474606175173283,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043848725676662087,
      "loss": 5.2159,
      "mean_token_accuracy": 0.18835197240114213,
      "num_tokens": 54329028.0,
      "step": 29455
    },
    {
      "entropy": 5.488767051696778,
      "epoch": 2.475026254988448,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004384666805384911,
      "loss": 5.1423,
      "mean_token_accuracy": 0.19273586720228195,
      "num_tokens": 54337986.0,
      "step": 29460
    },
    {
      "entropy": 5.559908866882324,
      "epoch": 2.4754463348036126,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004384461014146332,
      "loss": 5.2044,
      "mean_token_accuracy": 0.17797481566667556,
      "num_tokens": 54346135.0,
      "step": 29465
    },
    {
      "entropy": 5.521977424621582,
      "epoch": 2.4758664146187774,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004384255193954118,
      "loss": 5.1055,
      "mean_token_accuracy": 0.1984826534986496,
      "num_tokens": 54354914.0,
      "step": 29470
    },
    {
      "entropy": 5.564348220825195,
      "epoch": 2.4762864944339427,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004384049344811915,
      "loss": 5.1982,
      "mean_token_accuracy": 0.1781144544482231,
      "num_tokens": 54364697.0,
      "step": 29475
    },
    {
      "entropy": 5.543250989913941,
      "epoch": 2.4767065742491075,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004383843466723369,
      "loss": 5.1725,
      "mean_token_accuracy": 0.18636921644210816,
      "num_tokens": 54374196.0,
      "step": 29480
    },
    {
      "entropy": 5.575621557235718,
      "epoch": 2.4771266540642722,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043836375596921273,
      "loss": 5.2109,
      "mean_token_accuracy": 0.18940665125846862,
      "num_tokens": 54383708.0,
      "step": 29485
    },
    {
      "entropy": 5.603682184219361,
      "epoch": 2.477546733879437,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043834316237218373,
      "loss": 5.256,
      "mean_token_accuracy": 0.17799389213323594,
      "num_tokens": 54393906.0,
      "step": 29490
    },
    {
      "entropy": 5.6396044254302975,
      "epoch": 2.477966813694602,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004383225658816148,
      "loss": 5.1821,
      "mean_token_accuracy": 0.18947301357984542,
      "num_tokens": 54402640.0,
      "step": 29495
    },
    {
      "entropy": 5.504562759399414,
      "epoch": 2.478386893509767,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004383019664978707,
      "loss": 5.1079,
      "mean_token_accuracy": 0.18855546414852142,
      "num_tokens": 54412382.0,
      "step": 29500
    },
    {
      "entropy": 5.493575668334961,
      "epoch": 2.478806973324932,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043828136422131625,
      "loss": 5.098,
      "mean_token_accuracy": 0.1928175613284111,
      "num_tokens": 54421375.0,
      "step": 29505
    },
    {
      "entropy": 5.556856346130371,
      "epoch": 2.4792270531400966,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043826075905231655,
      "loss": 5.2728,
      "mean_token_accuracy": 0.19033616483211518,
      "num_tokens": 54430794.0,
      "step": 29510
    },
    {
      "entropy": 5.519356441497803,
      "epoch": 2.4796471329552614,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004382401509912366,
      "loss": 5.1556,
      "mean_token_accuracy": 0.18844982385635375,
      "num_tokens": 54440913.0,
      "step": 29515
    },
    {
      "entropy": 5.541060924530029,
      "epoch": 2.480067212770426,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004382195400384414,
      "loss": 5.2336,
      "mean_token_accuracy": 0.18486442863941194,
      "num_tokens": 54449978.0,
      "step": 29520
    },
    {
      "entropy": 5.56826605796814,
      "epoch": 2.4804872925855914,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043819892619429617,
      "loss": 5.1651,
      "mean_token_accuracy": 0.1897268548607826,
      "num_tokens": 54459187.0,
      "step": 29525
    },
    {
      "entropy": 5.597857093811035,
      "epoch": 2.480907372400756,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004381783094591659,
      "loss": 5.3547,
      "mean_token_accuracy": 0.18008258938789368,
      "num_tokens": 54468354.0,
      "step": 29530
    },
    {
      "entropy": 5.576054906845092,
      "epoch": 2.481327452215921,
      "grad_norm": 1.09375,
      "learning_rate": 0.000438157689833416,
      "loss": 5.2307,
      "mean_token_accuracy": 0.18333393931388856,
      "num_tokens": 54478331.0,
      "step": 29535
    },
    {
      "entropy": 5.496292304992676,
      "epoch": 2.481747532031086,
      "grad_norm": 1.0,
      "learning_rate": 0.00043813706731741153,
      "loss": 5.1676,
      "mean_token_accuracy": 0.1926803395152092,
      "num_tokens": 54488243.0,
      "step": 29540
    },
    {
      "entropy": 5.532199096679688,
      "epoch": 2.4821676118462506,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043811644191151796,
      "loss": 5.0913,
      "mean_token_accuracy": 0.1853126809000969,
      "num_tokens": 54496985.0,
      "step": 29545
    },
    {
      "entropy": 5.492210817337036,
      "epoch": 2.482587691661416,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004380958136161006,
      "loss": 5.2335,
      "mean_token_accuracy": 0.18056112825870513,
      "num_tokens": 54506012.0,
      "step": 29550
    },
    {
      "entropy": 5.631212949752808,
      "epoch": 2.4830077714765806,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004380751824315249,
      "loss": 5.2802,
      "mean_token_accuracy": 0.18105464726686477,
      "num_tokens": 54514883.0,
      "step": 29555
    },
    {
      "entropy": 5.493071985244751,
      "epoch": 2.4834278512917454,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043805454835815636,
      "loss": 5.1503,
      "mean_token_accuracy": 0.1863706275820732,
      "num_tokens": 54523590.0,
      "step": 29560
    },
    {
      "entropy": 5.445086193084717,
      "epoch": 2.48384793110691,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004380339113963604,
      "loss": 5.0725,
      "mean_token_accuracy": 0.1941179573535919,
      "num_tokens": 54532696.0,
      "step": 29565
    },
    {
      "entropy": 5.489297676086426,
      "epoch": 2.4842680109220754,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004380132715465027,
      "loss": 5.1411,
      "mean_token_accuracy": 0.1945464640855789,
      "num_tokens": 54541792.0,
      "step": 29570
    },
    {
      "entropy": 5.496041870117187,
      "epoch": 2.48468809073724,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043799262880894876,
      "loss": 5.1638,
      "mean_token_accuracy": 0.19377403259277343,
      "num_tokens": 54550625.0,
      "step": 29575
    },
    {
      "entropy": 5.576722192764282,
      "epoch": 2.485108170552405,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004379719831840644,
      "loss": 5.2518,
      "mean_token_accuracy": 0.18258660435676574,
      "num_tokens": 54560490.0,
      "step": 29580
    },
    {
      "entropy": 5.6214742183685305,
      "epoch": 2.4855282503675697,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004379513346722152,
      "loss": 5.2952,
      "mean_token_accuracy": 0.18141794800758362,
      "num_tokens": 54569897.0,
      "step": 29585
    },
    {
      "entropy": 5.643501663208008,
      "epoch": 2.4859483301827345,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004379306832737671,
      "loss": 5.2648,
      "mean_token_accuracy": 0.1751942291855812,
      "num_tokens": 54579711.0,
      "step": 29590
    },
    {
      "entropy": 5.447202110290528,
      "epoch": 2.4863684099978998,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004379100289890858,
      "loss": 5.1212,
      "mean_token_accuracy": 0.19162269681692123,
      "num_tokens": 54589354.0,
      "step": 29595
    },
    {
      "entropy": 5.5553652286529545,
      "epoch": 2.4867884898130646,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043788937181853723,
      "loss": 5.1841,
      "mean_token_accuracy": 0.18601470738649367,
      "num_tokens": 54598575.0,
      "step": 29600
    },
    {
      "entropy": 5.4052205085754395,
      "epoch": 2.4872085696282293,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004378687117624873,
      "loss": 5.0278,
      "mean_token_accuracy": 0.19140967279672622,
      "num_tokens": 54608014.0,
      "step": 29605
    },
    {
      "entropy": 5.4757280349731445,
      "epoch": 2.487628649443394,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043784804882130205,
      "loss": 5.1433,
      "mean_token_accuracy": 0.1896376609802246,
      "num_tokens": 54617300.0,
      "step": 29610
    },
    {
      "entropy": 5.560055780410766,
      "epoch": 2.4880487292585594,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004378273829953474,
      "loss": 5.1658,
      "mean_token_accuracy": 0.18210013806819916,
      "num_tokens": 54626777.0,
      "step": 29615
    },
    {
      "entropy": 5.482356548309326,
      "epoch": 2.488468809073724,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043780671428498945,
      "loss": 5.1239,
      "mean_token_accuracy": 0.18974536806344985,
      "num_tokens": 54636357.0,
      "step": 29620
    },
    {
      "entropy": 5.51355013847351,
      "epoch": 2.488888888888889,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043778604269059456,
      "loss": 5.0671,
      "mean_token_accuracy": 0.19305214881896973,
      "num_tokens": 54645126.0,
      "step": 29625
    },
    {
      "entropy": 5.597341012954712,
      "epoch": 2.4893089687040537,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043776536821252867,
      "loss": 5.2673,
      "mean_token_accuracy": 0.18975103348493577,
      "num_tokens": 54653920.0,
      "step": 29630
    },
    {
      "entropy": 5.606925344467163,
      "epoch": 2.4897290485192185,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000437744690851158,
      "loss": 5.224,
      "mean_token_accuracy": 0.18374820649623871,
      "num_tokens": 54663627.0,
      "step": 29635
    },
    {
      "entropy": 5.585060071945191,
      "epoch": 2.4901491283343837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043772401060684907,
      "loss": 5.2603,
      "mean_token_accuracy": 0.18325006812810898,
      "num_tokens": 54673741.0,
      "step": 29640
    },
    {
      "entropy": 5.580456876754761,
      "epoch": 2.4905692081495485,
      "grad_norm": 1.125,
      "learning_rate": 0.000437703327479968,
      "loss": 5.2091,
      "mean_token_accuracy": 0.17942784279584884,
      "num_tokens": 54683387.0,
      "step": 29645
    },
    {
      "entropy": 5.590901279449463,
      "epoch": 2.4909892879647133,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043768264147088135,
      "loss": 5.302,
      "mean_token_accuracy": 0.17858352065086364,
      "num_tokens": 54693807.0,
      "step": 29650
    },
    {
      "entropy": 5.535178995132446,
      "epoch": 2.491409367779878,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004376619525799554,
      "loss": 5.2001,
      "mean_token_accuracy": 0.1806879758834839,
      "num_tokens": 54703001.0,
      "step": 29655
    },
    {
      "entropy": 5.5171606063842775,
      "epoch": 2.491829447595043,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004376412608075567,
      "loss": 5.0821,
      "mean_token_accuracy": 0.19843075275421143,
      "num_tokens": 54712863.0,
      "step": 29660
    },
    {
      "entropy": 5.520681858062744,
      "epoch": 2.492249527410208,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043762056615405186,
      "loss": 5.1736,
      "mean_token_accuracy": 0.18493476063013076,
      "num_tokens": 54721784.0,
      "step": 29665
    },
    {
      "entropy": 5.535509443283081,
      "epoch": 2.492669607225373,
      "grad_norm": 1.125,
      "learning_rate": 0.0004375998686198074,
      "loss": 5.2234,
      "mean_token_accuracy": 0.1853698670864105,
      "num_tokens": 54730961.0,
      "step": 29670
    },
    {
      "entropy": 5.488910913467407,
      "epoch": 2.4930896870405377,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043757916820519,
      "loss": 5.1791,
      "mean_token_accuracy": 0.19193556755781174,
      "num_tokens": 54740124.0,
      "step": 29675
    },
    {
      "entropy": 5.519402503967285,
      "epoch": 2.4935097668557025,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004375584649105664,
      "loss": 5.1846,
      "mean_token_accuracy": 0.18220505267381668,
      "num_tokens": 54749486.0,
      "step": 29680
    },
    {
      "entropy": 5.53220796585083,
      "epoch": 2.4939298466708673,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004375377587363033,
      "loss": 5.1921,
      "mean_token_accuracy": 0.1875484734773636,
      "num_tokens": 54759167.0,
      "step": 29685
    },
    {
      "entropy": 5.487993574142456,
      "epoch": 2.4943499264860325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043751704968276745,
      "loss": 5.128,
      "mean_token_accuracy": 0.1980043426156044,
      "num_tokens": 54767403.0,
      "step": 29690
    },
    {
      "entropy": 5.451072454452515,
      "epoch": 2.4947700063011973,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004374963377503257,
      "loss": 5.1794,
      "mean_token_accuracy": 0.18114737123250962,
      "num_tokens": 54777373.0,
      "step": 29695
    },
    {
      "entropy": 5.5383893013000485,
      "epoch": 2.495190086116362,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043747562293934507,
      "loss": 5.1764,
      "mean_token_accuracy": 0.18932260423898697,
      "num_tokens": 54786082.0,
      "step": 29700
    },
    {
      "entropy": 5.590947580337525,
      "epoch": 2.495610165931527,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043745490525019245,
      "loss": 5.2073,
      "mean_token_accuracy": 0.18346758633852006,
      "num_tokens": 54794879.0,
      "step": 29705
    },
    {
      "entropy": 5.549365091323852,
      "epoch": 2.4960302457466916,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004374341846832348,
      "loss": 5.188,
      "mean_token_accuracy": 0.1844865545630455,
      "num_tokens": 54804095.0,
      "step": 29710
    },
    {
      "entropy": 5.524419641494751,
      "epoch": 2.496450325561857,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043741346123883927,
      "loss": 5.205,
      "mean_token_accuracy": 0.19380844980478287,
      "num_tokens": 54813135.0,
      "step": 29715
    },
    {
      "entropy": 5.509318733215332,
      "epoch": 2.4968704053770217,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004373927349173728,
      "loss": 5.1737,
      "mean_token_accuracy": 0.19625871628522873,
      "num_tokens": 54822129.0,
      "step": 29720
    },
    {
      "entropy": 5.609893465042115,
      "epoch": 2.4972904851921864,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004373720057192028,
      "loss": 5.2756,
      "mean_token_accuracy": 0.17780711352825165,
      "num_tokens": 54831342.0,
      "step": 29725
    },
    {
      "entropy": 5.529437017440796,
      "epoch": 2.4977105650073512,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004373512736446962,
      "loss": 5.1934,
      "mean_token_accuracy": 0.18865105807781218,
      "num_tokens": 54840327.0,
      "step": 29730
    },
    {
      "entropy": 5.55469970703125,
      "epoch": 2.4981306448225165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043733053869422046,
      "loss": 5.1777,
      "mean_token_accuracy": 0.1894258365035057,
      "num_tokens": 54849259.0,
      "step": 29735
    },
    {
      "entropy": 5.4866105079650875,
      "epoch": 2.4985507246376812,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004373098008681428,
      "loss": 5.0766,
      "mean_token_accuracy": 0.18916652649641036,
      "num_tokens": 54858217.0,
      "step": 29740
    },
    {
      "entropy": 5.536677360534668,
      "epoch": 2.498970804452846,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004372890601668305,
      "loss": 5.135,
      "mean_token_accuracy": 0.1903023138642311,
      "num_tokens": 54867256.0,
      "step": 29745
    },
    {
      "entropy": 5.54814133644104,
      "epoch": 2.499390884268011,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043726831659065114,
      "loss": 5.2048,
      "mean_token_accuracy": 0.1896450161933899,
      "num_tokens": 54875621.0,
      "step": 29750
    },
    {
      "entropy": 5.492396450042724,
      "epoch": 2.499810964083176,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004372475701399721,
      "loss": 5.0952,
      "mean_token_accuracy": 0.19565211236476898,
      "num_tokens": 54884153.0,
      "step": 29755
    },
    {
      "entropy": 5.471549654006958,
      "epoch": 2.500231043898341,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043722682081516093,
      "loss": 5.0635,
      "mean_token_accuracy": 0.19340321272611619,
      "num_tokens": 54892800.0,
      "step": 29760
    },
    {
      "entropy": 5.351401948928833,
      "epoch": 2.5006511237135056,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004372060686165851,
      "loss": 5.0373,
      "mean_token_accuracy": 0.1978205546736717,
      "num_tokens": 54902083.0,
      "step": 29765
    },
    {
      "entropy": 5.541549015045166,
      "epoch": 2.5010712035286704,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043718531354461235,
      "loss": 5.1847,
      "mean_token_accuracy": 0.18841346949338914,
      "num_tokens": 54911733.0,
      "step": 29770
    },
    {
      "entropy": 5.577776670455933,
      "epoch": 2.501491283343835,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004371645555996103,
      "loss": 5.2931,
      "mean_token_accuracy": 0.18347054123878478,
      "num_tokens": 54921835.0,
      "step": 29775
    },
    {
      "entropy": 5.471684503555298,
      "epoch": 2.5019113631590004,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004371437947819466,
      "loss": 5.0683,
      "mean_token_accuracy": 0.1932407632470131,
      "num_tokens": 54930452.0,
      "step": 29780
    },
    {
      "entropy": 5.517944765090943,
      "epoch": 2.502331442974165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043712303109198905,
      "loss": 5.1969,
      "mean_token_accuracy": 0.18487227857112884,
      "num_tokens": 54938989.0,
      "step": 29785
    },
    {
      "entropy": 5.484797382354737,
      "epoch": 2.50275152278933,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043710226453010553,
      "loss": 5.1184,
      "mean_token_accuracy": 0.18576253056526185,
      "num_tokens": 54948595.0,
      "step": 29790
    },
    {
      "entropy": 5.503661060333252,
      "epoch": 2.503171602604495,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043708149509666383,
      "loss": 5.2143,
      "mean_token_accuracy": 0.1813365787267685,
      "num_tokens": 54958561.0,
      "step": 29795
    },
    {
      "entropy": 5.526693058013916,
      "epoch": 2.5035916824196596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000437060722792032,
      "loss": 5.2254,
      "mean_token_accuracy": 0.18174044489860536,
      "num_tokens": 54968515.0,
      "step": 29800
    },
    {
      "entropy": 5.528715753555298,
      "epoch": 2.504011762234825,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043703994761657783,
      "loss": 5.0919,
      "mean_token_accuracy": 0.1925382360816002,
      "num_tokens": 54977645.0,
      "step": 29805
    },
    {
      "entropy": 5.566304874420166,
      "epoch": 2.5044318420499896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004370191695706694,
      "loss": 5.2797,
      "mean_token_accuracy": 0.18142885863780975,
      "num_tokens": 54987842.0,
      "step": 29810
    },
    {
      "entropy": 5.538144445419311,
      "epoch": 2.5048519218651544,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004369983886546748,
      "loss": 5.2273,
      "mean_token_accuracy": 0.18121739625930786,
      "num_tokens": 54997182.0,
      "step": 29815
    },
    {
      "entropy": 5.457459878921509,
      "epoch": 2.505272001680319,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004369776048689622,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19366969913244247,
      "num_tokens": 55005740.0,
      "step": 29820
    },
    {
      "entropy": 5.513277149200439,
      "epoch": 2.505692081495484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004369568182138998,
      "loss": 5.1689,
      "mean_token_accuracy": 0.18846119642257692,
      "num_tokens": 55015130.0,
      "step": 29825
    },
    {
      "entropy": 5.530158042907715,
      "epoch": 2.506112161310649,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043693602868985563,
      "loss": 5.1898,
      "mean_token_accuracy": 0.18987253308296204,
      "num_tokens": 55025206.0,
      "step": 29830
    },
    {
      "entropy": 5.5579041004180905,
      "epoch": 2.506532241125814,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004369152362971982,
      "loss": 5.1312,
      "mean_token_accuracy": 0.19291565865278243,
      "num_tokens": 55034258.0,
      "step": 29835
    },
    {
      "entropy": 5.536284923553467,
      "epoch": 2.5069523209409788,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004368944410362957,
      "loss": 5.1715,
      "mean_token_accuracy": 0.18790233582258226,
      "num_tokens": 55042732.0,
      "step": 29840
    },
    {
      "entropy": 5.507989311218262,
      "epoch": 2.5073724007561435,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004368736429075165,
      "loss": 5.2061,
      "mean_token_accuracy": 0.18181175589561463,
      "num_tokens": 55051728.0,
      "step": 29845
    },
    {
      "entropy": 5.447445869445801,
      "epoch": 2.5077924805713083,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043685284191122915,
      "loss": 5.1441,
      "mean_token_accuracy": 0.19600435197353364,
      "num_tokens": 55060509.0,
      "step": 29850
    },
    {
      "entropy": 5.53816180229187,
      "epoch": 2.5082125603864736,
      "grad_norm": 1.125,
      "learning_rate": 0.000436832038047802,
      "loss": 5.2201,
      "mean_token_accuracy": 0.1850486323237419,
      "num_tokens": 55069637.0,
      "step": 29855
    },
    {
      "entropy": 5.564800453186035,
      "epoch": 2.5086326402016383,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004368112313176037,
      "loss": 5.1952,
      "mean_token_accuracy": 0.18164317160844803,
      "num_tokens": 55079547.0,
      "step": 29860
    },
    {
      "entropy": 5.548403644561768,
      "epoch": 2.509052720016803,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004367904217210026,
      "loss": 5.1307,
      "mean_token_accuracy": 0.19530919194221497,
      "num_tokens": 55088126.0,
      "step": 29865
    },
    {
      "entropy": 5.543561697006226,
      "epoch": 2.509472799831968,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004367696092583676,
      "loss": 5.1625,
      "mean_token_accuracy": 0.18727702647447586,
      "num_tokens": 55097458.0,
      "step": 29870
    },
    {
      "entropy": 5.530220556259155,
      "epoch": 2.5098928796471327,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004367487939300673,
      "loss": 5.1366,
      "mean_token_accuracy": 0.19094910621643066,
      "num_tokens": 55105969.0,
      "step": 29875
    },
    {
      "entropy": 5.445891427993774,
      "epoch": 2.510312959462298,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043672797573647033,
      "loss": 5.0945,
      "mean_token_accuracy": 0.1946832299232483,
      "num_tokens": 55114985.0,
      "step": 29880
    },
    {
      "entropy": 5.529201555252075,
      "epoch": 2.5107330392774627,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004367071546779456,
      "loss": 5.1344,
      "mean_token_accuracy": 0.18974451571702958,
      "num_tokens": 55123576.0,
      "step": 29885
    },
    {
      "entropy": 5.559971475601197,
      "epoch": 2.5111531190926275,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004366863307548618,
      "loss": 5.13,
      "mean_token_accuracy": 0.19292028695344926,
      "num_tokens": 55131758.0,
      "step": 29890
    },
    {
      "entropy": 5.486197280883789,
      "epoch": 2.5115731989077927,
      "grad_norm": 1.078125,
      "learning_rate": 0.000436665503967588,
      "loss": 5.1552,
      "mean_token_accuracy": 0.19017743915319443,
      "num_tokens": 55141288.0,
      "step": 29895
    },
    {
      "entropy": 5.573207902908325,
      "epoch": 2.5119932787229575,
      "grad_norm": 1.171875,
      "learning_rate": 0.000436644674316493,
      "loss": 5.2669,
      "mean_token_accuracy": 0.1804138958454132,
      "num_tokens": 55150834.0,
      "step": 29900
    },
    {
      "entropy": 5.524005603790283,
      "epoch": 2.5124133585381223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043662384180194583,
      "loss": 5.1301,
      "mean_token_accuracy": 0.19026567190885543,
      "num_tokens": 55159581.0,
      "step": 29905
    },
    {
      "entropy": 5.463131332397461,
      "epoch": 2.512833438353287,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043660300642431554,
      "loss": 5.1183,
      "mean_token_accuracy": 0.19108158051967622,
      "num_tokens": 55167740.0,
      "step": 29910
    },
    {
      "entropy": 5.5488427639007565,
      "epoch": 2.513253518168452,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043658216818397123,
      "loss": 5.2228,
      "mean_token_accuracy": 0.19210165143013,
      "num_tokens": 55176788.0,
      "step": 29915
    },
    {
      "entropy": 5.496541547775268,
      "epoch": 2.513673597983617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000436561327081282,
      "loss": 5.0968,
      "mean_token_accuracy": 0.19238279461860658,
      "num_tokens": 55185234.0,
      "step": 29920
    },
    {
      "entropy": 5.487098455429077,
      "epoch": 2.514093677798782,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043654048311661704,
      "loss": 5.1453,
      "mean_token_accuracy": 0.19226481914520263,
      "num_tokens": 55193784.0,
      "step": 29925
    },
    {
      "entropy": 5.579278898239136,
      "epoch": 2.5145137576139467,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043651963629034557,
      "loss": 5.273,
      "mean_token_accuracy": 0.17708095759153367,
      "num_tokens": 55203271.0,
      "step": 29930
    },
    {
      "entropy": 5.523320245742798,
      "epoch": 2.5149338374291115,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004364987866028369,
      "loss": 5.1103,
      "mean_token_accuracy": 0.1901486724615097,
      "num_tokens": 55211888.0,
      "step": 29935
    },
    {
      "entropy": 5.435547924041748,
      "epoch": 2.5153539172442763,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004364779340544605,
      "loss": 5.1765,
      "mean_token_accuracy": 0.18238507807254792,
      "num_tokens": 55221261.0,
      "step": 29940
    },
    {
      "entropy": 5.612782573699951,
      "epoch": 2.5157739970594415,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043645707864558555,
      "loss": 5.2516,
      "mean_token_accuracy": 0.18088991791009904,
      "num_tokens": 55229792.0,
      "step": 29945
    },
    {
      "entropy": 5.572704410552978,
      "epoch": 2.5161940768746063,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043643622037658163,
      "loss": 5.2098,
      "mean_token_accuracy": 0.18706173300743104,
      "num_tokens": 55238908.0,
      "step": 29950
    },
    {
      "entropy": 5.499961614608765,
      "epoch": 2.516614156689771,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004364153592478182,
      "loss": 5.1464,
      "mean_token_accuracy": 0.18830306380987166,
      "num_tokens": 55248305.0,
      "step": 29955
    },
    {
      "entropy": 5.626817464828491,
      "epoch": 2.517034236504936,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004363944952596647,
      "loss": 5.2551,
      "mean_token_accuracy": 0.18370186686515808,
      "num_tokens": 55257430.0,
      "step": 29960
    },
    {
      "entropy": 5.521985387802124,
      "epoch": 2.5174543163201006,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004363736284124909,
      "loss": 5.1308,
      "mean_token_accuracy": 0.19426779001951217,
      "num_tokens": 55266088.0,
      "step": 29965
    },
    {
      "entropy": 5.491559028625488,
      "epoch": 2.517874396135266,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043635275870666643,
      "loss": 5.1707,
      "mean_token_accuracy": 0.188088496029377,
      "num_tokens": 55276530.0,
      "step": 29970
    },
    {
      "entropy": 5.520514631271363,
      "epoch": 2.5182944759504307,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043633188614256084,
      "loss": 5.1793,
      "mean_token_accuracy": 0.18852411806583405,
      "num_tokens": 55286465.0,
      "step": 29975
    },
    {
      "entropy": 5.46496434211731,
      "epoch": 2.5187145557655954,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043631101072054396,
      "loss": 5.0912,
      "mean_token_accuracy": 0.1926356926560402,
      "num_tokens": 55295577.0,
      "step": 29980
    },
    {
      "entropy": 5.489986610412598,
      "epoch": 2.5191346355807602,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004362901324409856,
      "loss": 5.1554,
      "mean_token_accuracy": 0.19067124724388124,
      "num_tokens": 55304201.0,
      "step": 29985
    },
    {
      "entropy": 5.505571699142456,
      "epoch": 2.519554715395925,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004362692513042556,
      "loss": 5.157,
      "mean_token_accuracy": 0.19214553087949754,
      "num_tokens": 55313269.0,
      "step": 29990
    },
    {
      "entropy": 5.497665500640869,
      "epoch": 2.5199747952110902,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043624836731072386,
      "loss": 5.1818,
      "mean_token_accuracy": 0.18502172976732253,
      "num_tokens": 55322301.0,
      "step": 29995
    },
    {
      "entropy": 5.521495389938354,
      "epoch": 2.520394875026255,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004362274804607604,
      "loss": 5.1428,
      "mean_token_accuracy": 0.18965448588132858,
      "num_tokens": 55331764.0,
      "step": 30000
    },
    {
      "epoch": 2.520394875026255,
      "eval_entropy": 5.3145340982991245,
      "eval_loss": 5.252910614013672,
      "eval_mean_token_accuracy": 0.19286997219835939,
      "eval_num_tokens": 55331764.0,
      "eval_runtime": 27.356,
      "eval_samples_per_second": 1365.915,
      "eval_steps_per_second": 170.748,
      "step": 30000
    },
    {
      "entropy": 5.540525197982788,
      "epoch": 2.52081495484142,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043620659075473497,
      "loss": 5.1422,
      "mean_token_accuracy": 0.18727703392505646,
      "num_tokens": 55340874.0,
      "step": 30005
    },
    {
      "entropy": 5.503339958190918,
      "epoch": 2.5212350346565846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004361856981930179,
      "loss": 5.1106,
      "mean_token_accuracy": 0.19143618792295455,
      "num_tokens": 55349846.0,
      "step": 30010
    },
    {
      "entropy": 5.5215651512146,
      "epoch": 2.5216551144717494,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004361648027759792,
      "loss": 5.1389,
      "mean_token_accuracy": 0.1902594655752182,
      "num_tokens": 55357865.0,
      "step": 30015
    },
    {
      "entropy": 5.4606170654296875,
      "epoch": 2.5220751942869146,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043614390450398896,
      "loss": 5.1971,
      "mean_token_accuracy": 0.19358139485120773,
      "num_tokens": 55367096.0,
      "step": 30020
    },
    {
      "entropy": 5.608029413223266,
      "epoch": 2.5224952741020794,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043612300337741744,
      "loss": 5.2503,
      "mean_token_accuracy": 0.18941752165555953,
      "num_tokens": 55376557.0,
      "step": 30025
    },
    {
      "entropy": 5.570345830917359,
      "epoch": 2.522915353917244,
      "grad_norm": 1.125,
      "learning_rate": 0.00043610209939663486,
      "loss": 5.1903,
      "mean_token_accuracy": 0.1908656746149063,
      "num_tokens": 55385336.0,
      "step": 30030
    },
    {
      "entropy": 5.452816486358643,
      "epoch": 2.5233354337324094,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004360811925620117,
      "loss": 5.1181,
      "mean_token_accuracy": 0.19138784259557723,
      "num_tokens": 55393349.0,
      "step": 30035
    },
    {
      "entropy": 5.569668102264404,
      "epoch": 2.5237555135475738,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043606028287391797,
      "loss": 5.2432,
      "mean_token_accuracy": 0.1817514941096306,
      "num_tokens": 55402964.0,
      "step": 30040
    },
    {
      "entropy": 5.458809566497803,
      "epoch": 2.524175593362739,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00043603937033272433,
      "loss": 5.1253,
      "mean_token_accuracy": 0.20039294809103012,
      "num_tokens": 55412861.0,
      "step": 30045
    },
    {
      "entropy": 5.463651704788208,
      "epoch": 2.524595673177904,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043601845493880113,
      "loss": 5.1672,
      "mean_token_accuracy": 0.1891668513417244,
      "num_tokens": 55421922.0,
      "step": 30050
    },
    {
      "entropy": 5.562634229660034,
      "epoch": 2.5250157529930686,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000435997536692519,
      "loss": 5.193,
      "mean_token_accuracy": 0.1790940836071968,
      "num_tokens": 55432043.0,
      "step": 30055
    },
    {
      "entropy": 5.572760152816772,
      "epoch": 2.525435832808234,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043597661559424835,
      "loss": 5.127,
      "mean_token_accuracy": 0.19014683216810227,
      "num_tokens": 55440659.0,
      "step": 30060
    },
    {
      "entropy": 5.411213397979736,
      "epoch": 2.5258559126233986,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004359556916443598,
      "loss": 5.0826,
      "mean_token_accuracy": 0.19399195164442062,
      "num_tokens": 55449680.0,
      "step": 30065
    },
    {
      "entropy": 5.5541685104370115,
      "epoch": 2.5262759924385634,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043593476484322415,
      "loss": 5.2485,
      "mean_token_accuracy": 0.18634578436613083,
      "num_tokens": 55459591.0,
      "step": 30070
    },
    {
      "entropy": 5.60334210395813,
      "epoch": 2.526696072253728,
      "grad_norm": 1.25,
      "learning_rate": 0.00043591383519121194,
      "loss": 5.2319,
      "mean_token_accuracy": 0.1825944647192955,
      "num_tokens": 55468921.0,
      "step": 30075
    },
    {
      "entropy": 5.5450492858886715,
      "epoch": 2.527116152068893,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043589290268869404,
      "loss": 5.2432,
      "mean_token_accuracy": 0.18597447276115417,
      "num_tokens": 55479077.0,
      "step": 30080
    },
    {
      "entropy": 5.5354760646820065,
      "epoch": 2.527536231884058,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004358719673360412,
      "loss": 5.1643,
      "mean_token_accuracy": 0.18760695010423661,
      "num_tokens": 55488098.0,
      "step": 30085
    },
    {
      "entropy": 5.5415184020996096,
      "epoch": 2.527956311699223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043585102913362424,
      "loss": 5.1671,
      "mean_token_accuracy": 0.1909402072429657,
      "num_tokens": 55497096.0,
      "step": 30090
    },
    {
      "entropy": 5.5192924499511715,
      "epoch": 2.5283763915143878,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043583008808181426,
      "loss": 5.1503,
      "mean_token_accuracy": 0.19068965762853624,
      "num_tokens": 55506612.0,
      "step": 30095
    },
    {
      "entropy": 5.4712005138397215,
      "epoch": 2.5287964713295525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043580914418098193,
      "loss": 5.1528,
      "mean_token_accuracy": 0.1890902817249298,
      "num_tokens": 55515660.0,
      "step": 30100
    },
    {
      "entropy": 5.502286291122436,
      "epoch": 2.5292165511447173,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004357881974314985,
      "loss": 5.1047,
      "mean_token_accuracy": 0.1952556148171425,
      "num_tokens": 55523935.0,
      "step": 30105
    },
    {
      "entropy": 5.536916017532349,
      "epoch": 2.5296366309598826,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004357672478337349,
      "loss": 5.1751,
      "mean_token_accuracy": 0.19286748319864272,
      "num_tokens": 55533209.0,
      "step": 30110
    },
    {
      "entropy": 5.499720048904419,
      "epoch": 2.5300567107750473,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004357462953880623,
      "loss": 5.0583,
      "mean_token_accuracy": 0.19382499754428864,
      "num_tokens": 55541954.0,
      "step": 30115
    },
    {
      "entropy": 5.477565622329712,
      "epoch": 2.530476790590212,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043572534009485195,
      "loss": 5.1512,
      "mean_token_accuracy": 0.18834263384342192,
      "num_tokens": 55551342.0,
      "step": 30120
    },
    {
      "entropy": 5.536240959167481,
      "epoch": 2.530896870405377,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043570438195447486,
      "loss": 5.1286,
      "mean_token_accuracy": 0.18240986168384551,
      "num_tokens": 55560105.0,
      "step": 30125
    },
    {
      "entropy": 5.523880195617676,
      "epoch": 2.5313169502205417,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043568342096730244,
      "loss": 5.2101,
      "mean_token_accuracy": 0.18494227677583694,
      "num_tokens": 55569594.0,
      "step": 30130
    },
    {
      "entropy": 5.53792405128479,
      "epoch": 2.531737030035707,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004356624571337059,
      "loss": 5.107,
      "mean_token_accuracy": 0.1896895706653595,
      "num_tokens": 55578442.0,
      "step": 30135
    },
    {
      "entropy": 5.451293516159057,
      "epoch": 2.5321571098508717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004356414904540567,
      "loss": 5.1069,
      "mean_token_accuracy": 0.18847065269947053,
      "num_tokens": 55587823.0,
      "step": 30140
    },
    {
      "entropy": 5.478387022018433,
      "epoch": 2.5325771896660365,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004356205209287262,
      "loss": 5.1294,
      "mean_token_accuracy": 0.19465380311012268,
      "num_tokens": 55597303.0,
      "step": 30145
    },
    {
      "entropy": 5.5038557052612305,
      "epoch": 2.5329972694812013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043559954855808595,
      "loss": 5.1788,
      "mean_token_accuracy": 0.18327670246362687,
      "num_tokens": 55606273.0,
      "step": 30150
    },
    {
      "entropy": 5.568985176086426,
      "epoch": 2.533417349296366,
      "grad_norm": 1.125,
      "learning_rate": 0.0004355785733425072,
      "loss": 5.2066,
      "mean_token_accuracy": 0.186585134267807,
      "num_tokens": 55615537.0,
      "step": 30155
    },
    {
      "entropy": 5.509389209747314,
      "epoch": 2.5338374291115313,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043555759528236186,
      "loss": 5.1123,
      "mean_token_accuracy": 0.18591154515743255,
      "num_tokens": 55624682.0,
      "step": 30160
    },
    {
      "entropy": 5.534445714950562,
      "epoch": 2.534257508926696,
      "grad_norm": 1.125,
      "learning_rate": 0.00043553661437802137,
      "loss": 5.1799,
      "mean_token_accuracy": 0.18333360105752944,
      "num_tokens": 55634684.0,
      "step": 30165
    },
    {
      "entropy": 5.574244165420533,
      "epoch": 2.534677588741861,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043551563062985745,
      "loss": 5.1863,
      "mean_token_accuracy": 0.19406710416078568,
      "num_tokens": 55644141.0,
      "step": 30170
    },
    {
      "entropy": 5.455415105819702,
      "epoch": 2.5350976685570257,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004354946440382418,
      "loss": 5.1494,
      "mean_token_accuracy": 0.20083874613046646,
      "num_tokens": 55653916.0,
      "step": 30175
    },
    {
      "entropy": 5.587460613250732,
      "epoch": 2.5355177483721905,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004354736546035461,
      "loss": 5.3034,
      "mean_token_accuracy": 0.17791556417942048,
      "num_tokens": 55664125.0,
      "step": 30180
    },
    {
      "entropy": 5.522809314727783,
      "epoch": 2.5359378281873557,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004354526623261423,
      "loss": 5.1306,
      "mean_token_accuracy": 0.1951425164937973,
      "num_tokens": 55673647.0,
      "step": 30185
    },
    {
      "entropy": 5.549458265304565,
      "epoch": 2.5363579080025205,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043543166720640223,
      "loss": 5.1973,
      "mean_token_accuracy": 0.18662489652633668,
      "num_tokens": 55683422.0,
      "step": 30190
    },
    {
      "entropy": 5.594909715652466,
      "epoch": 2.5367779878176853,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004354106692446977,
      "loss": 5.1916,
      "mean_token_accuracy": 0.1871662139892578,
      "num_tokens": 55694177.0,
      "step": 30195
    },
    {
      "entropy": 5.516188049316407,
      "epoch": 2.5371980676328505,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043538966844140076,
      "loss": 5.2321,
      "mean_token_accuracy": 0.1902655065059662,
      "num_tokens": 55703534.0,
      "step": 30200
    },
    {
      "entropy": 5.504040670394898,
      "epoch": 2.5376181474480153,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004353686647968835,
      "loss": 5.1249,
      "mean_token_accuracy": 0.18925627022981645,
      "num_tokens": 55712920.0,
      "step": 30205
    },
    {
      "entropy": 5.483880996704102,
      "epoch": 2.53803822726318,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004353476583115179,
      "loss": 5.1725,
      "mean_token_accuracy": 0.18093796521425248,
      "num_tokens": 55721706.0,
      "step": 30210
    },
    {
      "entropy": 5.584664630889892,
      "epoch": 2.538458307078345,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004353266489856761,
      "loss": 5.2194,
      "mean_token_accuracy": 0.19300937354564668,
      "num_tokens": 55730626.0,
      "step": 30215
    },
    {
      "entropy": 5.582037162780762,
      "epoch": 2.5388783868935096,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043530563681973023,
      "loss": 5.2148,
      "mean_token_accuracy": 0.18306079804897307,
      "num_tokens": 55740044.0,
      "step": 30220
    },
    {
      "entropy": 5.5028753757476805,
      "epoch": 2.539298466708675,
      "grad_norm": 1.125,
      "learning_rate": 0.0004352846218140526,
      "loss": 5.0275,
      "mean_token_accuracy": 0.1938682898879051,
      "num_tokens": 55748907.0,
      "step": 30225
    },
    {
      "entropy": 5.505010366439819,
      "epoch": 2.5397185465238397,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004352636039690154,
      "loss": 5.1446,
      "mean_token_accuracy": 0.1888437956571579,
      "num_tokens": 55758011.0,
      "step": 30230
    },
    {
      "entropy": 5.46423077583313,
      "epoch": 2.5401386263390044,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000435242583284991,
      "loss": 5.1225,
      "mean_token_accuracy": 0.1931807890534401,
      "num_tokens": 55766937.0,
      "step": 30235
    },
    {
      "entropy": 5.586774015426636,
      "epoch": 2.5405587061541692,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043522155976235177,
      "loss": 5.2585,
      "mean_token_accuracy": 0.18143410384654998,
      "num_tokens": 55776160.0,
      "step": 30240
    },
    {
      "entropy": 5.5589862823486325,
      "epoch": 2.540978785969334,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043520053340147003,
      "loss": 5.2039,
      "mean_token_accuracy": 0.18959733098745346,
      "num_tokens": 55785857.0,
      "step": 30245
    },
    {
      "entropy": 5.592557382583618,
      "epoch": 2.5413988657844993,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004351795042027184,
      "loss": 5.1556,
      "mean_token_accuracy": 0.19286150932312013,
      "num_tokens": 55795062.0,
      "step": 30250
    },
    {
      "entropy": 5.54801344871521,
      "epoch": 2.541818945599664,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043515847216646933,
      "loss": 5.2332,
      "mean_token_accuracy": 0.18461897671222688,
      "num_tokens": 55805099.0,
      "step": 30255
    },
    {
      "entropy": 5.449955320358276,
      "epoch": 2.542239025414829,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043513743729309533,
      "loss": 5.1285,
      "mean_token_accuracy": 0.19009532183408737,
      "num_tokens": 55815006.0,
      "step": 30260
    },
    {
      "entropy": 5.537880373001099,
      "epoch": 2.5426591052299936,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004351163995829692,
      "loss": 5.1615,
      "mean_token_accuracy": 0.19332342892885207,
      "num_tokens": 55823352.0,
      "step": 30265
    },
    {
      "entropy": 5.510927343368531,
      "epoch": 2.5430791850451584,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004350953590364634,
      "loss": 5.1639,
      "mean_token_accuracy": 0.1854093685746193,
      "num_tokens": 55832853.0,
      "step": 30270
    },
    {
      "entropy": 5.649024915695191,
      "epoch": 2.5434992648603236,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043507431565395076,
      "loss": 5.2399,
      "mean_token_accuracy": 0.18140957430005072,
      "num_tokens": 55842040.0,
      "step": 30275
    },
    {
      "entropy": 5.472314739227295,
      "epoch": 2.5439193446754884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004350532694358041,
      "loss": 5.1,
      "mean_token_accuracy": 0.19308952242136002,
      "num_tokens": 55850907.0,
      "step": 30280
    },
    {
      "entropy": 5.43353304862976,
      "epoch": 2.544339424490653,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004350322203823962,
      "loss": 5.1268,
      "mean_token_accuracy": 0.19043803513050078,
      "num_tokens": 55860271.0,
      "step": 30285
    },
    {
      "entropy": 5.602634716033935,
      "epoch": 2.544759504305818,
      "grad_norm": 1.25,
      "learning_rate": 0.0004350111684940998,
      "loss": 5.1853,
      "mean_token_accuracy": 0.1901439219713211,
      "num_tokens": 55868917.0,
      "step": 30290
    },
    {
      "entropy": 5.507132148742675,
      "epoch": 2.5451795841209828,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004349901137712881,
      "loss": 5.1752,
      "mean_token_accuracy": 0.19218392670154572,
      "num_tokens": 55877781.0,
      "step": 30295
    },
    {
      "entropy": 5.469120740890503,
      "epoch": 2.545599663936148,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004349690562143338,
      "loss": 5.1357,
      "mean_token_accuracy": 0.19314913153648378,
      "num_tokens": 55886829.0,
      "step": 30300
    },
    {
      "entropy": 5.535685062408447,
      "epoch": 2.546019743751313,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004349479958236101,
      "loss": 5.1647,
      "mean_token_accuracy": 0.18793170750141144,
      "num_tokens": 55896491.0,
      "step": 30305
    },
    {
      "entropy": 5.554475736618042,
      "epoch": 2.5464398235664776,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043492693259949,
      "loss": 5.2217,
      "mean_token_accuracy": 0.1849558338522911,
      "num_tokens": 55905616.0,
      "step": 30310
    },
    {
      "entropy": 5.493935775756836,
      "epoch": 2.5468599033816424,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004349058665423466,
      "loss": 5.0315,
      "mean_token_accuracy": 0.1967244416475296,
      "num_tokens": 55914175.0,
      "step": 30315
    },
    {
      "entropy": 5.496127080917359,
      "epoch": 2.547279983196807,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043488479765255313,
      "loss": 5.1165,
      "mean_token_accuracy": 0.19288550466299056,
      "num_tokens": 55923182.0,
      "step": 30320
    },
    {
      "entropy": 5.472088050842285,
      "epoch": 2.5477000630119724,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004348637259304828,
      "loss": 5.1141,
      "mean_token_accuracy": 0.1878558248281479,
      "num_tokens": 55932461.0,
      "step": 30325
    },
    {
      "entropy": 5.471930074691772,
      "epoch": 2.548120142827137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004348426513765089,
      "loss": 5.1533,
      "mean_token_accuracy": 0.18143776655197144,
      "num_tokens": 55941487.0,
      "step": 30330
    },
    {
      "entropy": 5.499337816238404,
      "epoch": 2.548540222642302,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004348215739910048,
      "loss": 5.2144,
      "mean_token_accuracy": 0.1883416712284088,
      "num_tokens": 55950503.0,
      "step": 30335
    },
    {
      "entropy": 5.572641515731812,
      "epoch": 2.548960302457467,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043480049377434373,
      "loss": 5.185,
      "mean_token_accuracy": 0.1836465284228325,
      "num_tokens": 55960012.0,
      "step": 30340
    },
    {
      "entropy": 5.582668209075928,
      "epoch": 2.5493803822726315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004347794107268992,
      "loss": 5.1262,
      "mean_token_accuracy": 0.1892571732401848,
      "num_tokens": 55969985.0,
      "step": 30345
    },
    {
      "entropy": 5.482986497879028,
      "epoch": 2.5498004620877968,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043475832484904473,
      "loss": 5.1245,
      "mean_token_accuracy": 0.19569194465875625,
      "num_tokens": 55978760.0,
      "step": 30350
    },
    {
      "entropy": 5.401320934295654,
      "epoch": 2.5502205419029615,
      "grad_norm": 1.125,
      "learning_rate": 0.00043473723614115376,
      "loss": 5.1041,
      "mean_token_accuracy": 0.18883305937051773,
      "num_tokens": 55986630.0,
      "step": 30355
    },
    {
      "entropy": 5.491638612747193,
      "epoch": 2.5506406217181263,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043471614460359985,
      "loss": 5.2346,
      "mean_token_accuracy": 0.18083437383174897,
      "num_tokens": 55995361.0,
      "step": 30360
    },
    {
      "entropy": 5.571425724029541,
      "epoch": 2.5510607015332916,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043469505023675684,
      "loss": 5.1768,
      "mean_token_accuracy": 0.19104745239019394,
      "num_tokens": 56004601.0,
      "step": 30365
    },
    {
      "entropy": 5.5091475486755375,
      "epoch": 2.5514807813484564,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004346739530409981,
      "loss": 5.163,
      "mean_token_accuracy": 0.18837871104478837,
      "num_tokens": 56014426.0,
      "step": 30370
    },
    {
      "entropy": 5.503758382797241,
      "epoch": 2.551900861163621,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043465285301669753,
      "loss": 5.0641,
      "mean_token_accuracy": 0.19328705668449403,
      "num_tokens": 56022943.0,
      "step": 30375
    },
    {
      "entropy": 5.545754146575928,
      "epoch": 2.552320940978786,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004346317501642289,
      "loss": 5.2114,
      "mean_token_accuracy": 0.18862257450819014,
      "num_tokens": 56032802.0,
      "step": 30380
    },
    {
      "entropy": 5.5107110977172855,
      "epoch": 2.5527410207939507,
      "grad_norm": 1.0625,
      "learning_rate": 0.000434610644483966,
      "loss": 5.2375,
      "mean_token_accuracy": 0.1791264072060585,
      "num_tokens": 56043074.0,
      "step": 30385
    },
    {
      "entropy": 5.646072292327881,
      "epoch": 2.553161100609116,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043458953597628276,
      "loss": 5.2744,
      "mean_token_accuracy": 0.18365247547626495,
      "num_tokens": 56053031.0,
      "step": 30390
    },
    {
      "entropy": 5.528688859939575,
      "epoch": 2.5535811804242807,
      "grad_norm": 1.125,
      "learning_rate": 0.00043456842464155307,
      "loss": 5.1608,
      "mean_token_accuracy": 0.18583139181137084,
      "num_tokens": 56062019.0,
      "step": 30395
    },
    {
      "entropy": 5.4633180618286135,
      "epoch": 2.5540012602394455,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043454731048015084,
      "loss": 5.1427,
      "mean_token_accuracy": 0.1858494833111763,
      "num_tokens": 56071050.0,
      "step": 30400
    },
    {
      "entropy": 5.559524726867676,
      "epoch": 2.5544213400546103,
      "grad_norm": 1.125,
      "learning_rate": 0.0004345261934924502,
      "loss": 5.1769,
      "mean_token_accuracy": 0.18463169485330583,
      "num_tokens": 56079911.0,
      "step": 30405
    },
    {
      "entropy": 5.521191644668579,
      "epoch": 2.554841419869775,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043450507367882515,
      "loss": 5.1194,
      "mean_token_accuracy": 0.1899896204471588,
      "num_tokens": 56089377.0,
      "step": 30410
    },
    {
      "entropy": 5.4749260425567625,
      "epoch": 2.5552614996849403,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043448395103964986,
      "loss": 5.1503,
      "mean_token_accuracy": 0.1905626878142357,
      "num_tokens": 56098374.0,
      "step": 30415
    },
    {
      "entropy": 5.413010025024414,
      "epoch": 2.555681579500105,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043446282557529854,
      "loss": 5.0361,
      "mean_token_accuracy": 0.19678647071123123,
      "num_tokens": 56107965.0,
      "step": 30420
    },
    {
      "entropy": 5.513029623031616,
      "epoch": 2.55610165931527,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043444169728614536,
      "loss": 5.1768,
      "mean_token_accuracy": 0.1883454814553261,
      "num_tokens": 56116795.0,
      "step": 30425
    },
    {
      "entropy": 5.387496471405029,
      "epoch": 2.5565217391304347,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004344205661725646,
      "loss": 5.1322,
      "mean_token_accuracy": 0.1899116411805153,
      "num_tokens": 56125653.0,
      "step": 30430
    },
    {
      "entropy": 5.493484830856323,
      "epoch": 2.5569418189455995,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043439943223493065,
      "loss": 5.1906,
      "mean_token_accuracy": 0.18970874845981597,
      "num_tokens": 56135451.0,
      "step": 30435
    },
    {
      "entropy": 5.534724330902099,
      "epoch": 2.5573618987607647,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043437829547361776,
      "loss": 5.1142,
      "mean_token_accuracy": 0.19025367200374604,
      "num_tokens": 56144454.0,
      "step": 30440
    },
    {
      "entropy": 5.572935152053833,
      "epoch": 2.5577819785759295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043435715588900045,
      "loss": 5.2217,
      "mean_token_accuracy": 0.17990583926439285,
      "num_tokens": 56154415.0,
      "step": 30445
    },
    {
      "entropy": 5.54227499961853,
      "epoch": 2.5582020583910943,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004343360134814532,
      "loss": 5.1187,
      "mean_token_accuracy": 0.1932160586118698,
      "num_tokens": 56163375.0,
      "step": 30450
    },
    {
      "entropy": 5.4982212543487545,
      "epoch": 2.558622138206259,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004343148682513506,
      "loss": 5.1183,
      "mean_token_accuracy": 0.18809721171855925,
      "num_tokens": 56172282.0,
      "step": 30455
    },
    {
      "entropy": 5.481016540527344,
      "epoch": 2.559042218021424,
      "grad_norm": 1.1875,
      "learning_rate": 0.000434293720199067,
      "loss": 5.193,
      "mean_token_accuracy": 0.17784777283668518,
      "num_tokens": 56181450.0,
      "step": 30460
    },
    {
      "entropy": 5.595362424850464,
      "epoch": 2.559462297836589,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004342725693249772,
      "loss": 5.2416,
      "mean_token_accuracy": 0.18653126806020737,
      "num_tokens": 56190837.0,
      "step": 30465
    },
    {
      "entropy": 5.641267251968384,
      "epoch": 2.559882377651754,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043425141562945585,
      "loss": 5.1596,
      "mean_token_accuracy": 0.19488691687583923,
      "num_tokens": 56200253.0,
      "step": 30470
    },
    {
      "entropy": 5.503944301605225,
      "epoch": 2.5603024574669186,
      "grad_norm": 1.125,
      "learning_rate": 0.0004342302591128777,
      "loss": 5.1142,
      "mean_token_accuracy": 0.19002165645360947,
      "num_tokens": 56209779.0,
      "step": 30475
    },
    {
      "entropy": 5.435065412521363,
      "epoch": 2.5607225372820834,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004342090997756175,
      "loss": 5.1495,
      "mean_token_accuracy": 0.1938638374209404,
      "num_tokens": 56218987.0,
      "step": 30480
    },
    {
      "entropy": 5.564526081085205,
      "epoch": 2.561142617097248,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004341879376180501,
      "loss": 5.1525,
      "mean_token_accuracy": 0.18257056027650834,
      "num_tokens": 56228155.0,
      "step": 30485
    },
    {
      "entropy": 5.5677734375,
      "epoch": 2.5615626969124135,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004341667726405503,
      "loss": 5.2086,
      "mean_token_accuracy": 0.18675844222307206,
      "num_tokens": 56237807.0,
      "step": 30490
    },
    {
      "entropy": 5.474724006652832,
      "epoch": 2.5619827767275782,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004341456048434932,
      "loss": 5.1733,
      "mean_token_accuracy": 0.1901071235537529,
      "num_tokens": 56247616.0,
      "step": 30495
    },
    {
      "entropy": 5.461116600036621,
      "epoch": 2.562402856542743,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004341244342272536,
      "loss": 5.1019,
      "mean_token_accuracy": 0.19533679783344268,
      "num_tokens": 56256586.0,
      "step": 30500
    },
    {
      "entropy": 5.5778439998626705,
      "epoch": 2.5628229363579083,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004341032607922065,
      "loss": 5.1988,
      "mean_token_accuracy": 0.1894515037536621,
      "num_tokens": 56266711.0,
      "step": 30505
    },
    {
      "entropy": 5.558708143234253,
      "epoch": 2.563243016173073,
      "grad_norm": 1.25,
      "learning_rate": 0.00043408208453872717,
      "loss": 5.1877,
      "mean_token_accuracy": 0.18690166622400284,
      "num_tokens": 56276650.0,
      "step": 30510
    },
    {
      "entropy": 5.5323365211486815,
      "epoch": 2.563663095988238,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043406090546719054,
      "loss": 5.1743,
      "mean_token_accuracy": 0.18713914304971696,
      "num_tokens": 56285364.0,
      "step": 30515
    },
    {
      "entropy": 5.559246301651001,
      "epoch": 2.5640831758034026,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004340397235779719,
      "loss": 5.1946,
      "mean_token_accuracy": 0.1884198248386383,
      "num_tokens": 56295243.0,
      "step": 30520
    },
    {
      "entropy": 5.6046027660369875,
      "epoch": 2.5645032556185674,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004340185388714465,
      "loss": 5.1828,
      "mean_token_accuracy": 0.18754370957612992,
      "num_tokens": 56305052.0,
      "step": 30525
    },
    {
      "entropy": 5.509028530120849,
      "epoch": 2.5649233354337326,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004339973513479895,
      "loss": 5.1606,
      "mean_token_accuracy": 0.18696104884147643,
      "num_tokens": 56314213.0,
      "step": 30530
    },
    {
      "entropy": 5.586784315109253,
      "epoch": 2.5653434152488974,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004339761610079763,
      "loss": 5.2498,
      "mean_token_accuracy": 0.1816104531288147,
      "num_tokens": 56324068.0,
      "step": 30535
    },
    {
      "entropy": 5.544496059417725,
      "epoch": 2.565763495064062,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004339549678517823,
      "loss": 5.1125,
      "mean_token_accuracy": 0.1880229890346527,
      "num_tokens": 56332423.0,
      "step": 30540
    },
    {
      "entropy": 5.470459508895874,
      "epoch": 2.566183574879227,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004339337718797828,
      "loss": 5.1389,
      "mean_token_accuracy": 0.18910287469625472,
      "num_tokens": 56341725.0,
      "step": 30545
    },
    {
      "entropy": 5.573567008972168,
      "epoch": 2.566603654694392,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043391257309235347,
      "loss": 5.2861,
      "mean_token_accuracy": 0.18588721454143525,
      "num_tokens": 56351079.0,
      "step": 30550
    },
    {
      "entropy": 5.539049386978149,
      "epoch": 2.567023734509557,
      "grad_norm": 1.125,
      "learning_rate": 0.00043389137148986973,
      "loss": 5.0818,
      "mean_token_accuracy": 0.18607227206230165,
      "num_tokens": 56361007.0,
      "step": 30555
    },
    {
      "entropy": 5.549571466445923,
      "epoch": 2.567443814324722,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004338701670727072,
      "loss": 5.2935,
      "mean_token_accuracy": 0.17945171892642975,
      "num_tokens": 56370881.0,
      "step": 30560
    },
    {
      "entropy": 5.551998710632324,
      "epoch": 2.5678638941398866,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004338489598412414,
      "loss": 5.2006,
      "mean_token_accuracy": 0.18891485929489135,
      "num_tokens": 56380148.0,
      "step": 30565
    },
    {
      "entropy": 5.530758380889893,
      "epoch": 2.5682839739550514,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043382774979584804,
      "loss": 5.1755,
      "mean_token_accuracy": 0.18459355235099792,
      "num_tokens": 56388931.0,
      "step": 30570
    },
    {
      "entropy": 5.489635992050171,
      "epoch": 2.568704053770216,
      "grad_norm": 1.125,
      "learning_rate": 0.000433806536936903,
      "loss": 5.1134,
      "mean_token_accuracy": 0.18688125610351564,
      "num_tokens": 56398249.0,
      "step": 30575
    },
    {
      "entropy": 5.520345687866211,
      "epoch": 2.5691241335853814,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004337853212647819,
      "loss": 5.1601,
      "mean_token_accuracy": 0.19216308891773223,
      "num_tokens": 56406761.0,
      "step": 30580
    },
    {
      "entropy": 5.5557859420776365,
      "epoch": 2.569544213400546,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043376410277986047,
      "loss": 5.2371,
      "mean_token_accuracy": 0.18071142882108687,
      "num_tokens": 56416460.0,
      "step": 30585
    },
    {
      "entropy": 5.493467378616333,
      "epoch": 2.569964293215711,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004337428814825148,
      "loss": 5.23,
      "mean_token_accuracy": 0.1910933256149292,
      "num_tokens": 56425867.0,
      "step": 30590
    },
    {
      "entropy": 5.494433069229126,
      "epoch": 2.5703843730308757,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004337216573731207,
      "loss": 5.055,
      "mean_token_accuracy": 0.19252495616674423,
      "num_tokens": 56434719.0,
      "step": 30595
    },
    {
      "entropy": 5.54516921043396,
      "epoch": 2.5708044528460405,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004337004304520542,
      "loss": 5.129,
      "mean_token_accuracy": 0.1932085335254669,
      "num_tokens": 56444260.0,
      "step": 30600
    },
    {
      "entropy": 5.5634198665618895,
      "epoch": 2.5712245326612058,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004336792007196913,
      "loss": 5.1377,
      "mean_token_accuracy": 0.19032558053731918,
      "num_tokens": 56453662.0,
      "step": 30605
    },
    {
      "entropy": 5.586282062530517,
      "epoch": 2.5716446124763706,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043365796817640797,
      "loss": 5.1845,
      "mean_token_accuracy": 0.18914655148983,
      "num_tokens": 56462521.0,
      "step": 30610
    },
    {
      "entropy": 5.4564110279083256,
      "epoch": 2.5720646922915353,
      "grad_norm": 1.125,
      "learning_rate": 0.00043363673282258055,
      "loss": 5.1322,
      "mean_token_accuracy": 0.18892084509134294,
      "num_tokens": 56471236.0,
      "step": 30615
    },
    {
      "entropy": 5.543896627426148,
      "epoch": 2.5724847721067,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004336154946585851,
      "loss": 5.1926,
      "mean_token_accuracy": 0.1927035704255104,
      "num_tokens": 56481094.0,
      "step": 30620
    },
    {
      "entropy": 5.50788197517395,
      "epoch": 2.572904851921865,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004335942536847977,
      "loss": 5.086,
      "mean_token_accuracy": 0.19093281775712967,
      "num_tokens": 56490997.0,
      "step": 30625
    },
    {
      "entropy": 5.560194683074951,
      "epoch": 2.57332493173703,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004335730099015949,
      "loss": 5.2,
      "mean_token_accuracy": 0.18550018966197968,
      "num_tokens": 56500118.0,
      "step": 30630
    },
    {
      "entropy": 5.607406949996948,
      "epoch": 2.573745011552195,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004335517633093528,
      "loss": 5.2841,
      "mean_token_accuracy": 0.17687698304653168,
      "num_tokens": 56509952.0,
      "step": 30635
    },
    {
      "entropy": 5.571618366241455,
      "epoch": 2.5741650913673597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043353051390844787,
      "loss": 5.1463,
      "mean_token_accuracy": 0.1932022601366043,
      "num_tokens": 56518637.0,
      "step": 30640
    },
    {
      "entropy": 5.507772302627563,
      "epoch": 2.574585171182525,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004335092616992565,
      "loss": 5.2457,
      "mean_token_accuracy": 0.18418538868427276,
      "num_tokens": 56529152.0,
      "step": 30645
    },
    {
      "entropy": 5.49394211769104,
      "epoch": 2.5750052509976893,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043348800668215526,
      "loss": 5.0814,
      "mean_token_accuracy": 0.19504878520965577,
      "num_tokens": 56538724.0,
      "step": 30650
    },
    {
      "entropy": 5.557313919067383,
      "epoch": 2.5754253308128545,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043346674885752056,
      "loss": 5.2382,
      "mean_token_accuracy": 0.17880548387765885,
      "num_tokens": 56547856.0,
      "step": 30655
    },
    {
      "entropy": 5.611384773254395,
      "epoch": 2.5758454106280193,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043344548822572896,
      "loss": 5.2059,
      "mean_token_accuracy": 0.18748241364955903,
      "num_tokens": 56556705.0,
      "step": 30660
    },
    {
      "entropy": 5.625724458694458,
      "epoch": 2.576265490443184,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043342422478715717,
      "loss": 5.273,
      "mean_token_accuracy": 0.18366764038801192,
      "num_tokens": 56565945.0,
      "step": 30665
    },
    {
      "entropy": 5.490627193450928,
      "epoch": 2.5766855702583493,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004334029585421818,
      "loss": 5.1604,
      "mean_token_accuracy": 0.1877441942691803,
      "num_tokens": 56574950.0,
      "step": 30670
    },
    {
      "entropy": 5.4424206733703615,
      "epoch": 2.577105650073514,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004333816894911795,
      "loss": 5.1335,
      "mean_token_accuracy": 0.1840498775243759,
      "num_tokens": 56584670.0,
      "step": 30675
    },
    {
      "entropy": 5.548323249816894,
      "epoch": 2.577525729888679,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043336041763452726,
      "loss": 5.1417,
      "mean_token_accuracy": 0.18330129235982895,
      "num_tokens": 56594776.0,
      "step": 30680
    },
    {
      "entropy": 5.564962244033813,
      "epoch": 2.5779458097038437,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004333391429726016,
      "loss": 5.1759,
      "mean_token_accuracy": 0.19003473222255707,
      "num_tokens": 56605302.0,
      "step": 30685
    },
    {
      "entropy": 5.564210700988769,
      "epoch": 2.5783658895190085,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043331786550577974,
      "loss": 5.2004,
      "mean_token_accuracy": 0.18254441022872925,
      "num_tokens": 56614388.0,
      "step": 30690
    },
    {
      "entropy": 5.444561958312988,
      "epoch": 2.5787859693341737,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043329658523443834,
      "loss": 5.0833,
      "mean_token_accuracy": 0.19499247670173644,
      "num_tokens": 56623349.0,
      "step": 30695
    },
    {
      "entropy": 5.461025381088257,
      "epoch": 2.5792060491493385,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043327530215895443,
      "loss": 5.0733,
      "mean_token_accuracy": 0.18814963102340698,
      "num_tokens": 56632503.0,
      "step": 30700
    },
    {
      "entropy": 5.539375877380371,
      "epoch": 2.5796261289645033,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043325401627970506,
      "loss": 5.199,
      "mean_token_accuracy": 0.18862173557281495,
      "num_tokens": 56642055.0,
      "step": 30705
    },
    {
      "entropy": 5.4634675025939945,
      "epoch": 2.580046208779668,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004332327275970672,
      "loss": 5.0672,
      "mean_token_accuracy": 0.1970888063311577,
      "num_tokens": 56651071.0,
      "step": 30710
    },
    {
      "entropy": 5.635152053833008,
      "epoch": 2.580466288594833,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043321143611141816,
      "loss": 5.2885,
      "mean_token_accuracy": 0.18024080246686935,
      "num_tokens": 56662091.0,
      "step": 30715
    },
    {
      "entropy": 5.58305401802063,
      "epoch": 2.580886368409998,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043319014182313496,
      "loss": 5.2156,
      "mean_token_accuracy": 0.18346319049596788,
      "num_tokens": 56671261.0,
      "step": 30720
    },
    {
      "entropy": 5.479212427139283,
      "epoch": 2.581306448225163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043316884473259485,
      "loss": 5.1105,
      "mean_token_accuracy": 0.1935817152261734,
      "num_tokens": 56680906.0,
      "step": 30725
    },
    {
      "entropy": 5.595008087158203,
      "epoch": 2.5817265280403277,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043314754484017503,
      "loss": 5.212,
      "mean_token_accuracy": 0.1891324833035469,
      "num_tokens": 56689933.0,
      "step": 30730
    },
    {
      "entropy": 5.457785415649414,
      "epoch": 2.5821466078554924,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000433126242146253,
      "loss": 5.1946,
      "mean_token_accuracy": 0.18495263904333115,
      "num_tokens": 56699065.0,
      "step": 30735
    },
    {
      "entropy": 5.502129030227661,
      "epoch": 2.5825666876706572,
      "grad_norm": 1.140625,
      "learning_rate": 0.000433104936651206,
      "loss": 5.206,
      "mean_token_accuracy": 0.18934895545244218,
      "num_tokens": 56707986.0,
      "step": 30740
    },
    {
      "entropy": 5.675470066070557,
      "epoch": 2.5829867674858225,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004330836283554114,
      "loss": 5.206,
      "mean_token_accuracy": 0.1820852726697922,
      "num_tokens": 56716190.0,
      "step": 30745
    },
    {
      "entropy": 5.579714965820313,
      "epoch": 2.5834068473009872,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043306231725924677,
      "loss": 5.1334,
      "mean_token_accuracy": 0.1934879407286644,
      "num_tokens": 56725097.0,
      "step": 30750
    },
    {
      "entropy": 5.573604345321655,
      "epoch": 2.583826927116152,
      "grad_norm": 1.609375,
      "learning_rate": 0.00043304100336308954,
      "loss": 5.2197,
      "mean_token_accuracy": 0.18042871803045274,
      "num_tokens": 56734440.0,
      "step": 30755
    },
    {
      "entropy": 5.606794548034668,
      "epoch": 2.584247006931317,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043301968666731734,
      "loss": 5.2773,
      "mean_token_accuracy": 0.1853393629193306,
      "num_tokens": 56744521.0,
      "step": 30760
    },
    {
      "entropy": 5.438584566116333,
      "epoch": 2.5846670867464816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004329983671723078,
      "loss": 5.1027,
      "mean_token_accuracy": 0.1945261001586914,
      "num_tokens": 56754134.0,
      "step": 30765
    },
    {
      "entropy": 5.471046066284179,
      "epoch": 2.585087166561647,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004329770448784384,
      "loss": 5.0745,
      "mean_token_accuracy": 0.1915004387497902,
      "num_tokens": 56762750.0,
      "step": 30770
    },
    {
      "entropy": 5.596000385284424,
      "epoch": 2.5855072463768116,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004329557197860872,
      "loss": 5.2147,
      "mean_token_accuracy": 0.1877630263566971,
      "num_tokens": 56771954.0,
      "step": 30775
    },
    {
      "entropy": 5.537437105178833,
      "epoch": 2.5859273261919764,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043293439189563167,
      "loss": 5.1325,
      "mean_token_accuracy": 0.19766118079423906,
      "num_tokens": 56780191.0,
      "step": 30780
    },
    {
      "entropy": 5.472856426239014,
      "epoch": 2.586347406007141,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004329130612074497,
      "loss": 5.171,
      "mean_token_accuracy": 0.18418103754520415,
      "num_tokens": 56790295.0,
      "step": 30785
    },
    {
      "entropy": 5.489522981643677,
      "epoch": 2.586767485822306,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004328917277219191,
      "loss": 5.0233,
      "mean_token_accuracy": 0.19508449882268905,
      "num_tokens": 56798601.0,
      "step": 30790
    },
    {
      "entropy": 5.380479192733764,
      "epoch": 2.587187565637471,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043287039143941785,
      "loss": 5.0006,
      "mean_token_accuracy": 0.19292248636484147,
      "num_tokens": 56808499.0,
      "step": 30795
    },
    {
      "entropy": 5.502881336212158,
      "epoch": 2.587607645452636,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000432849052360324,
      "loss": 5.1535,
      "mean_token_accuracy": 0.19044586271047592,
      "num_tokens": 56818073.0,
      "step": 30800
    },
    {
      "entropy": 5.5150402069091795,
      "epoch": 2.588027725267801,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004328277104850154,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19081406444311141,
      "num_tokens": 56828190.0,
      "step": 30805
    },
    {
      "entropy": 5.536669111251831,
      "epoch": 2.588447805082966,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043280636581387023,
      "loss": 5.1618,
      "mean_token_accuracy": 0.18940214812755585,
      "num_tokens": 56837032.0,
      "step": 30810
    },
    {
      "entropy": 5.411822843551636,
      "epoch": 2.588867884898131,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004327850183472665,
      "loss": 5.0384,
      "mean_token_accuracy": 0.20916508436203002,
      "num_tokens": 56846138.0,
      "step": 30815
    },
    {
      "entropy": 5.501022911071777,
      "epoch": 2.5892879647132956,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004327636680855824,
      "loss": 5.1562,
      "mean_token_accuracy": 0.18866044282913208,
      "num_tokens": 56856336.0,
      "step": 30820
    },
    {
      "entropy": 5.570519828796387,
      "epoch": 2.5897080445284604,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043274231502919626,
      "loss": 5.1568,
      "mean_token_accuracy": 0.18372846990823746,
      "num_tokens": 56865555.0,
      "step": 30825
    },
    {
      "entropy": 5.484909915924073,
      "epoch": 2.590128124343625,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043272095917848614,
      "loss": 5.1113,
      "mean_token_accuracy": 0.18954291194677353,
      "num_tokens": 56875194.0,
      "step": 30830
    },
    {
      "entropy": 5.5256775379180905,
      "epoch": 2.5905482041587904,
      "grad_norm": 1.125,
      "learning_rate": 0.0004326996005338305,
      "loss": 5.2136,
      "mean_token_accuracy": 0.1858087182044983,
      "num_tokens": 56884571.0,
      "step": 30835
    },
    {
      "entropy": 5.5895874977111815,
      "epoch": 2.590968283973955,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004326782390956076,
      "loss": 5.2039,
      "mean_token_accuracy": 0.1932423382997513,
      "num_tokens": 56893833.0,
      "step": 30840
    },
    {
      "entropy": 5.466413497924805,
      "epoch": 2.59138836378912,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004326568748641959,
      "loss": 5.1728,
      "mean_token_accuracy": 0.18868966549634933,
      "num_tokens": 56903382.0,
      "step": 30845
    },
    {
      "entropy": 5.450658893585205,
      "epoch": 2.5918084436042848,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004326355078399738,
      "loss": 5.0689,
      "mean_token_accuracy": 0.19063598066568374,
      "num_tokens": 56912317.0,
      "step": 30850
    },
    {
      "entropy": 5.50457534790039,
      "epoch": 2.5922285234194495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043261413802331993,
      "loss": 5.0325,
      "mean_token_accuracy": 0.19168474972248079,
      "num_tokens": 56921596.0,
      "step": 30855
    },
    {
      "entropy": 5.448745965957642,
      "epoch": 2.5926486032346148,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043259276541461275,
      "loss": 5.0591,
      "mean_token_accuracy": 0.19170855730772018,
      "num_tokens": 56930061.0,
      "step": 30860
    },
    {
      "entropy": 5.5345316410064695,
      "epoch": 2.5930686830497796,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043257139001423085,
      "loss": 5.2154,
      "mean_token_accuracy": 0.1943620502948761,
      "num_tokens": 56938622.0,
      "step": 30865
    },
    {
      "entropy": 5.506506013870239,
      "epoch": 2.5934887628649443,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043255001182255294,
      "loss": 5.1437,
      "mean_token_accuracy": 0.1891665890812874,
      "num_tokens": 56947379.0,
      "step": 30870
    },
    {
      "entropy": 5.535540437698364,
      "epoch": 2.593908842680109,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043252863083995764,
      "loss": 5.1857,
      "mean_token_accuracy": 0.1833522364497185,
      "num_tokens": 56956572.0,
      "step": 30875
    },
    {
      "entropy": 5.552693939208984,
      "epoch": 2.594328922495274,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004325072470668238,
      "loss": 5.2198,
      "mean_token_accuracy": 0.18516040444374085,
      "num_tokens": 56966038.0,
      "step": 30880
    },
    {
      "entropy": 5.574924516677856,
      "epoch": 2.594749002310439,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043248586050353017,
      "loss": 5.2024,
      "mean_token_accuracy": 0.18645667135715485,
      "num_tokens": 56975287.0,
      "step": 30885
    },
    {
      "entropy": 5.500284862518311,
      "epoch": 2.595169082125604,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043246447115045564,
      "loss": 5.1256,
      "mean_token_accuracy": 0.19056877344846726,
      "num_tokens": 56984798.0,
      "step": 30890
    },
    {
      "entropy": 5.431474304199218,
      "epoch": 2.5955891619407687,
      "grad_norm": 1.140625,
      "learning_rate": 0.000432443079007979,
      "loss": 5.157,
      "mean_token_accuracy": 0.18433244079351424,
      "num_tokens": 56993497.0,
      "step": 30895
    },
    {
      "entropy": 5.52392463684082,
      "epoch": 2.5960092417559335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043242168407647943,
      "loss": 5.0853,
      "mean_token_accuracy": 0.19564361572265626,
      "num_tokens": 57002667.0,
      "step": 30900
    },
    {
      "entropy": 5.558113718032837,
      "epoch": 2.5964293215710983,
      "grad_norm": 1.125,
      "learning_rate": 0.00043240028635633566,
      "loss": 5.1406,
      "mean_token_accuracy": 0.19301715195178987,
      "num_tokens": 57011051.0,
      "step": 30905
    },
    {
      "entropy": 5.491744565963745,
      "epoch": 2.5968494013862635,
      "grad_norm": 1.125,
      "learning_rate": 0.0004323788858479269,
      "loss": 5.1814,
      "mean_token_accuracy": 0.19306180775165557,
      "num_tokens": 57020777.0,
      "step": 30910
    },
    {
      "entropy": 5.553839063644409,
      "epoch": 2.5972694812014283,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043235748255163225,
      "loss": 5.1926,
      "mean_token_accuracy": 0.18773225992918013,
      "num_tokens": 57030009.0,
      "step": 30915
    },
    {
      "entropy": 5.532326364517212,
      "epoch": 2.597689561016593,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043233607646783084,
      "loss": 5.11,
      "mean_token_accuracy": 0.19465990513563156,
      "num_tokens": 57039128.0,
      "step": 30920
    },
    {
      "entropy": 5.444122552871704,
      "epoch": 2.598109640831758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004323146675969018,
      "loss": 5.1092,
      "mean_token_accuracy": 0.19785375744104386,
      "num_tokens": 57048067.0,
      "step": 30925
    },
    {
      "entropy": 5.4579915523529055,
      "epoch": 2.5985297206469227,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004322932559392244,
      "loss": 5.1388,
      "mean_token_accuracy": 0.18935923725366594,
      "num_tokens": 57057494.0,
      "step": 30930
    },
    {
      "entropy": 5.486630392074585,
      "epoch": 2.598949800462088,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043227184149517803,
      "loss": 5.1786,
      "mean_token_accuracy": 0.19278823882341384,
      "num_tokens": 57067259.0,
      "step": 30935
    },
    {
      "entropy": 5.415824604034424,
      "epoch": 2.5993698802772527,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043225042426514197,
      "loss": 5.0927,
      "mean_token_accuracy": 0.19226229041814805,
      "num_tokens": 57076650.0,
      "step": 30940
    },
    {
      "entropy": 5.487235355377197,
      "epoch": 2.5997899600924175,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004322290042494956,
      "loss": 5.244,
      "mean_token_accuracy": 0.1826328992843628,
      "num_tokens": 57087789.0,
      "step": 30945
    },
    {
      "entropy": 5.454976987838745,
      "epoch": 2.6002100399075827,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043220758144861836,
      "loss": 5.0211,
      "mean_token_accuracy": 0.1959989070892334,
      "num_tokens": 57096112.0,
      "step": 30950
    },
    {
      "entropy": 5.486496305465698,
      "epoch": 2.600630119722747,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004321861558628898,
      "loss": 5.1423,
      "mean_token_accuracy": 0.18781739920377732,
      "num_tokens": 57105342.0,
      "step": 30955
    },
    {
      "entropy": 5.502754402160645,
      "epoch": 2.6010501995379123,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043216472749268937,
      "loss": 5.08,
      "mean_token_accuracy": 0.19378931373357772,
      "num_tokens": 57114137.0,
      "step": 30960
    },
    {
      "entropy": 5.426181840896606,
      "epoch": 2.601470279353077,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043214329633839676,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19645897001028062,
      "num_tokens": 57122494.0,
      "step": 30965
    },
    {
      "entropy": 5.530596399307251,
      "epoch": 2.601890359168242,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004321218624003916,
      "loss": 5.2335,
      "mean_token_accuracy": 0.17556762099266052,
      "num_tokens": 57131546.0,
      "step": 30970
    },
    {
      "entropy": 5.5023589611053465,
      "epoch": 2.602310438983407,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043210042567905357,
      "loss": 5.0737,
      "mean_token_accuracy": 0.20274895876646043,
      "num_tokens": 57140356.0,
      "step": 30975
    },
    {
      "entropy": 5.4437871932983395,
      "epoch": 2.602730518798572,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043207898617476226,
      "loss": 5.103,
      "mean_token_accuracy": 0.19646032005548478,
      "num_tokens": 57149415.0,
      "step": 30980
    },
    {
      "entropy": 5.443943452835083,
      "epoch": 2.6031505986137367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004320575438878977,
      "loss": 5.2177,
      "mean_token_accuracy": 0.1862757295370102,
      "num_tokens": 57159330.0,
      "step": 30985
    },
    {
      "entropy": 5.4467872142791744,
      "epoch": 2.6035706784289014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004320360988188396,
      "loss": 5.0912,
      "mean_token_accuracy": 0.19785370230674743,
      "num_tokens": 57168950.0,
      "step": 30990
    },
    {
      "entropy": 5.53366003036499,
      "epoch": 2.6039907582440662,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004320146509679679,
      "loss": 5.1555,
      "mean_token_accuracy": 0.18768541365861893,
      "num_tokens": 57178094.0,
      "step": 30995
    },
    {
      "entropy": 5.520300388336182,
      "epoch": 2.6044108380592315,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004319932003356625,
      "loss": 5.1473,
      "mean_token_accuracy": 0.19241006523370743,
      "num_tokens": 57187179.0,
      "step": 31000
    },
    {
      "entropy": 5.46401047706604,
      "epoch": 2.6048309178743962,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004319717469223034,
      "loss": 5.0503,
      "mean_token_accuracy": 0.19473724216222763,
      "num_tokens": 57195933.0,
      "step": 31005
    },
    {
      "entropy": 5.553114891052246,
      "epoch": 2.605250997689561,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043195029072827057,
      "loss": 5.1973,
      "mean_token_accuracy": 0.18557906448841094,
      "num_tokens": 57205582.0,
      "step": 31010
    },
    {
      "entropy": 5.562368583679199,
      "epoch": 2.605671077504726,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004319288317539442,
      "loss": 5.2486,
      "mean_token_accuracy": 0.18530704081058502,
      "num_tokens": 57214178.0,
      "step": 31015
    },
    {
      "entropy": 5.4461225986480715,
      "epoch": 2.6060911573198906,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043190736999970436,
      "loss": 5.0541,
      "mean_token_accuracy": 0.187532439827919,
      "num_tokens": 57223646.0,
      "step": 31020
    },
    {
      "entropy": 5.488103437423706,
      "epoch": 2.606511237135056,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004318859054659313,
      "loss": 5.1328,
      "mean_token_accuracy": 0.18955357372760773,
      "num_tokens": 57232355.0,
      "step": 31025
    },
    {
      "entropy": 5.64519362449646,
      "epoch": 2.6069313169502206,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043186443815300515,
      "loss": 5.3316,
      "mean_token_accuracy": 0.18608778864145278,
      "num_tokens": 57241429.0,
      "step": 31030
    },
    {
      "entropy": 5.633120918273926,
      "epoch": 2.6073513967653854,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004318429680613063,
      "loss": 5.2406,
      "mean_token_accuracy": 0.18918096721172334,
      "num_tokens": 57250280.0,
      "step": 31035
    },
    {
      "entropy": 5.487821054458618,
      "epoch": 2.60777147658055,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043182149519121506,
      "loss": 5.0941,
      "mean_token_accuracy": 0.19509832710027694,
      "num_tokens": 57259469.0,
      "step": 31040
    },
    {
      "entropy": 5.4230131149292,
      "epoch": 2.608191556395715,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043180001954311174,
      "loss": 5.1612,
      "mean_token_accuracy": 0.18906352072954177,
      "num_tokens": 57269081.0,
      "step": 31045
    },
    {
      "entropy": 5.578999662399292,
      "epoch": 2.60861163621088,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004317785411173768,
      "loss": 5.1879,
      "mean_token_accuracy": 0.18904557824134827,
      "num_tokens": 57278744.0,
      "step": 31050
    },
    {
      "entropy": 5.580397415161133,
      "epoch": 2.609031716026045,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043175705991439073,
      "loss": 5.203,
      "mean_token_accuracy": 0.18975525945425034,
      "num_tokens": 57287041.0,
      "step": 31055
    },
    {
      "entropy": 5.489940166473389,
      "epoch": 2.60945179584121,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004317355759345341,
      "loss": 5.1681,
      "mean_token_accuracy": 0.18868425041437148,
      "num_tokens": 57296897.0,
      "step": 31060
    },
    {
      "entropy": 5.441902828216553,
      "epoch": 2.6098718756563746,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043171408917818744,
      "loss": 5.0645,
      "mean_token_accuracy": 0.19538479149341584,
      "num_tokens": 57306821.0,
      "step": 31065
    },
    {
      "entropy": 5.553781604766845,
      "epoch": 2.6102919554715394,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043169259964573136,
      "loss": 5.1582,
      "mean_token_accuracy": 0.1880461886525154,
      "num_tokens": 57315341.0,
      "step": 31070
    },
    {
      "entropy": 5.658563852310181,
      "epoch": 2.6107120352867046,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004316711073375466,
      "loss": 5.2841,
      "mean_token_accuracy": 0.18033913373947144,
      "num_tokens": 57325250.0,
      "step": 31075
    },
    {
      "entropy": 5.562008571624756,
      "epoch": 2.6111321151018694,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004316496122540138,
      "loss": 5.2076,
      "mean_token_accuracy": 0.18445756435394287,
      "num_tokens": 57333889.0,
      "step": 31080
    },
    {
      "entropy": 5.544509744644165,
      "epoch": 2.611552194917034,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004316281143955138,
      "loss": 5.2318,
      "mean_token_accuracy": 0.18187065571546554,
      "num_tokens": 57342779.0,
      "step": 31085
    },
    {
      "entropy": 5.530968236923218,
      "epoch": 2.6119722747321994,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004316066137624274,
      "loss": 5.1353,
      "mean_token_accuracy": 0.1923784539103508,
      "num_tokens": 57351955.0,
      "step": 31090
    },
    {
      "entropy": 5.523536300659179,
      "epoch": 2.6123923545473637,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043158511035513547,
      "loss": 5.1782,
      "mean_token_accuracy": 0.1884585961699486,
      "num_tokens": 57360540.0,
      "step": 31095
    },
    {
      "entropy": 5.459219884872437,
      "epoch": 2.612812434362529,
      "grad_norm": 1.234375,
      "learning_rate": 0.000431563604174019,
      "loss": 5.1226,
      "mean_token_accuracy": 0.18420231938362122,
      "num_tokens": 57370322.0,
      "step": 31100
    },
    {
      "entropy": 5.500577211380005,
      "epoch": 2.6132325141776938,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004315420952194588,
      "loss": 5.1316,
      "mean_token_accuracy": 0.18860035687685012,
      "num_tokens": 57379577.0,
      "step": 31105
    },
    {
      "entropy": 5.428093957901001,
      "epoch": 2.6136525939928585,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004315205834918361,
      "loss": 4.9849,
      "mean_token_accuracy": 0.19223637282848358,
      "num_tokens": 57388862.0,
      "step": 31110
    },
    {
      "entropy": 5.542025709152222,
      "epoch": 2.6140726738080238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004314990689915318,
      "loss": 5.2025,
      "mean_token_accuracy": 0.1933722823858261,
      "num_tokens": 57398323.0,
      "step": 31115
    },
    {
      "entropy": 5.513317537307739,
      "epoch": 2.6144927536231886,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004314775517189271,
      "loss": 5.1184,
      "mean_token_accuracy": 0.1917305514216423,
      "num_tokens": 57407317.0,
      "step": 31120
    },
    {
      "entropy": 5.50135989189148,
      "epoch": 2.6149128334383533,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004314560316744031,
      "loss": 5.1304,
      "mean_token_accuracy": 0.19680935144424438,
      "num_tokens": 57415936.0,
      "step": 31125
    },
    {
      "entropy": 5.432661485671997,
      "epoch": 2.615332913253518,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004314345088583411,
      "loss": 5.0619,
      "mean_token_accuracy": 0.19323383569717406,
      "num_tokens": 57424548.0,
      "step": 31130
    },
    {
      "entropy": 5.5351762771606445,
      "epoch": 2.615752993068683,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043141298327112234,
      "loss": 5.1258,
      "mean_token_accuracy": 0.1875014841556549,
      "num_tokens": 57433991.0,
      "step": 31135
    },
    {
      "entropy": 5.545192527770996,
      "epoch": 2.616173072883848,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043139145491312815,
      "loss": 5.1899,
      "mean_token_accuracy": 0.19004935026168823,
      "num_tokens": 57442921.0,
      "step": 31140
    },
    {
      "entropy": 5.646677684783936,
      "epoch": 2.616593152699013,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043136992378473975,
      "loss": 5.2719,
      "mean_token_accuracy": 0.180599345266819,
      "num_tokens": 57451015.0,
      "step": 31145
    },
    {
      "entropy": 5.519902992248535,
      "epoch": 2.6170132325141777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004313483898863387,
      "loss": 5.057,
      "mean_token_accuracy": 0.19488445669412613,
      "num_tokens": 57460117.0,
      "step": 31150
    },
    {
      "entropy": 5.485816097259521,
      "epoch": 2.6174333123293425,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004313268532183065,
      "loss": 5.1188,
      "mean_token_accuracy": 0.19786026626825332,
      "num_tokens": 57470319.0,
      "step": 31155
    },
    {
      "entropy": 5.474789428710937,
      "epoch": 2.6178533921445073,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043130531378102454,
      "loss": 5.117,
      "mean_token_accuracy": 0.19620194733142854,
      "num_tokens": 57479980.0,
      "step": 31160
    },
    {
      "entropy": 5.574428129196167,
      "epoch": 2.6182734719596725,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004312837715748745,
      "loss": 5.2954,
      "mean_token_accuracy": 0.18197956383228303,
      "num_tokens": 57490498.0,
      "step": 31165
    },
    {
      "entropy": 5.463033008575439,
      "epoch": 2.6186935517748373,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043126222660023783,
      "loss": 5.1282,
      "mean_token_accuracy": 0.18975790441036225,
      "num_tokens": 57500597.0,
      "step": 31170
    },
    {
      "entropy": 5.558431196212768,
      "epoch": 2.619113631590002,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043124067885749636,
      "loss": 5.1821,
      "mean_token_accuracy": 0.19410171508789062,
      "num_tokens": 57510107.0,
      "step": 31175
    },
    {
      "entropy": 5.4648314952850345,
      "epoch": 2.619533711405167,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004312191283470316,
      "loss": 5.066,
      "mean_token_accuracy": 0.19601387530565262,
      "num_tokens": 57519348.0,
      "step": 31180
    },
    {
      "entropy": 5.457971572875977,
      "epoch": 2.6199537912203317,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004311975750692255,
      "loss": 5.1396,
      "mean_token_accuracy": 0.18686210066080094,
      "num_tokens": 57528607.0,
      "step": 31185
    },
    {
      "entropy": 5.48379955291748,
      "epoch": 2.620373871035497,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043117601902445987,
      "loss": 5.1472,
      "mean_token_accuracy": 0.1918279767036438,
      "num_tokens": 57538024.0,
      "step": 31190
    },
    {
      "entropy": 5.528535461425781,
      "epoch": 2.6207939508506617,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043115446021311637,
      "loss": 5.2039,
      "mean_token_accuracy": 0.18182047158479692,
      "num_tokens": 57548534.0,
      "step": 31195
    },
    {
      "entropy": 5.511900138854981,
      "epoch": 2.6212140306658265,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043113289863557706,
      "loss": 5.1451,
      "mean_token_accuracy": 0.19597591310739518,
      "num_tokens": 57557614.0,
      "step": 31200
    },
    {
      "entropy": 5.575668048858643,
      "epoch": 2.6216341104809913,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004311113342922238,
      "loss": 5.1753,
      "mean_token_accuracy": 0.18963324576616286,
      "num_tokens": 57566720.0,
      "step": 31205
    },
    {
      "entropy": 5.461807727813721,
      "epoch": 2.622054190296156,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043108976718343874,
      "loss": 5.0991,
      "mean_token_accuracy": 0.18846842348575593,
      "num_tokens": 57575823.0,
      "step": 31210
    },
    {
      "entropy": 5.500463962554932,
      "epoch": 2.6224742701113213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004310681973096037,
      "loss": 5.125,
      "mean_token_accuracy": 0.18725589513778687,
      "num_tokens": 57585556.0,
      "step": 31215
    },
    {
      "entropy": 5.634766721725464,
      "epoch": 2.622894349926486,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043104662467110096,
      "loss": 5.268,
      "mean_token_accuracy": 0.18168987184762955,
      "num_tokens": 57594573.0,
      "step": 31220
    },
    {
      "entropy": 5.51279993057251,
      "epoch": 2.623314429741651,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004310250492683126,
      "loss": 5.1196,
      "mean_token_accuracy": 0.1938912183046341,
      "num_tokens": 57603600.0,
      "step": 31225
    },
    {
      "entropy": 5.4563792705535885,
      "epoch": 2.6237345095568156,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043100347110162094,
      "loss": 5.1138,
      "mean_token_accuracy": 0.19500478804111482,
      "num_tokens": 57612371.0,
      "step": 31230
    },
    {
      "entropy": 5.5459216117858885,
      "epoch": 2.6241545893719804,
      "grad_norm": 1.125,
      "learning_rate": 0.000430981890171408,
      "loss": 5.1536,
      "mean_token_accuracy": 0.18651970475912094,
      "num_tokens": 57620994.0,
      "step": 31235
    },
    {
      "entropy": 5.5392234325408936,
      "epoch": 2.6245746691871457,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004309603064780562,
      "loss": 5.2191,
      "mean_token_accuracy": 0.1785485029220581,
      "num_tokens": 57630953.0,
      "step": 31240
    },
    {
      "entropy": 5.48131537437439,
      "epoch": 2.6249947490023104,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004309387200219479,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19073591232299805,
      "num_tokens": 57640352.0,
      "step": 31245
    },
    {
      "entropy": 5.572226810455322,
      "epoch": 2.6254148288174752,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004309171308034655,
      "loss": 5.1803,
      "mean_token_accuracy": 0.19182629585266114,
      "num_tokens": 57649204.0,
      "step": 31250
    },
    {
      "entropy": 5.458486032485962,
      "epoch": 2.6258349086326405,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043089553882299137,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20207718163728713,
      "num_tokens": 57658609.0,
      "step": 31255
    },
    {
      "entropy": 5.484911012649536,
      "epoch": 2.626254988447805,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043087394408090814,
      "loss": 5.1947,
      "mean_token_accuracy": 0.19208331704139708,
      "num_tokens": 57667896.0,
      "step": 31260
    },
    {
      "entropy": 5.505950164794922,
      "epoch": 2.62667506826297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004308523465775981,
      "loss": 5.1795,
      "mean_token_accuracy": 0.1888086676597595,
      "num_tokens": 57676470.0,
      "step": 31265
    },
    {
      "entropy": 5.523262977600098,
      "epoch": 2.627095148078135,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043083074631344414,
      "loss": 5.1136,
      "mean_token_accuracy": 0.1939445525407791,
      "num_tokens": 57685428.0,
      "step": 31270
    },
    {
      "entropy": 5.506783151626587,
      "epoch": 2.6275152278932996,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043080914328882865,
      "loss": 5.1465,
      "mean_token_accuracy": 0.19306448400020598,
      "num_tokens": 57694926.0,
      "step": 31275
    },
    {
      "entropy": 5.435008335113525,
      "epoch": 2.627935307708465,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043078753750413443,
      "loss": 5.1128,
      "mean_token_accuracy": 0.19753591269254683,
      "num_tokens": 57703336.0,
      "step": 31280
    },
    {
      "entropy": 5.492234086990356,
      "epoch": 2.6283553875236296,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004307659289597442,
      "loss": 5.1728,
      "mean_token_accuracy": 0.19070515185594558,
      "num_tokens": 57712386.0,
      "step": 31285
    },
    {
      "entropy": 5.509941434860229,
      "epoch": 2.6287754673387944,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043074431765604084,
      "loss": 5.1168,
      "mean_token_accuracy": 0.1922464296221733,
      "num_tokens": 57721742.0,
      "step": 31290
    },
    {
      "entropy": 5.506961393356323,
      "epoch": 2.629195547153959,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004307227035934069,
      "loss": 5.1961,
      "mean_token_accuracy": 0.18460369110107422,
      "num_tokens": 57731700.0,
      "step": 31295
    },
    {
      "entropy": 5.49363899230957,
      "epoch": 2.629615626969124,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004307010867722256,
      "loss": 5.1458,
      "mean_token_accuracy": 0.1884138062596321,
      "num_tokens": 57739801.0,
      "step": 31300
    },
    {
      "entropy": 5.493569183349609,
      "epoch": 2.630035706784289,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004306794671928797,
      "loss": 5.0907,
      "mean_token_accuracy": 0.1887332707643509,
      "num_tokens": 57748755.0,
      "step": 31305
    },
    {
      "entropy": 5.547189092636108,
      "epoch": 2.630455786599454,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043065784485575223,
      "loss": 5.2366,
      "mean_token_accuracy": 0.1887480765581131,
      "num_tokens": 57758809.0,
      "step": 31310
    },
    {
      "entropy": 5.519438982009888,
      "epoch": 2.630875866414619,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004306362197612261,
      "loss": 5.1912,
      "mean_token_accuracy": 0.1789963185787201,
      "num_tokens": 57768576.0,
      "step": 31315
    },
    {
      "entropy": 5.4799243927001955,
      "epoch": 2.6312959462297836,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004306145919096846,
      "loss": 5.2063,
      "mean_token_accuracy": 0.19029152989387513,
      "num_tokens": 57776954.0,
      "step": 31320
    },
    {
      "entropy": 5.544125938415528,
      "epoch": 2.6317160260449484,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043059296130151066,
      "loss": 5.1346,
      "mean_token_accuracy": 0.19830731600522994,
      "num_tokens": 57785048.0,
      "step": 31325
    },
    {
      "entropy": 5.561526536941528,
      "epoch": 2.6321361058601136,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043057132793708746,
      "loss": 5.156,
      "mean_token_accuracy": 0.19166335910558702,
      "num_tokens": 57794467.0,
      "step": 31330
    },
    {
      "entropy": 5.461753225326538,
      "epoch": 2.6325561856752784,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004305496918167984,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1968449667096138,
      "num_tokens": 57803342.0,
      "step": 31335
    },
    {
      "entropy": 5.557968235015869,
      "epoch": 2.632976265490443,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043052805294102657,
      "loss": 5.1838,
      "mean_token_accuracy": 0.1862136095762253,
      "num_tokens": 57813290.0,
      "step": 31340
    },
    {
      "entropy": 5.582530498504639,
      "epoch": 2.633396345305608,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004305064113101554,
      "loss": 5.1615,
      "mean_token_accuracy": 0.19139678478240968,
      "num_tokens": 57822440.0,
      "step": 31345
    },
    {
      "entropy": 5.444601011276245,
      "epoch": 2.6338164251207727,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004304847669245683,
      "loss": 5.0958,
      "mean_token_accuracy": 0.18614824414253234,
      "num_tokens": 57831673.0,
      "step": 31350
    },
    {
      "entropy": 5.516133260726929,
      "epoch": 2.634236504935938,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043046311978464847,
      "loss": 5.143,
      "mean_token_accuracy": 0.18707426637411118,
      "num_tokens": 57840506.0,
      "step": 31355
    },
    {
      "entropy": 5.467253875732422,
      "epoch": 2.6346565847511028,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043044146989077966,
      "loss": 5.1168,
      "mean_token_accuracy": 0.1910016119480133,
      "num_tokens": 57849376.0,
      "step": 31360
    },
    {
      "entropy": 5.554364109039307,
      "epoch": 2.6350766645662675,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004304198172433452,
      "loss": 5.2442,
      "mean_token_accuracy": 0.1866356164216995,
      "num_tokens": 57859161.0,
      "step": 31365
    },
    {
      "entropy": 5.535153388977051,
      "epoch": 2.6354967443814323,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004303981618427287,
      "loss": 5.0843,
      "mean_token_accuracy": 0.19121146947145462,
      "num_tokens": 57867598.0,
      "step": 31370
    },
    {
      "entropy": 5.56626296043396,
      "epoch": 2.635916824196597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004303765036893138,
      "loss": 5.2064,
      "mean_token_accuracy": 0.18948093950748443,
      "num_tokens": 57877569.0,
      "step": 31375
    },
    {
      "entropy": 5.552335166931153,
      "epoch": 2.6363369040117623,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004303548427834841,
      "loss": 5.2095,
      "mean_token_accuracy": 0.1865539252758026,
      "num_tokens": 57887104.0,
      "step": 31380
    },
    {
      "entropy": 5.544705581665039,
      "epoch": 2.636756983826927,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004303331791256234,
      "loss": 5.1215,
      "mean_token_accuracy": 0.19487340301275252,
      "num_tokens": 57895699.0,
      "step": 31385
    },
    {
      "entropy": 5.478884220123291,
      "epoch": 2.637177063642092,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043031151271611534,
      "loss": 5.1455,
      "mean_token_accuracy": 0.18921138644218444,
      "num_tokens": 57904997.0,
      "step": 31390
    },
    {
      "entropy": 5.464394474029541,
      "epoch": 2.637597143457257,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043028984355534394,
      "loss": 5.1638,
      "mean_token_accuracy": 0.19951582401990892,
      "num_tokens": 57913934.0,
      "step": 31395
    },
    {
      "entropy": 5.528600072860717,
      "epoch": 2.6380172232724215,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043026817164369273,
      "loss": 5.1381,
      "mean_token_accuracy": 0.1884427085518837,
      "num_tokens": 57923114.0,
      "step": 31400
    },
    {
      "entropy": 5.491101551055908,
      "epoch": 2.6384373030875867,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000430246496981546,
      "loss": 5.1759,
      "mean_token_accuracy": 0.18784166276454925,
      "num_tokens": 57932462.0,
      "step": 31405
    },
    {
      "entropy": 5.438311195373535,
      "epoch": 2.6388573829027515,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043022481956928743,
      "loss": 5.0737,
      "mean_token_accuracy": 0.18979546725749968,
      "num_tokens": 57940833.0,
      "step": 31410
    },
    {
      "entropy": 5.4632368087768555,
      "epoch": 2.6392774627179163,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043020313940730123,
      "loss": 5.093,
      "mean_token_accuracy": 0.19510050117969513,
      "num_tokens": 57949462.0,
      "step": 31415
    },
    {
      "entropy": 5.548546600341797,
      "epoch": 2.6396975425330815,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043018145649597114,
      "loss": 5.2749,
      "mean_token_accuracy": 0.18303299844264984,
      "num_tokens": 57958772.0,
      "step": 31420
    },
    {
      "entropy": 5.529941654205322,
      "epoch": 2.6401176223482463,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004301597708356815,
      "loss": 5.1534,
      "mean_token_accuracy": 0.17777338922023772,
      "num_tokens": 57969200.0,
      "step": 31425
    },
    {
      "entropy": 5.5661680698394775,
      "epoch": 2.640537702163411,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043013808242681645,
      "loss": 5.1761,
      "mean_token_accuracy": 0.19186543226242064,
      "num_tokens": 57977488.0,
      "step": 31430
    },
    {
      "entropy": 5.459603691101075,
      "epoch": 2.640957781978576,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004301163912697602,
      "loss": 5.0546,
      "mean_token_accuracy": 0.19187266379594803,
      "num_tokens": 57985870.0,
      "step": 31435
    },
    {
      "entropy": 5.52421498298645,
      "epoch": 2.6413778617937407,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004300946973648969,
      "loss": 5.2079,
      "mean_token_accuracy": 0.18509847819805145,
      "num_tokens": 57995378.0,
      "step": 31440
    },
    {
      "entropy": 5.52674880027771,
      "epoch": 2.641797941608906,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004300730007126109,
      "loss": 5.1856,
      "mean_token_accuracy": 0.18523213714361192,
      "num_tokens": 58005623.0,
      "step": 31445
    },
    {
      "entropy": 5.547564458847046,
      "epoch": 2.6422180214240707,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043005130131328664,
      "loss": 5.1573,
      "mean_token_accuracy": 0.1960933342576027,
      "num_tokens": 58014123.0,
      "step": 31450
    },
    {
      "entropy": 5.525696039199829,
      "epoch": 2.6426381012392355,
      "grad_norm": 1.25,
      "learning_rate": 0.00043002959916730834,
      "loss": 5.2007,
      "mean_token_accuracy": 0.1755317896604538,
      "num_tokens": 58023292.0,
      "step": 31455
    },
    {
      "entropy": 5.466460800170898,
      "epoch": 2.6430581810544003,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004300078942750604,
      "loss": 5.1,
      "mean_token_accuracy": 0.19207976311445235,
      "num_tokens": 58032165.0,
      "step": 31460
    },
    {
      "entropy": 5.51910629272461,
      "epoch": 2.643478260869565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004299861866369276,
      "loss": 5.0554,
      "mean_token_accuracy": 0.19714820981025696,
      "num_tokens": 58041518.0,
      "step": 31465
    },
    {
      "entropy": 5.519585800170899,
      "epoch": 2.6438983406847303,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042996447625329425,
      "loss": 5.2023,
      "mean_token_accuracy": 0.1846597746014595,
      "num_tokens": 58051414.0,
      "step": 31470
    },
    {
      "entropy": 5.477161598205567,
      "epoch": 2.644318420499895,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042994276312454505,
      "loss": 5.1141,
      "mean_token_accuracy": 0.18930020332336425,
      "num_tokens": 58060594.0,
      "step": 31475
    },
    {
      "entropy": 5.626159381866455,
      "epoch": 2.64473850031506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004299210472510645,
      "loss": 5.2559,
      "mean_token_accuracy": 0.17781059592962264,
      "num_tokens": 58070255.0,
      "step": 31480
    },
    {
      "entropy": 5.541515636444092,
      "epoch": 2.6451585801302246,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042989932863323746,
      "loss": 5.1776,
      "mean_token_accuracy": 0.1872272863984108,
      "num_tokens": 58079892.0,
      "step": 31485
    },
    {
      "entropy": 5.571437692642212,
      "epoch": 2.6455786599453894,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004298776072714484,
      "loss": 5.1607,
      "mean_token_accuracy": 0.19151849746704103,
      "num_tokens": 58088551.0,
      "step": 31490
    },
    {
      "entropy": 5.621867513656616,
      "epoch": 2.6459987397605547,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004298558831660824,
      "loss": 5.2987,
      "mean_token_accuracy": 0.180108006298542,
      "num_tokens": 58098189.0,
      "step": 31495
    },
    {
      "entropy": 5.5386425971984865,
      "epoch": 2.6464188195757194,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004298341563175242,
      "loss": 5.205,
      "mean_token_accuracy": 0.19051268249750136,
      "num_tokens": 58107980.0,
      "step": 31500
    },
    {
      "entropy": 5.592724800109863,
      "epoch": 2.6468388993908842,
      "grad_norm": 1.125,
      "learning_rate": 0.0004298124267261586,
      "loss": 5.2586,
      "mean_token_accuracy": 0.18016956597566605,
      "num_tokens": 58117084.0,
      "step": 31505
    },
    {
      "entropy": 5.528514289855957,
      "epoch": 2.647258979206049,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004297906943923706,
      "loss": 5.2039,
      "mean_token_accuracy": 0.18701286762952804,
      "num_tokens": 58126304.0,
      "step": 31510
    },
    {
      "entropy": 5.470707130432129,
      "epoch": 2.647679059021214,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004297689593165451,
      "loss": 5.027,
      "mean_token_accuracy": 0.1975795015692711,
      "num_tokens": 58134643.0,
      "step": 31515
    },
    {
      "entropy": 5.49492301940918,
      "epoch": 2.648099138836379,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004297472214990673,
      "loss": 5.047,
      "mean_token_accuracy": 0.2002097949385643,
      "num_tokens": 58142815.0,
      "step": 31520
    },
    {
      "entropy": 5.3864490509033205,
      "epoch": 2.648519218651544,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042972548094032207,
      "loss": 5.0705,
      "mean_token_accuracy": 0.20020551532506942,
      "num_tokens": 58151618.0,
      "step": 31525
    },
    {
      "entropy": 5.45513334274292,
      "epoch": 2.6489392984667086,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042970373764069465,
      "loss": 5.1327,
      "mean_token_accuracy": 0.1918591096997261,
      "num_tokens": 58160371.0,
      "step": 31530
    },
    {
      "entropy": 5.5550535202026365,
      "epoch": 2.6493593782818734,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042968199160057016,
      "loss": 5.1709,
      "mean_token_accuracy": 0.1932663083076477,
      "num_tokens": 58170536.0,
      "step": 31535
    },
    {
      "entropy": 5.618098402023316,
      "epoch": 2.649779458097038,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004296602428203338,
      "loss": 5.2622,
      "mean_token_accuracy": 0.18471622467041016,
      "num_tokens": 58181271.0,
      "step": 31540
    },
    {
      "entropy": 5.54141001701355,
      "epoch": 2.6501995379122034,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004296384913003709,
      "loss": 5.2058,
      "mean_token_accuracy": 0.18000854253768922,
      "num_tokens": 58192011.0,
      "step": 31545
    },
    {
      "entropy": 5.581263017654419,
      "epoch": 2.650619617727368,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042961673704106676,
      "loss": 5.1244,
      "mean_token_accuracy": 0.18924420475959777,
      "num_tokens": 58201407.0,
      "step": 31550
    },
    {
      "entropy": 5.505290889739991,
      "epoch": 2.651039697542533,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004295949800428067,
      "loss": 5.0828,
      "mean_token_accuracy": 0.19111973494291307,
      "num_tokens": 58211298.0,
      "step": 31555
    },
    {
      "entropy": 5.550792646408081,
      "epoch": 2.651459777357698,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042957322030597625,
      "loss": 5.2253,
      "mean_token_accuracy": 0.18703700304031373,
      "num_tokens": 58220498.0,
      "step": 31560
    },
    {
      "entropy": 5.512788867950439,
      "epoch": 2.651879857172863,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042955145783096077,
      "loss": 5.1238,
      "mean_token_accuracy": 0.19135883897542955,
      "num_tokens": 58229620.0,
      "step": 31565
    },
    {
      "entropy": 5.518397951126099,
      "epoch": 2.652299936988028,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004295296926181458,
      "loss": 5.1294,
      "mean_token_accuracy": 0.1978673979640007,
      "num_tokens": 58237835.0,
      "step": 31570
    },
    {
      "entropy": 5.532335042953491,
      "epoch": 2.6527200168031926,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042950792466791687,
      "loss": 5.1169,
      "mean_token_accuracy": 0.19194632172584533,
      "num_tokens": 58246562.0,
      "step": 31575
    },
    {
      "entropy": 5.546910381317138,
      "epoch": 2.6531400966183574,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004294861539806596,
      "loss": 5.1668,
      "mean_token_accuracy": 0.1875628724694252,
      "num_tokens": 58256033.0,
      "step": 31580
    },
    {
      "entropy": 5.501591348648072,
      "epoch": 2.6535601764335226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004294643805567597,
      "loss": 5.1516,
      "mean_token_accuracy": 0.19477391093969346,
      "num_tokens": 58265407.0,
      "step": 31585
    },
    {
      "entropy": 5.533048915863037,
      "epoch": 2.6539802562486874,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004294426043966029,
      "loss": 5.121,
      "mean_token_accuracy": 0.18373804688453674,
      "num_tokens": 58275091.0,
      "step": 31590
    },
    {
      "entropy": 5.506830835342408,
      "epoch": 2.654400336063852,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042942082550057476,
      "loss": 5.156,
      "mean_token_accuracy": 0.2026807799935341,
      "num_tokens": 58283669.0,
      "step": 31595
    },
    {
      "entropy": 5.4901350975036625,
      "epoch": 2.654820415879017,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004293990438690613,
      "loss": 5.1492,
      "mean_token_accuracy": 0.1814541697502136,
      "num_tokens": 58293482.0,
      "step": 31600
    },
    {
      "entropy": 5.576620626449585,
      "epoch": 2.6552404956941817,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004293772595024482,
      "loss": 5.2221,
      "mean_token_accuracy": 0.18610115796327592,
      "num_tokens": 58302098.0,
      "step": 31605
    },
    {
      "entropy": 5.537191152572632,
      "epoch": 2.655660575509347,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004293554724011215,
      "loss": 5.0773,
      "mean_token_accuracy": 0.1913406491279602,
      "num_tokens": 58310785.0,
      "step": 31610
    },
    {
      "entropy": 5.470881843566895,
      "epoch": 2.6560806553245118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042933368256546704,
      "loss": 5.0962,
      "mean_token_accuracy": 0.19509725719690324,
      "num_tokens": 58319197.0,
      "step": 31615
    },
    {
      "entropy": 5.441579341888428,
      "epoch": 2.6565007351396766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042931188999587093,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19167399257421494,
      "num_tokens": 58328136.0,
      "step": 31620
    },
    {
      "entropy": 5.515025520324707,
      "epoch": 2.6569208149548413,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004292900946927191,
      "loss": 5.2913,
      "mean_token_accuracy": 0.17810261249542236,
      "num_tokens": 58337735.0,
      "step": 31625
    },
    {
      "entropy": 5.5356285572052,
      "epoch": 2.657340894770006,
      "grad_norm": 1.125,
      "learning_rate": 0.0004292682966563977,
      "loss": 5.124,
      "mean_token_accuracy": 0.1861994281411171,
      "num_tokens": 58347936.0,
      "step": 31630
    },
    {
      "entropy": 5.536416387557983,
      "epoch": 2.6577609745851714,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042924649588729286,
      "loss": 5.1942,
      "mean_token_accuracy": 0.18697964996099473,
      "num_tokens": 58357111.0,
      "step": 31635
    },
    {
      "entropy": 5.494350099563599,
      "epoch": 2.658181054400336,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042922469238579076,
      "loss": 5.0647,
      "mean_token_accuracy": 0.198074047267437,
      "num_tokens": 58366396.0,
      "step": 31640
    },
    {
      "entropy": 5.544517135620117,
      "epoch": 2.658601134215501,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004292028861522777,
      "loss": 5.1751,
      "mean_token_accuracy": 0.1899146243929863,
      "num_tokens": 58375197.0,
      "step": 31645
    },
    {
      "entropy": 5.543887186050415,
      "epoch": 2.6590212140306657,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042918107718713987,
      "loss": 5.247,
      "mean_token_accuracy": 0.18814198970794677,
      "num_tokens": 58383937.0,
      "step": 31650
    },
    {
      "entropy": 5.532442283630371,
      "epoch": 2.6594412938458305,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004291592654907637,
      "loss": 5.1535,
      "mean_token_accuracy": 0.18489030003547668,
      "num_tokens": 58392855.0,
      "step": 31655
    },
    {
      "entropy": 5.511371612548828,
      "epoch": 2.6598613736609957,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042913745106353543,
      "loss": 5.0876,
      "mean_token_accuracy": 0.195082788169384,
      "num_tokens": 58402359.0,
      "step": 31660
    },
    {
      "entropy": 5.450203800201416,
      "epoch": 2.6602814534761605,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004291156339058417,
      "loss": 5.052,
      "mean_token_accuracy": 0.18983241617679597,
      "num_tokens": 58411377.0,
      "step": 31665
    },
    {
      "entropy": 5.432679939270019,
      "epoch": 2.6607015332913253,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042909381401806874,
      "loss": 5.0558,
      "mean_token_accuracy": 0.19732677489519118,
      "num_tokens": 58420535.0,
      "step": 31670
    },
    {
      "entropy": 5.469066143035889,
      "epoch": 2.66112161310649,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042907199140060327,
      "loss": 5.1018,
      "mean_token_accuracy": 0.1889919012784958,
      "num_tokens": 58429122.0,
      "step": 31675
    },
    {
      "entropy": 5.538174819946289,
      "epoch": 2.661541692921655,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004290501660538318,
      "loss": 5.1015,
      "mean_token_accuracy": 0.18926112800836564,
      "num_tokens": 58437870.0,
      "step": 31680
    },
    {
      "entropy": 5.467972469329834,
      "epoch": 2.66196177273682,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042902833797814105,
      "loss": 5.143,
      "mean_token_accuracy": 0.19019912630319596,
      "num_tokens": 58446411.0,
      "step": 31685
    },
    {
      "entropy": 5.445688676834107,
      "epoch": 2.662381852551985,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004290065071739175,
      "loss": 5.0914,
      "mean_token_accuracy": 0.1906509056687355,
      "num_tokens": 58454629.0,
      "step": 31690
    },
    {
      "entropy": 5.581258392333984,
      "epoch": 2.6628019323671497,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042898467364154796,
      "loss": 5.1776,
      "mean_token_accuracy": 0.1860459938645363,
      "num_tokens": 58464340.0,
      "step": 31695
    },
    {
      "entropy": 5.531717538833618,
      "epoch": 2.663222012182315,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042896283738141927,
      "loss": 5.035,
      "mean_token_accuracy": 0.19607675522565843,
      "num_tokens": 58473540.0,
      "step": 31700
    },
    {
      "entropy": 5.4915813446044925,
      "epoch": 2.6636420919974793,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042894099839391826,
      "loss": 5.0922,
      "mean_token_accuracy": 0.19036860316991805,
      "num_tokens": 58482309.0,
      "step": 31705
    },
    {
      "entropy": 5.49208083152771,
      "epoch": 2.6640621718126445,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042891915667943157,
      "loss": 5.2099,
      "mean_token_accuracy": 0.19033290296792985,
      "num_tokens": 58492019.0,
      "step": 31710
    },
    {
      "entropy": 5.468537282943726,
      "epoch": 2.6644822516278093,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042889731223834637,
      "loss": 5.1293,
      "mean_token_accuracy": 0.19465629607439042,
      "num_tokens": 58501016.0,
      "step": 31715
    },
    {
      "entropy": 5.488776969909668,
      "epoch": 2.664902331442974,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004288754650710496,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19554009586572646,
      "num_tokens": 58511042.0,
      "step": 31720
    },
    {
      "entropy": 5.5140156745910645,
      "epoch": 2.6653224112581393,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042885361517792817,
      "loss": 5.1093,
      "mean_token_accuracy": 0.1860186979174614,
      "num_tokens": 58519446.0,
      "step": 31725
    },
    {
      "entropy": 5.555151176452637,
      "epoch": 2.665742491073304,
      "grad_norm": 1.0,
      "learning_rate": 0.00042883176255936907,
      "loss": 5.2163,
      "mean_token_accuracy": 0.1857264071702957,
      "num_tokens": 58529573.0,
      "step": 31730
    },
    {
      "entropy": 5.606505441665649,
      "epoch": 2.666162570888469,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004288099072157596,
      "loss": 5.3627,
      "mean_token_accuracy": 0.17831559330224991,
      "num_tokens": 58538602.0,
      "step": 31735
    },
    {
      "entropy": 5.5883471965789795,
      "epoch": 2.6665826507036337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004287880491474868,
      "loss": 5.2016,
      "mean_token_accuracy": 0.18650546073913574,
      "num_tokens": 58547776.0,
      "step": 31740
    },
    {
      "entropy": 5.4761754989624025,
      "epoch": 2.6670027305187984,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042876618835493784,
      "loss": 5.0988,
      "mean_token_accuracy": 0.18983179181814194,
      "num_tokens": 58556734.0,
      "step": 31745
    },
    {
      "entropy": 5.444047260284424,
      "epoch": 2.6674228103339637,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004287443248385001,
      "loss": 5.1211,
      "mean_token_accuracy": 0.19250666797161103,
      "num_tokens": 58567549.0,
      "step": 31750
    },
    {
      "entropy": 5.49845175743103,
      "epoch": 2.6678428901491285,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004287224585985608,
      "loss": 5.1074,
      "mean_token_accuracy": 0.1919595092535019,
      "num_tokens": 58577007.0,
      "step": 31755
    },
    {
      "entropy": 5.527373886108398,
      "epoch": 2.6682629699642932,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004287005896355072,
      "loss": 5.1591,
      "mean_token_accuracy": 0.19173274785280228,
      "num_tokens": 58586140.0,
      "step": 31760
    },
    {
      "entropy": 5.551871871948242,
      "epoch": 2.668683049779458,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042867871794972695,
      "loss": 5.2691,
      "mean_token_accuracy": 0.18577557653188706,
      "num_tokens": 58595382.0,
      "step": 31765
    },
    {
      "entropy": 5.543910121917724,
      "epoch": 2.669103129594623,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042865684354160736,
      "loss": 5.197,
      "mean_token_accuracy": 0.19011869430541992,
      "num_tokens": 58603525.0,
      "step": 31770
    },
    {
      "entropy": 5.490243148803711,
      "epoch": 2.669523209409788,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004286349664115358,
      "loss": 5.0961,
      "mean_token_accuracy": 0.20219926834106444,
      "num_tokens": 58611720.0,
      "step": 31775
    },
    {
      "entropy": 5.557237052917481,
      "epoch": 2.669943289224953,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00042861308655989987,
      "loss": 5.1884,
      "mean_token_accuracy": 0.19034357517957687,
      "num_tokens": 58620961.0,
      "step": 31780
    },
    {
      "entropy": 5.507651662826538,
      "epoch": 2.6703633690401176,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042859120398708734,
      "loss": 5.2059,
      "mean_token_accuracy": 0.1913668543100357,
      "num_tokens": 58630384.0,
      "step": 31785
    },
    {
      "entropy": 5.511829662322998,
      "epoch": 2.6707834488552824,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004285693186934857,
      "loss": 5.0941,
      "mean_token_accuracy": 0.1957175463438034,
      "num_tokens": 58640245.0,
      "step": 31790
    },
    {
      "entropy": 5.504929876327514,
      "epoch": 2.671203528670447,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042854743067948257,
      "loss": 5.1505,
      "mean_token_accuracy": 0.18726368099451066,
      "num_tokens": 58649780.0,
      "step": 31795
    },
    {
      "entropy": 5.558163213729858,
      "epoch": 2.6716236084856124,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004285255399454659,
      "loss": 5.2244,
      "mean_token_accuracy": 0.17595206201076508,
      "num_tokens": 58659322.0,
      "step": 31800
    },
    {
      "entropy": 5.523754262924195,
      "epoch": 2.672043688300777,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042850364649182324,
      "loss": 5.1399,
      "mean_token_accuracy": 0.18683224618434907,
      "num_tokens": 58668995.0,
      "step": 31805
    },
    {
      "entropy": 5.589215469360352,
      "epoch": 2.672463768115942,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004284817503189426,
      "loss": 5.1347,
      "mean_token_accuracy": 0.1913974776864052,
      "num_tokens": 58677329.0,
      "step": 31810
    },
    {
      "entropy": 5.4991631507873535,
      "epoch": 2.672883847931107,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004284598514272117,
      "loss": 5.1016,
      "mean_token_accuracy": 0.19326708018779754,
      "num_tokens": 58686619.0,
      "step": 31815
    },
    {
      "entropy": 5.461451625823974,
      "epoch": 2.6733039277462716,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042843794981701865,
      "loss": 5.1349,
      "mean_token_accuracy": 0.19421766996383666,
      "num_tokens": 58695637.0,
      "step": 31820
    },
    {
      "entropy": 5.586685514450073,
      "epoch": 2.673724007561437,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042841604548875126,
      "loss": 5.2356,
      "mean_token_accuracy": 0.18544958978891374,
      "num_tokens": 58706161.0,
      "step": 31825
    },
    {
      "entropy": 5.513773107528687,
      "epoch": 2.6741440873766016,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042839413844279755,
      "loss": 5.0778,
      "mean_token_accuracy": 0.20160541385412217,
      "num_tokens": 58714368.0,
      "step": 31830
    },
    {
      "entropy": 5.4690453052520756,
      "epoch": 2.6745641671917664,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004283722286795458,
      "loss": 5.1102,
      "mean_token_accuracy": 0.1941131353378296,
      "num_tokens": 58723518.0,
      "step": 31835
    },
    {
      "entropy": 5.4441859245300295,
      "epoch": 2.674984247006931,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004283503161993839,
      "loss": 5.0548,
      "mean_token_accuracy": 0.1935894027352333,
      "num_tokens": 58732533.0,
      "step": 31840
    },
    {
      "entropy": 5.490504264831543,
      "epoch": 2.675404326822096,
      "grad_norm": 1.125,
      "learning_rate": 0.0004283284010027002,
      "loss": 5.2046,
      "mean_token_accuracy": 0.18616004586219786,
      "num_tokens": 58741373.0,
      "step": 31845
    },
    {
      "entropy": 5.503035545349121,
      "epoch": 2.675824406637261,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042830648308988276,
      "loss": 5.1461,
      "mean_token_accuracy": 0.1915808841586113,
      "num_tokens": 58750867.0,
      "step": 31850
    },
    {
      "entropy": 5.4821981430053714,
      "epoch": 2.676244486452426,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004282845624613199,
      "loss": 5.0554,
      "mean_token_accuracy": 0.19006980657577516,
      "num_tokens": 58759389.0,
      "step": 31855
    },
    {
      "entropy": 5.517873668670655,
      "epoch": 2.6766645662675908,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042826263911740006,
      "loss": 5.1914,
      "mean_token_accuracy": 0.1943967491388321,
      "num_tokens": 58768034.0,
      "step": 31860
    },
    {
      "entropy": 5.428678226470947,
      "epoch": 2.677084646082756,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042824071305851136,
      "loss": 5.1412,
      "mean_token_accuracy": 0.19341378509998322,
      "num_tokens": 58777651.0,
      "step": 31865
    },
    {
      "entropy": 5.490052032470703,
      "epoch": 2.6775047258979208,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004282187842850424,
      "loss": 5.1239,
      "mean_token_accuracy": 0.19190748631954194,
      "num_tokens": 58787490.0,
      "step": 31870
    },
    {
      "entropy": 5.5302211284637455,
      "epoch": 2.6779248057130856,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042819685279738153,
      "loss": 5.1046,
      "mean_token_accuracy": 0.19678151607513428,
      "num_tokens": 58797134.0,
      "step": 31875
    },
    {
      "entropy": 5.521770048141479,
      "epoch": 2.6783448855282503,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042817491859591735,
      "loss": 5.1148,
      "mean_token_accuracy": 0.19922269880771637,
      "num_tokens": 58805500.0,
      "step": 31880
    },
    {
      "entropy": 5.4705085277557375,
      "epoch": 2.678764965343415,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042815298168103833,
      "loss": 5.1396,
      "mean_token_accuracy": 0.19213962703943252,
      "num_tokens": 58814429.0,
      "step": 31885
    },
    {
      "entropy": 5.488186025619507,
      "epoch": 2.6791850451585804,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004281310420531331,
      "loss": 5.0806,
      "mean_token_accuracy": 0.18901498019695281,
      "num_tokens": 58824250.0,
      "step": 31890
    },
    {
      "entropy": 5.552608585357666,
      "epoch": 2.679605124973745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004281090997125903,
      "loss": 5.2244,
      "mean_token_accuracy": 0.1826266899704933,
      "num_tokens": 58833773.0,
      "step": 31895
    },
    {
      "entropy": 5.5744469165802,
      "epoch": 2.68002520478891,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004280871546597987,
      "loss": 5.177,
      "mean_token_accuracy": 0.1914798676967621,
      "num_tokens": 58843557.0,
      "step": 31900
    },
    {
      "entropy": 5.608108043670654,
      "epoch": 2.6804452846040747,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004280652068951469,
      "loss": 5.2792,
      "mean_token_accuracy": 0.17604817003011702,
      "num_tokens": 58853699.0,
      "step": 31905
    },
    {
      "entropy": 5.490276527404785,
      "epoch": 2.6808653644192395,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042804325641902384,
      "loss": 5.076,
      "mean_token_accuracy": 0.19724421799182892,
      "num_tokens": 58862355.0,
      "step": 31910
    },
    {
      "entropy": 5.412453508377075,
      "epoch": 2.6812854442344047,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042802130323181824,
      "loss": 5.0623,
      "mean_token_accuracy": 0.1947748228907585,
      "num_tokens": 58870906.0,
      "step": 31915
    },
    {
      "entropy": 5.484709692001343,
      "epoch": 2.6817055240495695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004279993473339191,
      "loss": 5.0977,
      "mean_token_accuracy": 0.19399794638156892,
      "num_tokens": 58879843.0,
      "step": 31920
    },
    {
      "entropy": 5.499013233184814,
      "epoch": 2.6821256038647343,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004279773887257153,
      "loss": 5.0827,
      "mean_token_accuracy": 0.19315697699785234,
      "num_tokens": 58888837.0,
      "step": 31925
    },
    {
      "entropy": 5.549001407623291,
      "epoch": 2.682545683679899,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004279554274075958,
      "loss": 5.2403,
      "mean_token_accuracy": 0.18282705396413804,
      "num_tokens": 58898414.0,
      "step": 31930
    },
    {
      "entropy": 5.459571647644043,
      "epoch": 2.682965763495064,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004279334633799497,
      "loss": 5.0706,
      "mean_token_accuracy": 0.18843164592981337,
      "num_tokens": 58906877.0,
      "step": 31935
    },
    {
      "entropy": 5.487481737136841,
      "epoch": 2.683385843310229,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042791149664316605,
      "loss": 5.0874,
      "mean_token_accuracy": 0.19402543306350709,
      "num_tokens": 58915911.0,
      "step": 31940
    },
    {
      "entropy": 5.515584135055542,
      "epoch": 2.683805923125394,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004278895271976339,
      "loss": 5.1302,
      "mean_token_accuracy": 0.18875904828310014,
      "num_tokens": 58924992.0,
      "step": 31945
    },
    {
      "entropy": 5.4821250438690186,
      "epoch": 2.6842260029405587,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042786755504374254,
      "loss": 5.1696,
      "mean_token_accuracy": 0.19321332424879073,
      "num_tokens": 58933362.0,
      "step": 31950
    },
    {
      "entropy": 5.507419776916504,
      "epoch": 2.6846460827557235,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004278455801818812,
      "loss": 5.1208,
      "mean_token_accuracy": 0.19065092951059343,
      "num_tokens": 58942810.0,
      "step": 31955
    },
    {
      "entropy": 5.56959867477417,
      "epoch": 2.6850661625708883,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004278236026124391,
      "loss": 5.1931,
      "mean_token_accuracy": 0.19113135635852813,
      "num_tokens": 58952178.0,
      "step": 31960
    },
    {
      "entropy": 5.551672458648682,
      "epoch": 2.6854862423860535,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042780162233580556,
      "loss": 5.1006,
      "mean_token_accuracy": 0.1909159615635872,
      "num_tokens": 58960653.0,
      "step": 31965
    },
    {
      "entropy": 5.470548868179321,
      "epoch": 2.6859063222012183,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042777963935236997,
      "loss": 5.002,
      "mean_token_accuracy": 0.19552794247865676,
      "num_tokens": 58969201.0,
      "step": 31970
    },
    {
      "entropy": 5.47041802406311,
      "epoch": 2.686326402016383,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004277576536625217,
      "loss": 5.2218,
      "mean_token_accuracy": 0.18826536238193511,
      "num_tokens": 58978547.0,
      "step": 31975
    },
    {
      "entropy": 5.511387777328491,
      "epoch": 2.686746481831548,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004277356652666502,
      "loss": 5.1493,
      "mean_token_accuracy": 0.18864538818597792,
      "num_tokens": 58988872.0,
      "step": 31980
    },
    {
      "entropy": 5.5427467823028564,
      "epoch": 2.6871665616467126,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004277136741651452,
      "loss": 5.1107,
      "mean_token_accuracy": 0.18546447902917862,
      "num_tokens": 58997375.0,
      "step": 31985
    },
    {
      "entropy": 5.568742132186889,
      "epoch": 2.687586641461878,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042769168035839593,
      "loss": 5.1837,
      "mean_token_accuracy": 0.18966754674911498,
      "num_tokens": 59006138.0,
      "step": 31990
    },
    {
      "entropy": 5.42591347694397,
      "epoch": 2.6880067212770427,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004276696838467923,
      "loss": 5.0459,
      "mean_token_accuracy": 0.19670403152704238,
      "num_tokens": 59015352.0,
      "step": 31995
    },
    {
      "entropy": 5.4943183898925785,
      "epoch": 2.6884268010922074,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004276476846307237,
      "loss": 5.1154,
      "mean_token_accuracy": 0.19429924488067626,
      "num_tokens": 59024150.0,
      "step": 32000
    },
    {
      "entropy": 5.390085935592651,
      "epoch": 2.6888468809073727,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004276256827105802,
      "loss": 5.04,
      "mean_token_accuracy": 0.19955824166536332,
      "num_tokens": 59033031.0,
      "step": 32005
    },
    {
      "entropy": 5.561896705627442,
      "epoch": 2.689266960722537,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004276036780867511,
      "loss": 5.2183,
      "mean_token_accuracy": 0.18492126911878587,
      "num_tokens": 59041777.0,
      "step": 32010
    },
    {
      "entropy": 5.551880693435669,
      "epoch": 2.6896870405377022,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042758167075962646,
      "loss": 5.1278,
      "mean_token_accuracy": 0.20371403992176057,
      "num_tokens": 59049954.0,
      "step": 32015
    },
    {
      "entropy": 5.468247318267823,
      "epoch": 2.690107120352867,
      "grad_norm": 1.421875,
      "learning_rate": 0.00042755966072959613,
      "loss": 5.2168,
      "mean_token_accuracy": 0.18303443342447281,
      "num_tokens": 59058490.0,
      "step": 32020
    },
    {
      "entropy": 5.519580602645874,
      "epoch": 2.690527200168032,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042753764799705,
      "loss": 5.1527,
      "mean_token_accuracy": 0.1907810941338539,
      "num_tokens": 59066855.0,
      "step": 32025
    },
    {
      "entropy": 5.530676746368409,
      "epoch": 2.690947279983197,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004275156325623779,
      "loss": 5.1392,
      "mean_token_accuracy": 0.18832884579896927,
      "num_tokens": 59077160.0,
      "step": 32030
    },
    {
      "entropy": 5.555582189559937,
      "epoch": 2.691367359798362,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004274936144259699,
      "loss": 5.1739,
      "mean_token_accuracy": 0.18896360993385314,
      "num_tokens": 59086001.0,
      "step": 32035
    },
    {
      "entropy": 5.447295379638672,
      "epoch": 2.6917874396135266,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042747159358821614,
      "loss": 5.1073,
      "mean_token_accuracy": 0.1876748189330101,
      "num_tokens": 59095140.0,
      "step": 32040
    },
    {
      "entropy": 5.471684885025025,
      "epoch": 2.6922075194286914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004274495700495065,
      "loss": 5.0475,
      "mean_token_accuracy": 0.1930280938744545,
      "num_tokens": 59103832.0,
      "step": 32045
    },
    {
      "entropy": 5.468412208557129,
      "epoch": 2.692627599243856,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042742754381023136,
      "loss": 5.0555,
      "mean_token_accuracy": 0.19336768090724946,
      "num_tokens": 59112555.0,
      "step": 32050
    },
    {
      "entropy": 5.488690948486328,
      "epoch": 2.6930476790590214,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042740551487078067,
      "loss": 5.0829,
      "mean_token_accuracy": 0.18846149146556854,
      "num_tokens": 59122001.0,
      "step": 32055
    },
    {
      "entropy": 5.472227764129639,
      "epoch": 2.693467758874186,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004273834832315448,
      "loss": 5.1006,
      "mean_token_accuracy": 0.19156733453273772,
      "num_tokens": 59130396.0,
      "step": 32060
    },
    {
      "entropy": 5.4928569316864015,
      "epoch": 2.693887838689351,
      "grad_norm": 1.078125,
      "learning_rate": 0.000427361448892914,
      "loss": 5.1956,
      "mean_token_accuracy": 0.18581036031246184,
      "num_tokens": 59139877.0,
      "step": 32065
    },
    {
      "entropy": 5.472042179107666,
      "epoch": 2.694307918504516,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042733941185527845,
      "loss": 5.1303,
      "mean_token_accuracy": 0.19105488508939744,
      "num_tokens": 59148079.0,
      "step": 32070
    },
    {
      "entropy": 5.536334753036499,
      "epoch": 2.6947279983196806,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042731737211902887,
      "loss": 5.0417,
      "mean_token_accuracy": 0.20235313922166825,
      "num_tokens": 59156469.0,
      "step": 32075
    },
    {
      "entropy": 5.563534736633301,
      "epoch": 2.695148078134846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004272953296845554,
      "loss": 5.1169,
      "mean_token_accuracy": 0.1924784228205681,
      "num_tokens": 59166004.0,
      "step": 32080
    },
    {
      "entropy": 5.507740592956543,
      "epoch": 2.6955681579500106,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042727328455224856,
      "loss": 5.1654,
      "mean_token_accuracy": 0.1915137767791748,
      "num_tokens": 59176669.0,
      "step": 32085
    },
    {
      "entropy": 5.470322751998902,
      "epoch": 2.6959882377651754,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042725123672249883,
      "loss": 5.1646,
      "mean_token_accuracy": 0.18390549272298812,
      "num_tokens": 59186434.0,
      "step": 32090
    },
    {
      "entropy": 5.489650011062622,
      "epoch": 2.69640831758034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042722918619569693,
      "loss": 5.1216,
      "mean_token_accuracy": 0.19243430495262145,
      "num_tokens": 59196959.0,
      "step": 32095
    },
    {
      "entropy": 5.534192276000977,
      "epoch": 2.696828397395505,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042720713297223335,
      "loss": 5.1523,
      "mean_token_accuracy": 0.19676709175109863,
      "num_tokens": 59206193.0,
      "step": 32100
    },
    {
      "entropy": 5.561839771270752,
      "epoch": 2.69724847721067,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004271850770524988,
      "loss": 5.183,
      "mean_token_accuracy": 0.19080325067043305,
      "num_tokens": 59214841.0,
      "step": 32105
    },
    {
      "entropy": 5.509930801391602,
      "epoch": 2.697668557025835,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000427163018436884,
      "loss": 5.0408,
      "mean_token_accuracy": 0.20583930015563964,
      "num_tokens": 59223929.0,
      "step": 32110
    },
    {
      "entropy": 5.442484712600708,
      "epoch": 2.6980886368409998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004271409571257797,
      "loss": 5.1038,
      "mean_token_accuracy": 0.1960267961025238,
      "num_tokens": 59234159.0,
      "step": 32115
    },
    {
      "entropy": 5.500023555755615,
      "epoch": 2.6985087166561645,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004271188931195766,
      "loss": 5.1634,
      "mean_token_accuracy": 0.1892283245921135,
      "num_tokens": 59243642.0,
      "step": 32120
    },
    {
      "entropy": 5.515969085693359,
      "epoch": 2.6989287964713293,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004270968264186656,
      "loss": 5.1816,
      "mean_token_accuracy": 0.18755628317594528,
      "num_tokens": 59253164.0,
      "step": 32125
    },
    {
      "entropy": 5.5666951656341555,
      "epoch": 2.6993488762864946,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042707475702343775,
      "loss": 5.1477,
      "mean_token_accuracy": 0.19343290627002716,
      "num_tokens": 59261697.0,
      "step": 32130
    },
    {
      "entropy": 5.585147333145142,
      "epoch": 2.6997689561016593,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042705268493428385,
      "loss": 5.2078,
      "mean_token_accuracy": 0.19036654233932496,
      "num_tokens": 59271583.0,
      "step": 32135
    },
    {
      "entropy": 5.4965235710144045,
      "epoch": 2.700189035916824,
      "grad_norm": 1.125,
      "learning_rate": 0.00042703061015159495,
      "loss": 5.1042,
      "mean_token_accuracy": 0.1912648782134056,
      "num_tokens": 59281024.0,
      "step": 32140
    },
    {
      "entropy": 5.463832712173462,
      "epoch": 2.700609115731989,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042700853267576205,
      "loss": 5.0757,
      "mean_token_accuracy": 0.1989748328924179,
      "num_tokens": 59289719.0,
      "step": 32145
    },
    {
      "entropy": 5.503789377212525,
      "epoch": 2.7010291955471537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042698645250717623,
      "loss": 5.0617,
      "mean_token_accuracy": 0.19962834119796752,
      "num_tokens": 59299242.0,
      "step": 32150
    },
    {
      "entropy": 5.531918334960937,
      "epoch": 2.701449275362319,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004269643696462287,
      "loss": 5.2149,
      "mean_token_accuracy": 0.1846102461218834,
      "num_tokens": 59308075.0,
      "step": 32155
    },
    {
      "entropy": 5.526486825942993,
      "epoch": 2.7018693551774837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042694228409331065,
      "loss": 5.1652,
      "mean_token_accuracy": 0.1826823204755783,
      "num_tokens": 59317347.0,
      "step": 32160
    },
    {
      "entropy": 5.510262823104858,
      "epoch": 2.7022894349926485,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042692019584881325,
      "loss": 5.1737,
      "mean_token_accuracy": 0.1912167489528656,
      "num_tokens": 59327237.0,
      "step": 32165
    },
    {
      "entropy": 5.520627403259278,
      "epoch": 2.7027095148078137,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004268981049131278,
      "loss": 5.0402,
      "mean_token_accuracy": 0.196823912858963,
      "num_tokens": 59335769.0,
      "step": 32170
    },
    {
      "entropy": 5.497651195526123,
      "epoch": 2.7031295946229785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004268760112866456,
      "loss": 5.1285,
      "mean_token_accuracy": 0.20023357421159743,
      "num_tokens": 59344677.0,
      "step": 32175
    },
    {
      "entropy": 5.5516383171081545,
      "epoch": 2.7035496744381433,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042685391496975804,
      "loss": 5.1807,
      "mean_token_accuracy": 0.18447633385658263,
      "num_tokens": 59354089.0,
      "step": 32180
    },
    {
      "entropy": 5.5661173343658445,
      "epoch": 2.703969754253308,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004268318159628567,
      "loss": 5.2528,
      "mean_token_accuracy": 0.18304607421159744,
      "num_tokens": 59362965.0,
      "step": 32185
    },
    {
      "entropy": 5.559704208374024,
      "epoch": 2.704389834068473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004268097142663329,
      "loss": 5.1307,
      "mean_token_accuracy": 0.18512456268072128,
      "num_tokens": 59372137.0,
      "step": 32190
    },
    {
      "entropy": 5.50511736869812,
      "epoch": 2.704809913883638,
      "grad_norm": 1.25,
      "learning_rate": 0.0004267876098805781,
      "loss": 5.096,
      "mean_token_accuracy": 0.1917126312851906,
      "num_tokens": 59381269.0,
      "step": 32195
    },
    {
      "entropy": 5.437407398223877,
      "epoch": 2.705229993698803,
      "grad_norm": 1.109375,
      "learning_rate": 0.000426765502805984,
      "loss": 5.0661,
      "mean_token_accuracy": 0.19924466460943221,
      "num_tokens": 59389713.0,
      "step": 32200
    },
    {
      "entropy": 5.545135164260865,
      "epoch": 2.7056500735139677,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042674339304294215,
      "loss": 5.1667,
      "mean_token_accuracy": 0.1919005483388901,
      "num_tokens": 59399756.0,
      "step": 32205
    },
    {
      "entropy": 5.543938827514649,
      "epoch": 2.7060701533291325,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004267212805918442,
      "loss": 5.1285,
      "mean_token_accuracy": 0.19105158895254135,
      "num_tokens": 59408504.0,
      "step": 32210
    },
    {
      "entropy": 5.413490104675293,
      "epoch": 2.7064902331442973,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042669916545308196,
      "loss": 5.0371,
      "mean_token_accuracy": 0.1940462574362755,
      "num_tokens": 59417241.0,
      "step": 32215
    },
    {
      "entropy": 5.38635025024414,
      "epoch": 2.7069103129594625,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004266770476270471,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19508740305900574,
      "num_tokens": 59426730.0,
      "step": 32220
    },
    {
      "entropy": 5.436523675918579,
      "epoch": 2.7073303927746273,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004266549271141314,
      "loss": 5.011,
      "mean_token_accuracy": 0.20319963097572327,
      "num_tokens": 59435197.0,
      "step": 32225
    },
    {
      "entropy": 5.412104511260987,
      "epoch": 2.707750472589792,
      "grad_norm": 1.0,
      "learning_rate": 0.00042663280391472676,
      "loss": 5.0888,
      "mean_token_accuracy": 0.1914111077785492,
      "num_tokens": 59444684.0,
      "step": 32230
    },
    {
      "entropy": 5.446088886260986,
      "epoch": 2.708170552404957,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004266106780292251,
      "loss": 5.0278,
      "mean_token_accuracy": 0.19228901118040084,
      "num_tokens": 59453869.0,
      "step": 32235
    },
    {
      "entropy": 5.478064060211182,
      "epoch": 2.7085906322201216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042658854945801835,
      "loss": 5.0853,
      "mean_token_accuracy": 0.1885649636387825,
      "num_tokens": 59462713.0,
      "step": 32240
    },
    {
      "entropy": 5.551276540756225,
      "epoch": 2.709010712035287,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004265664182014985,
      "loss": 5.1568,
      "mean_token_accuracy": 0.19691064655780793,
      "num_tokens": 59471636.0,
      "step": 32245
    },
    {
      "entropy": 5.4169495582580565,
      "epoch": 2.7094307918504517,
      "grad_norm": 1.0,
      "learning_rate": 0.00042654428426005755,
      "loss": 5.0524,
      "mean_token_accuracy": 0.20149677991867065,
      "num_tokens": 59480871.0,
      "step": 32250
    },
    {
      "entropy": 5.486738443374634,
      "epoch": 2.7098508716656164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004265221476340877,
      "loss": 5.1366,
      "mean_token_accuracy": 0.19104245752096177,
      "num_tokens": 59489483.0,
      "step": 32255
    },
    {
      "entropy": 5.49877610206604,
      "epoch": 2.7102709514807812,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000426500008323981,
      "loss": 5.018,
      "mean_token_accuracy": 0.18826273530721666,
      "num_tokens": 59498369.0,
      "step": 32260
    },
    {
      "entropy": 5.568576097488403,
      "epoch": 2.710691031295946,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004264778663301296,
      "loss": 5.1672,
      "mean_token_accuracy": 0.1973414435982704,
      "num_tokens": 59507611.0,
      "step": 32265
    },
    {
      "entropy": 5.475124549865723,
      "epoch": 2.7111111111111112,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042645572165292596,
      "loss": 5.0827,
      "mean_token_accuracy": 0.19035061448812485,
      "num_tokens": 59517842.0,
      "step": 32270
    },
    {
      "entropy": 5.495505523681641,
      "epoch": 2.711531190926276,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004264335742927621,
      "loss": 5.1201,
      "mean_token_accuracy": 0.19027052521705629,
      "num_tokens": 59527081.0,
      "step": 32275
    },
    {
      "entropy": 5.477044677734375,
      "epoch": 2.711951270741441,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004264114242500304,
      "loss": 5.2179,
      "mean_token_accuracy": 0.18826461732387542,
      "num_tokens": 59535862.0,
      "step": 32280
    },
    {
      "entropy": 5.510193157196045,
      "epoch": 2.7123713505566056,
      "grad_norm": 1.125,
      "learning_rate": 0.00042638927152512335,
      "loss": 5.0662,
      "mean_token_accuracy": 0.19466468691825867,
      "num_tokens": 59544417.0,
      "step": 32285
    },
    {
      "entropy": 5.510507869720459,
      "epoch": 2.7127914303717704,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004263671161184333,
      "loss": 5.0972,
      "mean_token_accuracy": 0.19552571475505828,
      "num_tokens": 59554089.0,
      "step": 32290
    },
    {
      "entropy": 5.582606172561645,
      "epoch": 2.7132115101869356,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042634495803035274,
      "loss": 5.1672,
      "mean_token_accuracy": 0.19629413336515428,
      "num_tokens": 59563049.0,
      "step": 32295
    },
    {
      "entropy": 5.408268213272095,
      "epoch": 2.7136315900021004,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042632279726127417,
      "loss": 4.9864,
      "mean_token_accuracy": 0.2022179901599884,
      "num_tokens": 59571677.0,
      "step": 32300
    },
    {
      "entropy": 5.4951869487762455,
      "epoch": 2.714051669817265,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004263006338115902,
      "loss": 5.17,
      "mean_token_accuracy": 0.18492477089166642,
      "num_tokens": 59580859.0,
      "step": 32305
    },
    {
      "entropy": 5.444460105895996,
      "epoch": 2.7144717496324304,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004262784676816933,
      "loss": 5.0584,
      "mean_token_accuracy": 0.1947973534464836,
      "num_tokens": 59590495.0,
      "step": 32310
    },
    {
      "entropy": 5.539727449417114,
      "epoch": 2.7148918294475948,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004262562988719763,
      "loss": 5.1871,
      "mean_token_accuracy": 0.18202164322137832,
      "num_tokens": 59599647.0,
      "step": 32315
    },
    {
      "entropy": 5.610881805419922,
      "epoch": 2.71531190926276,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042623412738283193,
      "loss": 5.285,
      "mean_token_accuracy": 0.18135855346918106,
      "num_tokens": 59609762.0,
      "step": 32320
    },
    {
      "entropy": 5.528363227844238,
      "epoch": 2.715731989077925,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004262119532146528,
      "loss": 5.221,
      "mean_token_accuracy": 0.18956512063741685,
      "num_tokens": 59618849.0,
      "step": 32325
    },
    {
      "entropy": 5.510046672821045,
      "epoch": 2.7161520688930896,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004261897763678318,
      "loss": 5.1429,
      "mean_token_accuracy": 0.18607606887817382,
      "num_tokens": 59628205.0,
      "step": 32330
    },
    {
      "entropy": 5.439441299438476,
      "epoch": 2.716572148708255,
      "grad_norm": 1.125,
      "learning_rate": 0.00042616759684276176,
      "loss": 5.1213,
      "mean_token_accuracy": 0.19467079043388366,
      "num_tokens": 59637418.0,
      "step": 32335
    },
    {
      "entropy": 5.511707401275634,
      "epoch": 2.7169922285234196,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042614541463983555,
      "loss": 5.0633,
      "mean_token_accuracy": 0.19550118148326873,
      "num_tokens": 59646622.0,
      "step": 32340
    },
    {
      "entropy": 5.51901683807373,
      "epoch": 2.7174123083385844,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042612322975944615,
      "loss": 5.2073,
      "mean_token_accuracy": 0.19237185418605804,
      "num_tokens": 59655802.0,
      "step": 32345
    },
    {
      "entropy": 5.502399110794068,
      "epoch": 2.717832388153749,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004261010422019866,
      "loss": 5.0997,
      "mean_token_accuracy": 0.19707415550947188,
      "num_tokens": 59664479.0,
      "step": 32350
    },
    {
      "entropy": 5.479755926132202,
      "epoch": 2.718252467968914,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042607885196784977,
      "loss": 5.1389,
      "mean_token_accuracy": 0.18639302849769593,
      "num_tokens": 59673794.0,
      "step": 32355
    },
    {
      "entropy": 5.497097635269165,
      "epoch": 2.718672547784079,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042605665905742903,
      "loss": 5.1183,
      "mean_token_accuracy": 0.190339158475399,
      "num_tokens": 59682865.0,
      "step": 32360
    },
    {
      "entropy": 5.546522855758667,
      "epoch": 2.719092627599244,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042603446347111727,
      "loss": 5.1519,
      "mean_token_accuracy": 0.1885865345597267,
      "num_tokens": 59692768.0,
      "step": 32365
    },
    {
      "entropy": 5.5862006664276125,
      "epoch": 2.7195127074144088,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004260122652093077,
      "loss": 5.2064,
      "mean_token_accuracy": 0.18018047213554383,
      "num_tokens": 59702504.0,
      "step": 32370
    },
    {
      "entropy": 5.506681156158447,
      "epoch": 2.7199327872295735,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004259900642723938,
      "loss": 5.1026,
      "mean_token_accuracy": 0.19638999551534653,
      "num_tokens": 59712162.0,
      "step": 32375
    },
    {
      "entropy": 5.422830104827881,
      "epoch": 2.7203528670447383,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042596786066076853,
      "loss": 5.0273,
      "mean_token_accuracy": 0.1953793004155159,
      "num_tokens": 59720984.0,
      "step": 32380
    },
    {
      "entropy": 5.495920705795288,
      "epoch": 2.7207729468599036,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042594565437482543,
      "loss": 5.1089,
      "mean_token_accuracy": 0.19170984625816345,
      "num_tokens": 59730338.0,
      "step": 32385
    },
    {
      "entropy": 5.532445192337036,
      "epoch": 2.7211930266750683,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042592344541495766,
      "loss": 5.1986,
      "mean_token_accuracy": 0.18853844255208968,
      "num_tokens": 59739504.0,
      "step": 32390
    },
    {
      "entropy": 5.486055374145508,
      "epoch": 2.721613106490233,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004259012337815588,
      "loss": 5.0311,
      "mean_token_accuracy": 0.19679437428712845,
      "num_tokens": 59747975.0,
      "step": 32395
    },
    {
      "entropy": 5.558631229400635,
      "epoch": 2.722033186305398,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042587901947502234,
      "loss": 5.2043,
      "mean_token_accuracy": 0.18962817937135695,
      "num_tokens": 59756351.0,
      "step": 32400
    },
    {
      "entropy": 5.45105676651001,
      "epoch": 2.7224532661205627,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042585680249574174,
      "loss": 5.019,
      "mean_token_accuracy": 0.1916288197040558,
      "num_tokens": 59764528.0,
      "step": 32405
    },
    {
      "entropy": 5.498947620391846,
      "epoch": 2.722873345935728,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042583458284411053,
      "loss": 5.117,
      "mean_token_accuracy": 0.18973957896232604,
      "num_tokens": 59773348.0,
      "step": 32410
    },
    {
      "entropy": 5.4592186450958256,
      "epoch": 2.7232934257508927,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042581236052052237,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19425807297229766,
      "num_tokens": 59782817.0,
      "step": 32415
    },
    {
      "entropy": 5.478377008438111,
      "epoch": 2.7237135055660575,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042579013552537093,
      "loss": 5.1557,
      "mean_token_accuracy": 0.17947444915771485,
      "num_tokens": 59792798.0,
      "step": 32420
    },
    {
      "entropy": 5.520164108276367,
      "epoch": 2.7241335853812223,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042576790785904984,
      "loss": 5.1971,
      "mean_token_accuracy": 0.19097830057144166,
      "num_tokens": 59802152.0,
      "step": 32425
    },
    {
      "entropy": 5.503011274337768,
      "epoch": 2.724553665196387,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004257456775219529,
      "loss": 5.0465,
      "mean_token_accuracy": 0.19678816795349122,
      "num_tokens": 59811415.0,
      "step": 32430
    },
    {
      "entropy": 5.482586097717285,
      "epoch": 2.7249737450115523,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042572344451447387,
      "loss": 5.0959,
      "mean_token_accuracy": 0.1878418132662773,
      "num_tokens": 59820246.0,
      "step": 32435
    },
    {
      "entropy": 5.48310694694519,
      "epoch": 2.725393824826717,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042570120883700667,
      "loss": 5.1391,
      "mean_token_accuracy": 0.19351733177900315,
      "num_tokens": 59829258.0,
      "step": 32440
    },
    {
      "entropy": 5.544810581207275,
      "epoch": 2.725813904641882,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042567897048994527,
      "loss": 5.1448,
      "mean_token_accuracy": 0.19001207053661345,
      "num_tokens": 59838442.0,
      "step": 32445
    },
    {
      "entropy": 5.56279821395874,
      "epoch": 2.7262339844570467,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042565672947368334,
      "loss": 5.1614,
      "mean_token_accuracy": 0.190788896381855,
      "num_tokens": 59847696.0,
      "step": 32450
    },
    {
      "entropy": 5.461150979995727,
      "epoch": 2.7266540642722115,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004256344857886151,
      "loss": 5.1091,
      "mean_token_accuracy": 0.19289017468690872,
      "num_tokens": 59856074.0,
      "step": 32455
    },
    {
      "entropy": 5.547749853134155,
      "epoch": 2.7270741440873767,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004256122394351344,
      "loss": 5.2983,
      "mean_token_accuracy": 0.1797018364071846,
      "num_tokens": 59866113.0,
      "step": 32460
    },
    {
      "entropy": 5.615870428085327,
      "epoch": 2.7274942239025415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004255899904136356,
      "loss": 5.1649,
      "mean_token_accuracy": 0.19137729108333587,
      "num_tokens": 59875175.0,
      "step": 32465
    },
    {
      "entropy": 5.585115242004394,
      "epoch": 2.7279143037177063,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042556773872451256,
      "loss": 5.1812,
      "mean_token_accuracy": 0.18440643548965455,
      "num_tokens": 59884690.0,
      "step": 32470
    },
    {
      "entropy": 5.483697080612183,
      "epoch": 2.7283343835328715,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042554548436815956,
      "loss": 5.1539,
      "mean_token_accuracy": 0.1980432763695717,
      "num_tokens": 59893711.0,
      "step": 32475
    },
    {
      "entropy": 5.452315092086792,
      "epoch": 2.7287544633480363,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042552322734497084,
      "loss": 5.0984,
      "mean_token_accuracy": 0.19209940433502198,
      "num_tokens": 59902552.0,
      "step": 32480
    },
    {
      "entropy": 5.524556064605713,
      "epoch": 2.729174543163201,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004255009676553406,
      "loss": 5.1164,
      "mean_token_accuracy": 0.19067761898040772,
      "num_tokens": 59912832.0,
      "step": 32485
    },
    {
      "entropy": 5.552696704864502,
      "epoch": 2.729594622978366,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00042547870529966326,
      "loss": 5.1317,
      "mean_token_accuracy": 0.19287499487400056,
      "num_tokens": 59922407.0,
      "step": 32490
    },
    {
      "entropy": 5.508346128463745,
      "epoch": 2.7300147027935306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004254564402783332,
      "loss": 5.1055,
      "mean_token_accuracy": 0.19809230118989946,
      "num_tokens": 59930619.0,
      "step": 32495
    },
    {
      "entropy": 5.3644750118255615,
      "epoch": 2.730434782608696,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004254341725917447,
      "loss": 5.0433,
      "mean_token_accuracy": 0.19423172622919083,
      "num_tokens": 59939386.0,
      "step": 32500
    },
    {
      "entropy": 5.476773500442505,
      "epoch": 2.7308548624238607,
      "grad_norm": 1.125,
      "learning_rate": 0.00042541190224029235,
      "loss": 5.1805,
      "mean_token_accuracy": 0.1902864933013916,
      "num_tokens": 59949174.0,
      "step": 32505
    },
    {
      "entropy": 5.558675241470337,
      "epoch": 2.7312749422390254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004253896292243705,
      "loss": 5.1518,
      "mean_token_accuracy": 0.19449598342180252,
      "num_tokens": 59957658.0,
      "step": 32510
    },
    {
      "entropy": 5.455645370483398,
      "epoch": 2.7316950220541902,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004253673535443739,
      "loss": 5.0615,
      "mean_token_accuracy": 0.19893271178007127,
      "num_tokens": 59966441.0,
      "step": 32515
    },
    {
      "entropy": 5.450817251205445,
      "epoch": 2.732115101869355,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042534507520069706,
      "loss": 5.0361,
      "mean_token_accuracy": 0.19376906156539916,
      "num_tokens": 59974917.0,
      "step": 32520
    },
    {
      "entropy": 5.447136068344117,
      "epoch": 2.7325351816845203,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004253227941937346,
      "loss": 5.0932,
      "mean_token_accuracy": 0.19378408342599868,
      "num_tokens": 59983931.0,
      "step": 32525
    },
    {
      "entropy": 5.462182283401489,
      "epoch": 2.732955261499685,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042530051052388124,
      "loss": 5.0379,
      "mean_token_accuracy": 0.1935834690928459,
      "num_tokens": 59992210.0,
      "step": 32530
    },
    {
      "entropy": 5.52638840675354,
      "epoch": 2.73337534131485,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042527822419153173,
      "loss": 5.2068,
      "mean_token_accuracy": 0.18592821359634398,
      "num_tokens": 60001775.0,
      "step": 32535
    },
    {
      "entropy": 5.624679183959961,
      "epoch": 2.7337954211300146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042525593519708094,
      "loss": 5.3093,
      "mean_token_accuracy": 0.18261929005384445,
      "num_tokens": 60012385.0,
      "step": 32540
    },
    {
      "entropy": 5.539332580566406,
      "epoch": 2.7342155009451794,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004252336435409235,
      "loss": 5.1095,
      "mean_token_accuracy": 0.18875256776809693,
      "num_tokens": 60021745.0,
      "step": 32545
    },
    {
      "entropy": 5.497437191009522,
      "epoch": 2.7346355807603446,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004252113492234545,
      "loss": 5.109,
      "mean_token_accuracy": 0.1960751846432686,
      "num_tokens": 60031359.0,
      "step": 32550
    },
    {
      "entropy": 5.52957592010498,
      "epoch": 2.7350556605755094,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042518905224506876,
      "loss": 5.1737,
      "mean_token_accuracy": 0.18350592255592346,
      "num_tokens": 60041293.0,
      "step": 32555
    },
    {
      "entropy": 5.5757698059082035,
      "epoch": 2.735475740390674,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042516675260616135,
      "loss": 5.2892,
      "mean_token_accuracy": 0.18292323052883147,
      "num_tokens": 60050766.0,
      "step": 32560
    },
    {
      "entropy": 5.59047122001648,
      "epoch": 2.735895820205839,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004251444503071272,
      "loss": 5.1807,
      "mean_token_accuracy": 0.18681995421648026,
      "num_tokens": 60060034.0,
      "step": 32565
    },
    {
      "entropy": 5.590848207473755,
      "epoch": 2.7363159000210038,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004251221453483614,
      "loss": 5.1878,
      "mean_token_accuracy": 0.1914323404431343,
      "num_tokens": 60068821.0,
      "step": 32570
    },
    {
      "entropy": 5.5001232624053955,
      "epoch": 2.736735979836169,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004250998377302591,
      "loss": 5.1007,
      "mean_token_accuracy": 0.18685579746961595,
      "num_tokens": 60078220.0,
      "step": 32575
    },
    {
      "entropy": 5.423198413848877,
      "epoch": 2.737156059651334,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042507752745321554,
      "loss": 5.0862,
      "mean_token_accuracy": 0.1937178909778595,
      "num_tokens": 60086620.0,
      "step": 32580
    },
    {
      "entropy": 5.519389343261719,
      "epoch": 2.7375761394664986,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004250552145176258,
      "loss": 5.1253,
      "mean_token_accuracy": 0.19495875239372254,
      "num_tokens": 60096208.0,
      "step": 32585
    },
    {
      "entropy": 5.483314275741577,
      "epoch": 2.7379962192816634,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004250328989238852,
      "loss": 5.0872,
      "mean_token_accuracy": 0.19431654512882232,
      "num_tokens": 60105640.0,
      "step": 32590
    },
    {
      "entropy": 5.521010160446167,
      "epoch": 2.738416299096828,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004250105806723891,
      "loss": 5.152,
      "mean_token_accuracy": 0.18532235771417618,
      "num_tokens": 60114824.0,
      "step": 32595
    },
    {
      "entropy": 5.607889986038208,
      "epoch": 2.7388363789119934,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004249882597635328,
      "loss": 5.2098,
      "mean_token_accuracy": 0.1896708145737648,
      "num_tokens": 60124453.0,
      "step": 32600
    },
    {
      "entropy": 5.583510017395019,
      "epoch": 2.739256458727158,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004249659361977116,
      "loss": 5.2207,
      "mean_token_accuracy": 0.18541543036699296,
      "num_tokens": 60134515.0,
      "step": 32605
    },
    {
      "entropy": 5.489931631088257,
      "epoch": 2.739676538542323,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042494360997532123,
      "loss": 5.1118,
      "mean_token_accuracy": 0.19110747724771499,
      "num_tokens": 60143738.0,
      "step": 32610
    },
    {
      "entropy": 5.530751323699951,
      "epoch": 2.740096618357488,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004249212810967569,
      "loss": 5.1563,
      "mean_token_accuracy": 0.1921055570244789,
      "num_tokens": 60154378.0,
      "step": 32615
    },
    {
      "entropy": 5.564778184890747,
      "epoch": 2.7405166981726525,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004248989495624143,
      "loss": 5.2123,
      "mean_token_accuracy": 0.18183356076478957,
      "num_tokens": 60165174.0,
      "step": 32620
    },
    {
      "entropy": 5.553151988983155,
      "epoch": 2.7409367779878178,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042487661537268903,
      "loss": 5.1023,
      "mean_token_accuracy": 0.19720169454813002,
      "num_tokens": 60174954.0,
      "step": 32625
    },
    {
      "entropy": 5.595531463623047,
      "epoch": 2.7413568578029825,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004248542785279766,
      "loss": 5.2808,
      "mean_token_accuracy": 0.17891852408647538,
      "num_tokens": 60183879.0,
      "step": 32630
    },
    {
      "entropy": 5.564065170288086,
      "epoch": 2.7417769376181473,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004248319390286729,
      "loss": 5.1605,
      "mean_token_accuracy": 0.1929906651377678,
      "num_tokens": 60192824.0,
      "step": 32635
    },
    {
      "entropy": 5.556074428558349,
      "epoch": 2.7421970174333126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042480959687517353,
      "loss": 5.185,
      "mean_token_accuracy": 0.19175364971160888,
      "num_tokens": 60201337.0,
      "step": 32640
    },
    {
      "entropy": 5.550010299682617,
      "epoch": 2.7426170972484774,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042478725206787426,
      "loss": 5.2055,
      "mean_token_accuracy": 0.19180342704057693,
      "num_tokens": 60210622.0,
      "step": 32645
    },
    {
      "entropy": 5.4889068603515625,
      "epoch": 2.743037177063642,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042476490460717086,
      "loss": 5.1014,
      "mean_token_accuracy": 0.19386740922927856,
      "num_tokens": 60220121.0,
      "step": 32650
    },
    {
      "entropy": 5.452874994277954,
      "epoch": 2.743457256878807,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042474255449345943,
      "loss": 5.0543,
      "mean_token_accuracy": 0.19057636708021164,
      "num_tokens": 60228305.0,
      "step": 32655
    },
    {
      "entropy": 5.5232831001281735,
      "epoch": 2.7438773366939717,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004247202017271357,
      "loss": 5.2362,
      "mean_token_accuracy": 0.18313243389129638,
      "num_tokens": 60237560.0,
      "step": 32660
    },
    {
      "entropy": 5.499138641357422,
      "epoch": 2.744297416509137,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042469784630859577,
      "loss": 5.07,
      "mean_token_accuracy": 0.19837523698806764,
      "num_tokens": 60247254.0,
      "step": 32665
    },
    {
      "entropy": 5.544473886489868,
      "epoch": 2.7447174963243017,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004246754882382355,
      "loss": 5.1926,
      "mean_token_accuracy": 0.1792330875992775,
      "num_tokens": 60256286.0,
      "step": 32670
    },
    {
      "entropy": 5.555413913726807,
      "epoch": 2.7451375761394665,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000424653127516451,
      "loss": 5.2002,
      "mean_token_accuracy": 0.18963598757982253,
      "num_tokens": 60265984.0,
      "step": 32675
    },
    {
      "entropy": 5.505923843383789,
      "epoch": 2.7455576559546313,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042463076414363845,
      "loss": 5.1463,
      "mean_token_accuracy": 0.19182963222265242,
      "num_tokens": 60274929.0,
      "step": 32680
    },
    {
      "entropy": 5.430939102172852,
      "epoch": 2.745977735769796,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042460839812019397,
      "loss": 5.0923,
      "mean_token_accuracy": 0.20333992689847946,
      "num_tokens": 60283978.0,
      "step": 32685
    },
    {
      "entropy": 5.483720541000366,
      "epoch": 2.7463978155849613,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042458602944651377,
      "loss": 5.1477,
      "mean_token_accuracy": 0.19412756562232972,
      "num_tokens": 60292831.0,
      "step": 32690
    },
    {
      "entropy": 5.436821222305298,
      "epoch": 2.746817895400126,
      "grad_norm": 1.015625,
      "learning_rate": 0.000424563658122994,
      "loss": 4.9681,
      "mean_token_accuracy": 0.1940985068678856,
      "num_tokens": 60302129.0,
      "step": 32695
    },
    {
      "entropy": 5.469022130966186,
      "epoch": 2.747237975215291,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004245412841500311,
      "loss": 5.0344,
      "mean_token_accuracy": 0.19903966039419174,
      "num_tokens": 60310327.0,
      "step": 32700
    },
    {
      "entropy": 5.51338791847229,
      "epoch": 2.7476580550304557,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042451890752802135,
      "loss": 5.1179,
      "mean_token_accuracy": 0.1950961783528328,
      "num_tokens": 60320174.0,
      "step": 32705
    },
    {
      "entropy": 5.508333969116211,
      "epoch": 2.7480781348456205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042449652825736115,
      "loss": 5.2391,
      "mean_token_accuracy": 0.17869880199432372,
      "num_tokens": 60329901.0,
      "step": 32710
    },
    {
      "entropy": 5.610521030426026,
      "epoch": 2.7484982146607857,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004244741463384469,
      "loss": 5.3142,
      "mean_token_accuracy": 0.17716512382030486,
      "num_tokens": 60339219.0,
      "step": 32715
    },
    {
      "entropy": 5.560701513290406,
      "epoch": 2.7489182944759505,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004244517617716751,
      "loss": 5.1397,
      "mean_token_accuracy": 0.18511488437652587,
      "num_tokens": 60348745.0,
      "step": 32720
    },
    {
      "entropy": 5.537802076339721,
      "epoch": 2.7493383742911153,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042442937455744234,
      "loss": 5.1766,
      "mean_token_accuracy": 0.18373243063688277,
      "num_tokens": 60358362.0,
      "step": 32725
    },
    {
      "entropy": 5.543797302246094,
      "epoch": 2.74975845410628,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042440698469614515,
      "loss": 5.1648,
      "mean_token_accuracy": 0.18776392042636872,
      "num_tokens": 60368169.0,
      "step": 32730
    },
    {
      "entropy": 5.621547651290894,
      "epoch": 2.750178533921445,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042438459218818015,
      "loss": 5.2182,
      "mean_token_accuracy": 0.17984139323234558,
      "num_tokens": 60377710.0,
      "step": 32735
    },
    {
      "entropy": 5.42676682472229,
      "epoch": 2.75059861373661,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004243621970339441,
      "loss": 5.0423,
      "mean_token_accuracy": 0.19275061786174774,
      "num_tokens": 60387138.0,
      "step": 32740
    },
    {
      "entropy": 5.568715143203735,
      "epoch": 2.751018693551775,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004243397992338335,
      "loss": 5.2315,
      "mean_token_accuracy": 0.186509969830513,
      "num_tokens": 60397271.0,
      "step": 32745
    },
    {
      "entropy": 5.599147605895996,
      "epoch": 2.7514387733669396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042431739878824533,
      "loss": 5.1944,
      "mean_token_accuracy": 0.18236073181033136,
      "num_tokens": 60407058.0,
      "step": 32750
    },
    {
      "entropy": 5.576691436767578,
      "epoch": 2.7518588531821044,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042429499569757634,
      "loss": 5.1749,
      "mean_token_accuracy": 0.19010416120290757,
      "num_tokens": 60416455.0,
      "step": 32755
    },
    {
      "entropy": 5.465377283096314,
      "epoch": 2.7522789329972692,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004242725899622234,
      "loss": 5.1394,
      "mean_token_accuracy": 0.18889058232307435,
      "num_tokens": 60425943.0,
      "step": 32760
    },
    {
      "entropy": 5.53840126991272,
      "epoch": 2.7526990128124345,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004242501815825833,
      "loss": 5.2663,
      "mean_token_accuracy": 0.17914967685937883,
      "num_tokens": 60435649.0,
      "step": 32765
    },
    {
      "entropy": 5.5550134658813475,
      "epoch": 2.7531190926275992,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004242277705590531,
      "loss": 5.0578,
      "mean_token_accuracy": 0.1956307992339134,
      "num_tokens": 60445147.0,
      "step": 32770
    },
    {
      "entropy": 5.499627876281738,
      "epoch": 2.753539172442764,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042420535689202975,
      "loss": 5.1537,
      "mean_token_accuracy": 0.19450152814388275,
      "num_tokens": 60455345.0,
      "step": 32775
    },
    {
      "entropy": 5.458238649368286,
      "epoch": 2.7539592522579293,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004241829405819105,
      "loss": 5.1241,
      "mean_token_accuracy": 0.19407746195793152,
      "num_tokens": 60464247.0,
      "step": 32780
    },
    {
      "entropy": 5.574519109725952,
      "epoch": 2.754379332073094,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004241605216290921,
      "loss": 5.2198,
      "mean_token_accuracy": 0.18852973133325576,
      "num_tokens": 60472703.0,
      "step": 32785
    },
    {
      "entropy": 5.5077423572540285,
      "epoch": 2.754799411888259,
      "grad_norm": 1.125,
      "learning_rate": 0.000424138100033972,
      "loss": 5.0734,
      "mean_token_accuracy": 0.19240954667329788,
      "num_tokens": 60481895.0,
      "step": 32790
    },
    {
      "entropy": 5.512294292449951,
      "epoch": 2.7552194917034236,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042411567579694713,
      "loss": 5.2557,
      "mean_token_accuracy": 0.18019042760133744,
      "num_tokens": 60491477.0,
      "step": 32795
    },
    {
      "entropy": 5.573426055908203,
      "epoch": 2.7556395715185884,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000424093248918415,
      "loss": 5.2118,
      "mean_token_accuracy": 0.1868919849395752,
      "num_tokens": 60500683.0,
      "step": 32800
    },
    {
      "entropy": 5.578431463241577,
      "epoch": 2.7560596513337536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004240708193987726,
      "loss": 5.1746,
      "mean_token_accuracy": 0.18644829094409943,
      "num_tokens": 60510302.0,
      "step": 32805
    },
    {
      "entropy": 5.4230907440185545,
      "epoch": 2.7564797311489184,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004240483872384175,
      "loss": 5.0438,
      "mean_token_accuracy": 0.19079239070415496,
      "num_tokens": 60518720.0,
      "step": 32810
    },
    {
      "entropy": 5.558521270751953,
      "epoch": 2.756899810964083,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042402595243774686,
      "loss": 5.201,
      "mean_token_accuracy": 0.19057183414697648,
      "num_tokens": 60528800.0,
      "step": 32815
    },
    {
      "entropy": 5.481593418121338,
      "epoch": 2.757319890779248,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042400351499715825,
      "loss": 5.0241,
      "mean_token_accuracy": 0.19625092744827272,
      "num_tokens": 60537962.0,
      "step": 32820
    },
    {
      "entropy": 5.434565830230713,
      "epoch": 2.757739970594413,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042398107491704916,
      "loss": 5.1415,
      "mean_token_accuracy": 0.19663738012313842,
      "num_tokens": 60547020.0,
      "step": 32825
    },
    {
      "entropy": 5.600077104568482,
      "epoch": 2.758160050409578,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042395863219781693,
      "loss": 5.1469,
      "mean_token_accuracy": 0.19843304455280303,
      "num_tokens": 60555768.0,
      "step": 32830
    },
    {
      "entropy": 5.498507261276245,
      "epoch": 2.758580130224743,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042393618683985937,
      "loss": 5.0822,
      "mean_token_accuracy": 0.19673628956079484,
      "num_tokens": 60564712.0,
      "step": 32835
    },
    {
      "entropy": 5.454414701461792,
      "epoch": 2.7590002100399076,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004239137388435739,
      "loss": 5.0521,
      "mean_token_accuracy": 0.18754678666591645,
      "num_tokens": 60573315.0,
      "step": 32840
    },
    {
      "entropy": 5.590971660614014,
      "epoch": 2.7594202898550724,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004238912882093581,
      "loss": 5.2832,
      "mean_token_accuracy": 0.1799099013209343,
      "num_tokens": 60582387.0,
      "step": 32845
    },
    {
      "entropy": 5.485367155075073,
      "epoch": 2.759840369670237,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004238688349376099,
      "loss": 5.0994,
      "mean_token_accuracy": 0.1918117105960846,
      "num_tokens": 60592304.0,
      "step": 32850
    },
    {
      "entropy": 5.496688032150269,
      "epoch": 2.7602604494854024,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042384637902872694,
      "loss": 5.1368,
      "mean_token_accuracy": 0.1849253162741661,
      "num_tokens": 60602031.0,
      "step": 32855
    },
    {
      "entropy": 5.5591834545135494,
      "epoch": 2.760680529300567,
      "grad_norm": 1.109375,
      "learning_rate": 0.000423823920483107,
      "loss": 5.1217,
      "mean_token_accuracy": 0.19156631082296371,
      "num_tokens": 60611313.0,
      "step": 32860
    },
    {
      "entropy": 5.563255739212036,
      "epoch": 2.761100609115732,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004238014593011481,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19538293033838272,
      "num_tokens": 60620217.0,
      "step": 32865
    },
    {
      "entropy": 5.407671928405762,
      "epoch": 2.7615206889308967,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042377899548324774,
      "loss": 5.1028,
      "mean_token_accuracy": 0.18768367618322374,
      "num_tokens": 60629257.0,
      "step": 32870
    },
    {
      "entropy": 5.580098485946655,
      "epoch": 2.7619407687460615,
      "grad_norm": 1.125,
      "learning_rate": 0.00042375652902980414,
      "loss": 5.2427,
      "mean_token_accuracy": 0.181588077545166,
      "num_tokens": 60640377.0,
      "step": 32875
    },
    {
      "entropy": 5.611979532241821,
      "epoch": 2.7623608485612268,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004237340599412153,
      "loss": 5.2192,
      "mean_token_accuracy": 0.18363846242427825,
      "num_tokens": 60649794.0,
      "step": 32880
    },
    {
      "entropy": 5.543249082565308,
      "epoch": 2.7627809283763916,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042371158821787915,
      "loss": 5.202,
      "mean_token_accuracy": 0.18444195985794068,
      "num_tokens": 60659476.0,
      "step": 32885
    },
    {
      "entropy": 5.539347791671753,
      "epoch": 2.7632010081915563,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004236891138601938,
      "loss": 5.1686,
      "mean_token_accuracy": 0.18983530551195144,
      "num_tokens": 60668386.0,
      "step": 32890
    },
    {
      "entropy": 5.453588056564331,
      "epoch": 2.763621088006721,
      "grad_norm": 1.125,
      "learning_rate": 0.00042366663686855735,
      "loss": 5.115,
      "mean_token_accuracy": 0.19873028695583345,
      "num_tokens": 60677333.0,
      "step": 32895
    },
    {
      "entropy": 5.470369815826416,
      "epoch": 2.764041167821886,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004236441572433679,
      "loss": 5.1496,
      "mean_token_accuracy": 0.1926627427339554,
      "num_tokens": 60686627.0,
      "step": 32900
    },
    {
      "entropy": 5.443140983581543,
      "epoch": 2.764461247637051,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004236216749850238,
      "loss": 5.0221,
      "mean_token_accuracy": 0.19668327420949935,
      "num_tokens": 60695510.0,
      "step": 32905
    },
    {
      "entropy": 5.49115023612976,
      "epoch": 2.764881327452216,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004235991900939233,
      "loss": 5.15,
      "mean_token_accuracy": 0.18799378126859664,
      "num_tokens": 60704852.0,
      "step": 32910
    },
    {
      "entropy": 5.51422700881958,
      "epoch": 2.7653014072673807,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042357670257046464,
      "loss": 5.2319,
      "mean_token_accuracy": 0.19363542944192885,
      "num_tokens": 60714649.0,
      "step": 32915
    },
    {
      "entropy": 5.50185112953186,
      "epoch": 2.765721487082546,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042355421241504614,
      "loss": 5.1252,
      "mean_token_accuracy": 0.18387941569089888,
      "num_tokens": 60723960.0,
      "step": 32920
    },
    {
      "entropy": 5.470899963378907,
      "epoch": 2.7661415668977103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042353171962806633,
      "loss": 5.1123,
      "mean_token_accuracy": 0.19177275598049165,
      "num_tokens": 60733035.0,
      "step": 32925
    },
    {
      "entropy": 5.592600011825562,
      "epoch": 2.7665616467128755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004235092242099236,
      "loss": 5.1669,
      "mean_token_accuracy": 0.1919972136616707,
      "num_tokens": 60742183.0,
      "step": 32930
    },
    {
      "entropy": 5.4849992275238035,
      "epoch": 2.7669817265280403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004234867261610164,
      "loss": 5.1315,
      "mean_token_accuracy": 0.19238951057195663,
      "num_tokens": 60752163.0,
      "step": 32935
    },
    {
      "entropy": 5.3852368831634525,
      "epoch": 2.767401806343205,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004234642254817433,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20612553507089615,
      "num_tokens": 60760921.0,
      "step": 32940
    },
    {
      "entropy": 5.397820711135864,
      "epoch": 2.7678218861583703,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042344172217250294,
      "loss": 5.0182,
      "mean_token_accuracy": 0.20407064706087114,
      "num_tokens": 60769245.0,
      "step": 32945
    },
    {
      "entropy": 5.548063325881958,
      "epoch": 2.768241965973535,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042341921623369384,
      "loss": 5.1746,
      "mean_token_accuracy": 0.1928075671195984,
      "num_tokens": 60780246.0,
      "step": 32950
    },
    {
      "entropy": 5.55889401435852,
      "epoch": 2.7686620457887,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004233967076657148,
      "loss": 5.259,
      "mean_token_accuracy": 0.189474418759346,
      "num_tokens": 60790180.0,
      "step": 32955
    },
    {
      "entropy": 5.517692613601684,
      "epoch": 2.7690821256038647,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042337419646896454,
      "loss": 5.108,
      "mean_token_accuracy": 0.19383364617824556,
      "num_tokens": 60799097.0,
      "step": 32960
    },
    {
      "entropy": 5.601240730285644,
      "epoch": 2.7695022054190295,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004233516826438417,
      "loss": 5.2201,
      "mean_token_accuracy": 0.1891783818602562,
      "num_tokens": 60808180.0,
      "step": 32965
    },
    {
      "entropy": 5.54994387626648,
      "epoch": 2.7699222852341947,
      "grad_norm": 1.125,
      "learning_rate": 0.0004233291661907452,
      "loss": 5.2088,
      "mean_token_accuracy": 0.18697039484977723,
      "num_tokens": 60817564.0,
      "step": 32970
    },
    {
      "entropy": 5.557264041900635,
      "epoch": 2.7703423650493595,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042330664711007394,
      "loss": 5.1742,
      "mean_token_accuracy": 0.19287987053394318,
      "num_tokens": 60826461.0,
      "step": 32975
    },
    {
      "entropy": 5.468770837783813,
      "epoch": 2.7707624448645243,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004232841254022267,
      "loss": 5.1135,
      "mean_token_accuracy": 0.18998779803514482,
      "num_tokens": 60836514.0,
      "step": 32980
    },
    {
      "entropy": 5.508577632904053,
      "epoch": 2.771182524679689,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004232616010676026,
      "loss": 5.1488,
      "mean_token_accuracy": 0.19604662358760833,
      "num_tokens": 60846896.0,
      "step": 32985
    },
    {
      "entropy": 5.518606233596802,
      "epoch": 2.771602604494854,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004232390741066007,
      "loss": 5.2278,
      "mean_token_accuracy": 0.18273012787103654,
      "num_tokens": 60856629.0,
      "step": 32990
    },
    {
      "entropy": 5.478359174728394,
      "epoch": 2.772022684310019,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004232165445196198,
      "loss": 5.0893,
      "mean_token_accuracy": 0.19710645377635955,
      "num_tokens": 60865783.0,
      "step": 32995
    },
    {
      "entropy": 5.499603605270385,
      "epoch": 2.772442764125184,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004231940123070591,
      "loss": 5.0836,
      "mean_token_accuracy": 0.19181227087974548,
      "num_tokens": 60875626.0,
      "step": 33000
    },
    {
      "epoch": 2.772442764125184,
      "eval_entropy": 5.273637854499139,
      "eval_loss": 5.200925827026367,
      "eval_mean_token_accuracy": 0.19730805873701773,
      "eval_num_tokens": 60875626.0,
      "eval_runtime": 27.4908,
      "eval_samples_per_second": 1359.22,
      "eval_steps_per_second": 169.912,
      "step": 33000
    },
    {
      "entropy": 5.443844270706177,
      "epoch": 2.7728628439403487,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004231714774693179,
      "loss": 5.0289,
      "mean_token_accuracy": 0.19453359097242356,
      "num_tokens": 60884227.0,
      "step": 33005
    },
    {
      "entropy": 5.436780691146851,
      "epoch": 2.7732829237555134,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042314894000679515,
      "loss": 5.08,
      "mean_token_accuracy": 0.19905056804418564,
      "num_tokens": 60892937.0,
      "step": 33010
    },
    {
      "entropy": 5.408236885070801,
      "epoch": 2.7737030035706782,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004231263999198903,
      "loss": 5.0441,
      "mean_token_accuracy": 0.19232673197984695,
      "num_tokens": 60901494.0,
      "step": 33015
    },
    {
      "entropy": 5.542442750930786,
      "epoch": 2.7741230833858435,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042310385720900254,
      "loss": 5.1414,
      "mean_token_accuracy": 0.19348904937505723,
      "num_tokens": 60910708.0,
      "step": 33020
    },
    {
      "entropy": 5.5006355285644535,
      "epoch": 2.7745431632010082,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004230813118745312,
      "loss": 5.1603,
      "mean_token_accuracy": 0.18924813717603683,
      "num_tokens": 60920042.0,
      "step": 33025
    },
    {
      "entropy": 5.452403354644775,
      "epoch": 2.774963243016173,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004230587639168757,
      "loss": 5.0832,
      "mean_token_accuracy": 0.20034782588481903,
      "num_tokens": 60930052.0,
      "step": 33030
    },
    {
      "entropy": 5.493816709518432,
      "epoch": 2.775383322831338,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004230362133364354,
      "loss": 5.1383,
      "mean_token_accuracy": 0.18391549587249756,
      "num_tokens": 60939448.0,
      "step": 33035
    },
    {
      "entropy": 5.555425357818604,
      "epoch": 2.7758034026465026,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042301366013360975,
      "loss": 5.2656,
      "mean_token_accuracy": 0.19120437130331994,
      "num_tokens": 60948523.0,
      "step": 33040
    },
    {
      "entropy": 5.609965991973877,
      "epoch": 2.776223482461668,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004229911043087984,
      "loss": 5.1533,
      "mean_token_accuracy": 0.19301903396844863,
      "num_tokens": 60957527.0,
      "step": 33045
    },
    {
      "entropy": 5.512027359008789,
      "epoch": 2.7766435622768326,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004229685458624009,
      "loss": 5.1531,
      "mean_token_accuracy": 0.18805545270442964,
      "num_tokens": 60966538.0,
      "step": 33050
    },
    {
      "entropy": 5.44506516456604,
      "epoch": 2.7770636420919974,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042294598479481664,
      "loss": 4.985,
      "mean_token_accuracy": 0.19894355088472365,
      "num_tokens": 60974934.0,
      "step": 33055
    },
    {
      "entropy": 5.456914186477661,
      "epoch": 2.777483721907162,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004229234211064456,
      "loss": 5.1717,
      "mean_token_accuracy": 0.185769023001194,
      "num_tokens": 60984151.0,
      "step": 33060
    },
    {
      "entropy": 5.462401580810547,
      "epoch": 2.777903801722327,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004229008547976872,
      "loss": 5.1514,
      "mean_token_accuracy": 0.18811849057674407,
      "num_tokens": 60993898.0,
      "step": 33065
    },
    {
      "entropy": 5.44525671005249,
      "epoch": 2.778323881537492,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004228782858689414,
      "loss": 5.0589,
      "mean_token_accuracy": 0.2058633178472519,
      "num_tokens": 61002709.0,
      "step": 33070
    },
    {
      "entropy": 5.533148193359375,
      "epoch": 2.778743961352657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004228557143206079,
      "loss": 5.1473,
      "mean_token_accuracy": 0.19124349802732468,
      "num_tokens": 61012120.0,
      "step": 33075
    },
    {
      "entropy": 5.530851364135742,
      "epoch": 2.779164041167822,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004228331401530865,
      "loss": 5.0787,
      "mean_token_accuracy": 0.1970926508307457,
      "num_tokens": 61021012.0,
      "step": 33080
    },
    {
      "entropy": 5.5337042808532715,
      "epoch": 2.779584120982987,
      "grad_norm": 1.125,
      "learning_rate": 0.00042281056336677714,
      "loss": 5.1423,
      "mean_token_accuracy": 0.1915345698595047,
      "num_tokens": 61031229.0,
      "step": 33085
    },
    {
      "entropy": 5.412232065200806,
      "epoch": 2.780004200798152,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004227879839620798,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20124713629484176,
      "num_tokens": 61040875.0,
      "step": 33090
    },
    {
      "entropy": 5.4293825149536135,
      "epoch": 2.7804242806133166,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004227654019393945,
      "loss": 5.0824,
      "mean_token_accuracy": 0.19279139935970308,
      "num_tokens": 61049966.0,
      "step": 33095
    },
    {
      "entropy": 5.544611501693725,
      "epoch": 2.7808443604284814,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004227428172991211,
      "loss": 5.1544,
      "mean_token_accuracy": 0.1818712532520294,
      "num_tokens": 61059263.0,
      "step": 33100
    },
    {
      "entropy": 5.516089630126953,
      "epoch": 2.781264440243646,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004227202300416598,
      "loss": 5.134,
      "mean_token_accuracy": 0.19252004325389863,
      "num_tokens": 61068413.0,
      "step": 33105
    },
    {
      "entropy": 5.438155746459961,
      "epoch": 2.7816845200588114,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042269764016741065,
      "loss": 5.0748,
      "mean_token_accuracy": 0.1979958087205887,
      "num_tokens": 61078085.0,
      "step": 33110
    },
    {
      "entropy": 5.45478515625,
      "epoch": 2.782104599873976,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042267504767677384,
      "loss": 5.104,
      "mean_token_accuracy": 0.1954660177230835,
      "num_tokens": 61086813.0,
      "step": 33115
    },
    {
      "entropy": 5.528433609008789,
      "epoch": 2.782524679689141,
      "grad_norm": 0.984375,
      "learning_rate": 0.00042265245257014963,
      "loss": 5.1377,
      "mean_token_accuracy": 0.1881802946329117,
      "num_tokens": 61097134.0,
      "step": 33120
    },
    {
      "entropy": 5.448878622055053,
      "epoch": 2.7829447595043058,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042262985484793825,
      "loss": 5.0882,
      "mean_token_accuracy": 0.198402801156044,
      "num_tokens": 61106282.0,
      "step": 33125
    },
    {
      "entropy": 5.584585666656494,
      "epoch": 2.7833648393194705,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042260725451054006,
      "loss": 5.2599,
      "mean_token_accuracy": 0.18191756457090377,
      "num_tokens": 61115792.0,
      "step": 33130
    },
    {
      "entropy": 5.513132524490357,
      "epoch": 2.7837849191346358,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042258465155835533,
      "loss": 5.1898,
      "mean_token_accuracy": 0.18081435859203338,
      "num_tokens": 61124524.0,
      "step": 33135
    },
    {
      "entropy": 5.5383350372314455,
      "epoch": 2.7842049989498006,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004225620459917845,
      "loss": 5.1383,
      "mean_token_accuracy": 0.19061129838228225,
      "num_tokens": 61133587.0,
      "step": 33140
    },
    {
      "entropy": 5.583036994934082,
      "epoch": 2.7846250787649653,
      "grad_norm": 1.125,
      "learning_rate": 0.000422539437811228,
      "loss": 5.2183,
      "mean_token_accuracy": 0.1901163637638092,
      "num_tokens": 61142040.0,
      "step": 33145
    },
    {
      "entropy": 5.547698926925659,
      "epoch": 2.78504515858013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042251682701708633,
      "loss": 5.1285,
      "mean_token_accuracy": 0.18883834928274154,
      "num_tokens": 61151705.0,
      "step": 33150
    },
    {
      "entropy": 5.517924833297729,
      "epoch": 2.785465238395295,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042249421360976,
      "loss": 5.0823,
      "mean_token_accuracy": 0.19590977281332017,
      "num_tokens": 61159467.0,
      "step": 33155
    },
    {
      "entropy": 5.439500713348389,
      "epoch": 2.78588531821046,
      "grad_norm": 1.25,
      "learning_rate": 0.0004224715975896497,
      "loss": 5.1283,
      "mean_token_accuracy": 0.19127618223428727,
      "num_tokens": 61169794.0,
      "step": 33160
    },
    {
      "entropy": 5.628601360321045,
      "epoch": 2.786305398025625,
      "grad_norm": 1.0625,
      "learning_rate": 0.000422448978957156,
      "loss": 5.2466,
      "mean_token_accuracy": 0.17899647057056428,
      "num_tokens": 61178998.0,
      "step": 33165
    },
    {
      "entropy": 5.565878963470459,
      "epoch": 2.7867254778407897,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042242635771267946,
      "loss": 5.0842,
      "mean_token_accuracy": 0.19834268987178802,
      "num_tokens": 61187969.0,
      "step": 33170
    },
    {
      "entropy": 5.531174612045288,
      "epoch": 2.7871455576559545,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004224037338566209,
      "loss": 5.2124,
      "mean_token_accuracy": 0.18125469982624054,
      "num_tokens": 61198302.0,
      "step": 33175
    },
    {
      "entropy": 5.582902288436889,
      "epoch": 2.7875656374711193,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004223811073893812,
      "loss": 5.2353,
      "mean_token_accuracy": 0.182644884288311,
      "num_tokens": 61208021.0,
      "step": 33180
    },
    {
      "entropy": 5.56622633934021,
      "epoch": 2.7879857172862845,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042235847831136096,
      "loss": 5.2249,
      "mean_token_accuracy": 0.18324110358953477,
      "num_tokens": 61216987.0,
      "step": 33185
    },
    {
      "entropy": 5.571833610534668,
      "epoch": 2.7884057971014493,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042233584662296127,
      "loss": 5.1165,
      "mean_token_accuracy": 0.1888119474053383,
      "num_tokens": 61226813.0,
      "step": 33190
    },
    {
      "entropy": 5.491798019409179,
      "epoch": 2.788825876916614,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004223132123245828,
      "loss": 5.1303,
      "mean_token_accuracy": 0.19074346870183945,
      "num_tokens": 61236954.0,
      "step": 33195
    },
    {
      "entropy": 5.499544429779053,
      "epoch": 2.789245956731779,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042229057541662675,
      "loss": 5.1588,
      "mean_token_accuracy": 0.18412894010543823,
      "num_tokens": 61246134.0,
      "step": 33200
    },
    {
      "entropy": 5.453973817825317,
      "epoch": 2.7896660365469437,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042226793589949396,
      "loss": 5.0437,
      "mean_token_accuracy": 0.20029927641153336,
      "num_tokens": 61255229.0,
      "step": 33205
    },
    {
      "entropy": 5.50882658958435,
      "epoch": 2.790086116362109,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042224529377358545,
      "loss": 5.1728,
      "mean_token_accuracy": 0.19133198708295823,
      "num_tokens": 61263814.0,
      "step": 33210
    },
    {
      "entropy": 5.527130556106568,
      "epoch": 2.7905061961772737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004222226490393025,
      "loss": 5.1839,
      "mean_token_accuracy": 0.18934765905141832,
      "num_tokens": 61273669.0,
      "step": 33215
    },
    {
      "entropy": 5.444549798965454,
      "epoch": 2.7909262759924385,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042220000169704606,
      "loss": 4.9803,
      "mean_token_accuracy": 0.19649519324302672,
      "num_tokens": 61282594.0,
      "step": 33220
    },
    {
      "entropy": 5.486295604705811,
      "epoch": 2.7913463558076037,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004221773517472174,
      "loss": 5.0763,
      "mean_token_accuracy": 0.1942099004983902,
      "num_tokens": 61291262.0,
      "step": 33225
    },
    {
      "entropy": 5.437606954574585,
      "epoch": 2.791766435622768,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042215469919021766,
      "loss": 5.1383,
      "mean_token_accuracy": 0.19049413055181502,
      "num_tokens": 61300562.0,
      "step": 33230
    },
    {
      "entropy": 5.520679378509522,
      "epoch": 2.7921865154379333,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042213204402644826,
      "loss": 5.134,
      "mean_token_accuracy": 0.19199418425559997,
      "num_tokens": 61310522.0,
      "step": 33235
    },
    {
      "entropy": 5.543746376037598,
      "epoch": 2.792606595253098,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042210938625631046,
      "loss": 5.1736,
      "mean_token_accuracy": 0.19497582763433458,
      "num_tokens": 61319654.0,
      "step": 33240
    },
    {
      "entropy": 5.485003328323364,
      "epoch": 2.793026675068263,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004220867258802056,
      "loss": 5.1174,
      "mean_token_accuracy": 0.19896575808525085,
      "num_tokens": 61329002.0,
      "step": 33245
    },
    {
      "entropy": 5.45067982673645,
      "epoch": 2.793446754883428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004220640628985352,
      "loss": 5.0729,
      "mean_token_accuracy": 0.19353261291980745,
      "num_tokens": 61338884.0,
      "step": 33250
    },
    {
      "entropy": 5.518244123458862,
      "epoch": 2.793866834698593,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042204139731170064,
      "loss": 5.136,
      "mean_token_accuracy": 0.19356969743967056,
      "num_tokens": 61348248.0,
      "step": 33255
    },
    {
      "entropy": 5.564935159683228,
      "epoch": 2.7942869145137577,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004220187291201034,
      "loss": 5.2361,
      "mean_token_accuracy": 0.18301527798175812,
      "num_tokens": 61358154.0,
      "step": 33260
    },
    {
      "entropy": 5.538912582397461,
      "epoch": 2.7947069943289224,
      "grad_norm": 1.0,
      "learning_rate": 0.0004219960583241451,
      "loss": 5.1296,
      "mean_token_accuracy": 0.18879839777946472,
      "num_tokens": 61368225.0,
      "step": 33265
    },
    {
      "entropy": 5.389934968948364,
      "epoch": 2.7951270741440872,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004219733849242274,
      "loss": 4.9447,
      "mean_token_accuracy": 0.20502110272645951,
      "num_tokens": 61376721.0,
      "step": 33270
    },
    {
      "entropy": 5.454366159439087,
      "epoch": 2.7955471539592525,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004219507089207518,
      "loss": 5.1533,
      "mean_token_accuracy": 0.18677590936422347,
      "num_tokens": 61386333.0,
      "step": 33275
    },
    {
      "entropy": 5.374199247360229,
      "epoch": 2.7959672337744172,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042192803031412,
      "loss": 5.0432,
      "mean_token_accuracy": 0.1889810636639595,
      "num_tokens": 61394993.0,
      "step": 33280
    },
    {
      "entropy": 5.580218553543091,
      "epoch": 2.796387313589582,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004219053491047338,
      "loss": 5.233,
      "mean_token_accuracy": 0.17894484996795654,
      "num_tokens": 61404703.0,
      "step": 33285
    },
    {
      "entropy": 5.521010684967041,
      "epoch": 2.796807393404747,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042188266529299506,
      "loss": 5.0855,
      "mean_token_accuracy": 0.19230024367570878,
      "num_tokens": 61413433.0,
      "step": 33290
    },
    {
      "entropy": 5.528345394134521,
      "epoch": 2.7972274732199116,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004218599788793055,
      "loss": 5.2161,
      "mean_token_accuracy": 0.18924675285816192,
      "num_tokens": 61422979.0,
      "step": 33295
    },
    {
      "entropy": 5.5700953006744385,
      "epoch": 2.797647553035077,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004218372898640671,
      "loss": 5.1739,
      "mean_token_accuracy": 0.18398604393005372,
      "num_tokens": 61432637.0,
      "step": 33300
    },
    {
      "entropy": 5.503483486175537,
      "epoch": 2.7980676328502416,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042181459824768166,
      "loss": 5.1239,
      "mean_token_accuracy": 0.19097760021686555,
      "num_tokens": 61441951.0,
      "step": 33305
    },
    {
      "entropy": 5.508897924423218,
      "epoch": 2.7984877126654064,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004217919040305512,
      "loss": 5.0707,
      "mean_token_accuracy": 0.19191221594810487,
      "num_tokens": 61451564.0,
      "step": 33310
    },
    {
      "entropy": 5.4745887279510494,
      "epoch": 2.798907792480571,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004217692072130777,
      "loss": 5.1807,
      "mean_token_accuracy": 0.18688035309314727,
      "num_tokens": 61460374.0,
      "step": 33315
    },
    {
      "entropy": 5.501093101501465,
      "epoch": 2.799327872295736,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042174650779566334,
      "loss": 5.1654,
      "mean_token_accuracy": 0.18375305086374283,
      "num_tokens": 61469420.0,
      "step": 33320
    },
    {
      "entropy": 5.581950664520264,
      "epoch": 2.799747952110901,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004217238057787102,
      "loss": 5.1436,
      "mean_token_accuracy": 0.19261240512132644,
      "num_tokens": 61479835.0,
      "step": 33325
    },
    {
      "entropy": 5.552571773529053,
      "epoch": 2.800168031926066,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042170110116262035,
      "loss": 5.1819,
      "mean_token_accuracy": 0.18721874356269835,
      "num_tokens": 61490176.0,
      "step": 33330
    },
    {
      "entropy": 5.328046178817749,
      "epoch": 2.800588111741231,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000421678393947796,
      "loss": 4.9398,
      "mean_token_accuracy": 0.20450820475816728,
      "num_tokens": 61499893.0,
      "step": 33335
    },
    {
      "entropy": 5.497897148132324,
      "epoch": 2.8010081915563956,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004216556841346395,
      "loss": 5.1753,
      "mean_token_accuracy": 0.183819542825222,
      "num_tokens": 61508610.0,
      "step": 33340
    },
    {
      "entropy": 5.5182616233825685,
      "epoch": 2.8014282713715604,
      "grad_norm": 1.140625,
      "learning_rate": 0.000421632971723553,
      "loss": 5.1035,
      "mean_token_accuracy": 0.19582903981208802,
      "num_tokens": 61518118.0,
      "step": 33345
    },
    {
      "entropy": 5.529508876800537,
      "epoch": 2.8018483511867256,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042161025671493885,
      "loss": 5.1592,
      "mean_token_accuracy": 0.19036707580089568,
      "num_tokens": 61527930.0,
      "step": 33350
    },
    {
      "entropy": 5.545594692230225,
      "epoch": 2.8022684310018904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004215875391091996,
      "loss": 5.137,
      "mean_token_accuracy": 0.1901567906141281,
      "num_tokens": 61537232.0,
      "step": 33355
    },
    {
      "entropy": 5.511520147323608,
      "epoch": 2.802688510817055,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004215648189067376,
      "loss": 5.1269,
      "mean_token_accuracy": 0.1943872094154358,
      "num_tokens": 61546752.0,
      "step": 33360
    },
    {
      "entropy": 5.427586889266967,
      "epoch": 2.8031085906322204,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042154209610795526,
      "loss": 5.0436,
      "mean_token_accuracy": 0.2015354037284851,
      "num_tokens": 61555999.0,
      "step": 33365
    },
    {
      "entropy": 5.469909620285034,
      "epoch": 2.8035286704473847,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042151937071325515,
      "loss": 5.1781,
      "mean_token_accuracy": 0.19389662891626358,
      "num_tokens": 61565854.0,
      "step": 33370
    },
    {
      "entropy": 5.536516571044922,
      "epoch": 2.80394875026255,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042149664272303977,
      "loss": 5.0314,
      "mean_token_accuracy": 0.20370899587869645,
      "num_tokens": 61573382.0,
      "step": 33375
    },
    {
      "entropy": 5.48340950012207,
      "epoch": 2.8043688300777148,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004214739121377119,
      "loss": 5.1127,
      "mean_token_accuracy": 0.1991887167096138,
      "num_tokens": 61582361.0,
      "step": 33380
    },
    {
      "entropy": 5.540915298461914,
      "epoch": 2.8047889098928795,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000421451178957674,
      "loss": 5.126,
      "mean_token_accuracy": 0.19459474086761475,
      "num_tokens": 61591797.0,
      "step": 33385
    },
    {
      "entropy": 5.426541185379028,
      "epoch": 2.8052089897080448,
      "grad_norm": 1.078125,
      "learning_rate": 0.000421428443183329,
      "loss": 5.0109,
      "mean_token_accuracy": 0.203157676756382,
      "num_tokens": 61600380.0,
      "step": 33390
    },
    {
      "entropy": 5.484256362915039,
      "epoch": 2.8056290695232096,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042140570481507937,
      "loss": 5.175,
      "mean_token_accuracy": 0.1839839279651642,
      "num_tokens": 61610544.0,
      "step": 33395
    },
    {
      "entropy": 5.455849885940552,
      "epoch": 2.8060491493383743,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004213829638533282,
      "loss": 5.0576,
      "mean_token_accuracy": 0.1973906561732292,
      "num_tokens": 61619444.0,
      "step": 33400
    },
    {
      "entropy": 5.463327789306641,
      "epoch": 2.806469229153539,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042136022029847815,
      "loss": 5.0556,
      "mean_token_accuracy": 0.19961448311805724,
      "num_tokens": 61628018.0,
      "step": 33405
    },
    {
      "entropy": 5.349229621887207,
      "epoch": 2.806889308968704,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042133747415093215,
      "loss": 4.9692,
      "mean_token_accuracy": 0.2117154359817505,
      "num_tokens": 61636884.0,
      "step": 33410
    },
    {
      "entropy": 5.468451833724975,
      "epoch": 2.807309388783869,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004213147254110932,
      "loss": 5.1693,
      "mean_token_accuracy": 0.18471244126558303,
      "num_tokens": 61647461.0,
      "step": 33415
    },
    {
      "entropy": 5.542232370376587,
      "epoch": 2.807729468599034,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004212919740793642,
      "loss": 5.0884,
      "mean_token_accuracy": 0.19516003429889678,
      "num_tokens": 61655641.0,
      "step": 33420
    },
    {
      "entropy": 5.529873466491699,
      "epoch": 2.8081495484141987,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004212692201561482,
      "loss": 5.1492,
      "mean_token_accuracy": 0.19410317987203599,
      "num_tokens": 61665131.0,
      "step": 33425
    },
    {
      "entropy": 5.451307868957519,
      "epoch": 2.8085696282293635,
      "grad_norm": 1.125,
      "learning_rate": 0.00042124646364184834,
      "loss": 5.0309,
      "mean_token_accuracy": 0.19553417414426805,
      "num_tokens": 61674645.0,
      "step": 33430
    },
    {
      "entropy": 5.3870563983917235,
      "epoch": 2.8089897080445283,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00042122370453686757,
      "loss": 5.0915,
      "mean_token_accuracy": 0.19182055443525314,
      "num_tokens": 61683457.0,
      "step": 33435
    },
    {
      "entropy": 5.565767478942871,
      "epoch": 2.8094097878596935,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042120094284160925,
      "loss": 5.193,
      "mean_token_accuracy": 0.1929889813065529,
      "num_tokens": 61691648.0,
      "step": 33440
    },
    {
      "entropy": 5.585212469100952,
      "epoch": 2.8098298676748583,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004211781785564765,
      "loss": 5.1765,
      "mean_token_accuracy": 0.1881064683198929,
      "num_tokens": 61701518.0,
      "step": 33445
    },
    {
      "entropy": 5.544825172424316,
      "epoch": 2.810249947490023,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042115541168187255,
      "loss": 5.2492,
      "mean_token_accuracy": 0.18502815663814545,
      "num_tokens": 61711517.0,
      "step": 33450
    },
    {
      "entropy": 5.4229736328125,
      "epoch": 2.810670027305188,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004211326422182009,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2035339206457138,
      "num_tokens": 61719931.0,
      "step": 33455
    },
    {
      "entropy": 5.523689556121826,
      "epoch": 2.8110901071203527,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042110987016586457,
      "loss": 5.1574,
      "mean_token_accuracy": 0.18871065229177475,
      "num_tokens": 61730038.0,
      "step": 33460
    },
    {
      "entropy": 5.509160327911377,
      "epoch": 2.811510186935518,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004210870955252671,
      "loss": 5.0865,
      "mean_token_accuracy": 0.19614242315292357,
      "num_tokens": 61738589.0,
      "step": 33465
    },
    {
      "entropy": 5.5036262512207035,
      "epoch": 2.8119302667506827,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042106431829681194,
      "loss": 5.1683,
      "mean_token_accuracy": 0.19379590451717377,
      "num_tokens": 61748602.0,
      "step": 33470
    },
    {
      "entropy": 5.434737682342529,
      "epoch": 2.8123503465658475,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004210415384809027,
      "loss": 5.1265,
      "mean_token_accuracy": 0.19243243038654329,
      "num_tokens": 61758381.0,
      "step": 33475
    },
    {
      "entropy": 5.567957162857056,
      "epoch": 2.8127704263810123,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004210187560779428,
      "loss": 5.1181,
      "mean_token_accuracy": 0.19714441448450087,
      "num_tokens": 61767556.0,
      "step": 33480
    },
    {
      "entropy": 5.4411458492279055,
      "epoch": 2.813190506196177,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042099597108833585,
      "loss": 5.131,
      "mean_token_accuracy": 0.19339325726032258,
      "num_tokens": 61776892.0,
      "step": 33485
    },
    {
      "entropy": 5.451198530197144,
      "epoch": 2.8136105860113423,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042097318351248537,
      "loss": 5.1303,
      "mean_token_accuracy": 0.1967436835169792,
      "num_tokens": 61786915.0,
      "step": 33490
    },
    {
      "entropy": 5.5777421474456785,
      "epoch": 2.814030665826507,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042095039335079515,
      "loss": 5.2175,
      "mean_token_accuracy": 0.18423013240098954,
      "num_tokens": 61796835.0,
      "step": 33495
    },
    {
      "entropy": 5.560711622238159,
      "epoch": 2.814450745641672,
      "grad_norm": 1.125,
      "learning_rate": 0.0004209276006036689,
      "loss": 5.1857,
      "mean_token_accuracy": 0.19189511686563493,
      "num_tokens": 61806154.0,
      "step": 33500
    },
    {
      "entropy": 5.486426210403442,
      "epoch": 2.8148708254568366,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042090480527151033,
      "loss": 5.0743,
      "mean_token_accuracy": 0.19450336843729019,
      "num_tokens": 61815610.0,
      "step": 33505
    },
    {
      "entropy": 5.529978942871094,
      "epoch": 2.8152909052720014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004208820073547232,
      "loss": 5.1479,
      "mean_token_accuracy": 0.1935117870569229,
      "num_tokens": 61825017.0,
      "step": 33510
    },
    {
      "entropy": 5.4804566383361815,
      "epoch": 2.8157109850871667,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004208592068537115,
      "loss": 5.1455,
      "mean_token_accuracy": 0.19135744571685792,
      "num_tokens": 61834535.0,
      "step": 33515
    },
    {
      "entropy": 5.439340019226075,
      "epoch": 2.8161310649023314,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042083640376887896,
      "loss": 5.065,
      "mean_token_accuracy": 0.19146513789892197,
      "num_tokens": 61843042.0,
      "step": 33520
    },
    {
      "entropy": 5.45637321472168,
      "epoch": 2.8165511447174962,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004208135981006297,
      "loss": 5.071,
      "mean_token_accuracy": 0.20019577145576478,
      "num_tokens": 61852019.0,
      "step": 33525
    },
    {
      "entropy": 5.459341239929199,
      "epoch": 2.8169712245326615,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004207907898493676,
      "loss": 5.1372,
      "mean_token_accuracy": 0.19684213995933533,
      "num_tokens": 61860599.0,
      "step": 33530
    },
    {
      "entropy": 5.569476175308227,
      "epoch": 2.8173913043478263,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004207679790154967,
      "loss": 5.1255,
      "mean_token_accuracy": 0.19502735286951065,
      "num_tokens": 61869048.0,
      "step": 33535
    },
    {
      "entropy": 5.532147932052612,
      "epoch": 2.817811384162991,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004207451655994212,
      "loss": 5.1185,
      "mean_token_accuracy": 0.19183538854122162,
      "num_tokens": 61878075.0,
      "step": 33540
    },
    {
      "entropy": 5.495197772979736,
      "epoch": 2.818231463978156,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004207223496015451,
      "loss": 5.1965,
      "mean_token_accuracy": 0.1869797706604004,
      "num_tokens": 61887570.0,
      "step": 33545
    },
    {
      "entropy": 5.465198802947998,
      "epoch": 2.8186515437933206,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042069953102227267,
      "loss": 5.1221,
      "mean_token_accuracy": 0.19052191078662872,
      "num_tokens": 61897518.0,
      "step": 33550
    },
    {
      "entropy": 5.45495753288269,
      "epoch": 2.819071623608486,
      "grad_norm": 1.609375,
      "learning_rate": 0.00042067670986200797,
      "loss": 4.9841,
      "mean_token_accuracy": 0.2001306563615799,
      "num_tokens": 61907467.0,
      "step": 33555
    },
    {
      "entropy": 5.482865238189698,
      "epoch": 2.8194917034236506,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004206538861211554,
      "loss": 5.0562,
      "mean_token_accuracy": 0.19727532118558883,
      "num_tokens": 61916663.0,
      "step": 33560
    },
    {
      "entropy": 5.400060415267944,
      "epoch": 2.8199117832388154,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004206310598001192,
      "loss": 5.0249,
      "mean_token_accuracy": 0.20178867131471634,
      "num_tokens": 61925238.0,
      "step": 33565
    },
    {
      "entropy": 5.45782585144043,
      "epoch": 2.82033186305398,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042060823089930385,
      "loss": 5.1694,
      "mean_token_accuracy": 0.1904926687479019,
      "num_tokens": 61934405.0,
      "step": 33570
    },
    {
      "entropy": 5.527444648742676,
      "epoch": 2.820751942869145,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042058539941911363,
      "loss": 5.2063,
      "mean_token_accuracy": 0.19105783402919768,
      "num_tokens": 61944608.0,
      "step": 33575
    },
    {
      "entropy": 5.453639030456543,
      "epoch": 2.82117202268431,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000420562565359953,
      "loss": 5.0431,
      "mean_token_accuracy": 0.1979518011212349,
      "num_tokens": 61954217.0,
      "step": 33580
    },
    {
      "entropy": 5.519848585128784,
      "epoch": 2.821592102499475,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004205397287222265,
      "loss": 5.178,
      "mean_token_accuracy": 0.18065424710512162,
      "num_tokens": 61965676.0,
      "step": 33585
    },
    {
      "entropy": 5.456130838394165,
      "epoch": 2.82201218231464,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042051688950633866,
      "loss": 5.0452,
      "mean_token_accuracy": 0.2001536399126053,
      "num_tokens": 61975557.0,
      "step": 33590
    },
    {
      "entropy": 5.513027238845825,
      "epoch": 2.8224322621298046,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042049404771269403,
      "loss": 5.1115,
      "mean_token_accuracy": 0.19324278086423874,
      "num_tokens": 61984303.0,
      "step": 33595
    },
    {
      "entropy": 5.515049171447754,
      "epoch": 2.8228523419449694,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004204712033416973,
      "loss": 5.1814,
      "mean_token_accuracy": 0.18800653666257858,
      "num_tokens": 61993634.0,
      "step": 33600
    },
    {
      "entropy": 5.450329732894898,
      "epoch": 2.8232724217601346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004204483563937531,
      "loss": 5.0259,
      "mean_token_accuracy": 0.19571841806173323,
      "num_tokens": 62002375.0,
      "step": 33605
    },
    {
      "entropy": 5.470686578750611,
      "epoch": 2.8236925015752994,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004204255068692661,
      "loss": 5.119,
      "mean_token_accuracy": 0.19020138084888458,
      "num_tokens": 62011266.0,
      "step": 33610
    },
    {
      "entropy": 5.464394235610962,
      "epoch": 2.824112581390464,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004204026547686412,
      "loss": 5.0673,
      "mean_token_accuracy": 0.19492223411798476,
      "num_tokens": 62020513.0,
      "step": 33615
    },
    {
      "entropy": 5.485654592514038,
      "epoch": 2.824532661205629,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004203798000922831,
      "loss": 5.2106,
      "mean_token_accuracy": 0.19158849269151687,
      "num_tokens": 62029947.0,
      "step": 33620
    },
    {
      "entropy": 5.571140432357788,
      "epoch": 2.8249527410207937,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004203569428405967,
      "loss": 5.2079,
      "mean_token_accuracy": 0.18541137874126434,
      "num_tokens": 62038619.0,
      "step": 33625
    },
    {
      "entropy": 5.52348952293396,
      "epoch": 2.825372820835959,
      "grad_norm": 1.0625,
      "learning_rate": 0.000420334083013987,
      "loss": 5.0744,
      "mean_token_accuracy": 0.19572861194610597,
      "num_tokens": 62047051.0,
      "step": 33630
    },
    {
      "entropy": 5.4238917350769045,
      "epoch": 2.8257929006511238,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004203112206128587,
      "loss": 5.0344,
      "mean_token_accuracy": 0.19574928879737855,
      "num_tokens": 62055721.0,
      "step": 33635
    },
    {
      "entropy": 5.408810806274414,
      "epoch": 2.8262129804662885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000420288355637617,
      "loss": 5.0863,
      "mean_token_accuracy": 0.1959209054708481,
      "num_tokens": 62064769.0,
      "step": 33640
    },
    {
      "entropy": 5.5009400844573975,
      "epoch": 2.8266330602814533,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042026548808866694,
      "loss": 5.1254,
      "mean_token_accuracy": 0.1906157836318016,
      "num_tokens": 62074246.0,
      "step": 33645
    },
    {
      "entropy": 5.469218015670776,
      "epoch": 2.827053140096618,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042024261796641353,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19620139598846437,
      "num_tokens": 62082542.0,
      "step": 33650
    },
    {
      "entropy": 5.453561210632325,
      "epoch": 2.8274732199117834,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042021974527126196,
      "loss": 5.0984,
      "mean_token_accuracy": 0.1884379580616951,
      "num_tokens": 62092543.0,
      "step": 33655
    },
    {
      "entropy": 5.575100994110107,
      "epoch": 2.827893299726948,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042019687000361734,
      "loss": 5.2579,
      "mean_token_accuracy": 0.18889280408620834,
      "num_tokens": 62101357.0,
      "step": 33660
    },
    {
      "entropy": 5.479830312728882,
      "epoch": 2.828313379542113,
      "grad_norm": 1.046875,
      "learning_rate": 0.000420173992163885,
      "loss": 5.0546,
      "mean_token_accuracy": 0.20096427351236343,
      "num_tokens": 62110867.0,
      "step": 33665
    },
    {
      "entropy": 5.424468994140625,
      "epoch": 2.828733459357278,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004201511117524701,
      "loss": 5.0161,
      "mean_token_accuracy": 0.19911625534296035,
      "num_tokens": 62120156.0,
      "step": 33670
    },
    {
      "entropy": 5.508634757995606,
      "epoch": 2.8291535391724425,
      "grad_norm": 1.0,
      "learning_rate": 0.00042012822876977794,
      "loss": 5.0865,
      "mean_token_accuracy": 0.19290708005428314,
      "num_tokens": 62129382.0,
      "step": 33675
    },
    {
      "entropy": 5.497140312194825,
      "epoch": 2.8295736189876077,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000420105343216214,
      "loss": 5.0621,
      "mean_token_accuracy": 0.19968939423561097,
      "num_tokens": 62138508.0,
      "step": 33680
    },
    {
      "entropy": 5.5444474697113035,
      "epoch": 2.8299936988027725,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004200824550921836,
      "loss": 5.185,
      "mean_token_accuracy": 0.19184849113225938,
      "num_tokens": 62146812.0,
      "step": 33685
    },
    {
      "entropy": 5.551350355148315,
      "epoch": 2.8304137786179373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042005956439809225,
      "loss": 5.1694,
      "mean_token_accuracy": 0.18907350152730942,
      "num_tokens": 62156460.0,
      "step": 33690
    },
    {
      "entropy": 5.4768946170806885,
      "epoch": 2.8308338584331025,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042003667113434535,
      "loss": 5.1217,
      "mean_token_accuracy": 0.19072835892438889,
      "num_tokens": 62165916.0,
      "step": 33695
    },
    {
      "entropy": 5.437673282623291,
      "epoch": 2.8312539382482673,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004200137753013486,
      "loss": 5.1011,
      "mean_token_accuracy": 0.19133810549974442,
      "num_tokens": 62175394.0,
      "step": 33700
    },
    {
      "entropy": 5.501942825317383,
      "epoch": 2.831674018063432,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00041999087689950746,
      "loss": 5.0748,
      "mean_token_accuracy": 0.19463516175746917,
      "num_tokens": 62185026.0,
      "step": 33705
    },
    {
      "entropy": 5.443242502212525,
      "epoch": 2.832094097878597,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041996797592922755,
      "loss": 5.0695,
      "mean_token_accuracy": 0.19674620777368546,
      "num_tokens": 62194282.0,
      "step": 33710
    },
    {
      "entropy": 5.4145272731781,
      "epoch": 2.8325141776937617,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004199450723909146,
      "loss": 5.074,
      "mean_token_accuracy": 0.19217370301485062,
      "num_tokens": 62203693.0,
      "step": 33715
    },
    {
      "entropy": 5.558022928237915,
      "epoch": 2.832934257508927,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004199221662849743,
      "loss": 5.1686,
      "mean_token_accuracy": 0.1900000676512718,
      "num_tokens": 62213553.0,
      "step": 33720
    },
    {
      "entropy": 5.533995914459228,
      "epoch": 2.8333543373240917,
      "grad_norm": 1.0,
      "learning_rate": 0.00041989925761181247,
      "loss": 5.1246,
      "mean_token_accuracy": 0.19587934166193008,
      "num_tokens": 62222663.0,
      "step": 33725
    },
    {
      "entropy": 5.448152637481689,
      "epoch": 2.8337744171392565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041987634637183493,
      "loss": 5.0899,
      "mean_token_accuracy": 0.19004160463809966,
      "num_tokens": 62231028.0,
      "step": 33730
    },
    {
      "entropy": 5.393830013275147,
      "epoch": 2.8341944969544213,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004198534325654475,
      "loss": 5.0808,
      "mean_token_accuracy": 0.2032485619187355,
      "num_tokens": 62240015.0,
      "step": 33735
    },
    {
      "entropy": 5.466780471801758,
      "epoch": 2.834614576769586,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041983051619305607,
      "loss": 5.0931,
      "mean_token_accuracy": 0.1908632054924965,
      "num_tokens": 62249265.0,
      "step": 33740
    },
    {
      "entropy": 5.471548461914063,
      "epoch": 2.8350346565847513,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004198075972550667,
      "loss": 5.0927,
      "mean_token_accuracy": 0.19085631221532823,
      "num_tokens": 62258267.0,
      "step": 33745
    },
    {
      "entropy": 5.483861780166626,
      "epoch": 2.835454736399916,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004197846757518853,
      "loss": 5.0956,
      "mean_token_accuracy": 0.19315665066242219,
      "num_tokens": 62268384.0,
      "step": 33750
    },
    {
      "entropy": 5.425298690795898,
      "epoch": 2.835874816215081,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041976175168391785,
      "loss": 4.9902,
      "mean_token_accuracy": 0.2011181131005287,
      "num_tokens": 62277262.0,
      "step": 33755
    },
    {
      "entropy": 5.540938854217529,
      "epoch": 2.8362948960302456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041973882505157053,
      "loss": 5.0913,
      "mean_token_accuracy": 0.1889183074235916,
      "num_tokens": 62285271.0,
      "step": 33760
    },
    {
      "entropy": 5.440981864929199,
      "epoch": 2.8367149758454104,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004197158958552495,
      "loss": 5.0383,
      "mean_token_accuracy": 0.20040292143821717,
      "num_tokens": 62293555.0,
      "step": 33765
    },
    {
      "entropy": 5.455485105514526,
      "epoch": 2.8371350556605757,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041969296409536083,
      "loss": 5.1044,
      "mean_token_accuracy": 0.18841795027256011,
      "num_tokens": 62303563.0,
      "step": 33770
    },
    {
      "entropy": 5.469950151443482,
      "epoch": 2.8375551354757405,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004196700297723109,
      "loss": 5.1034,
      "mean_token_accuracy": 0.1953592523932457,
      "num_tokens": 62312040.0,
      "step": 33775
    },
    {
      "entropy": 5.533995771408081,
      "epoch": 2.8379752152909052,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004196470928865058,
      "loss": 5.1984,
      "mean_token_accuracy": 0.18816658109426498,
      "num_tokens": 62322147.0,
      "step": 33780
    },
    {
      "entropy": 5.509495258331299,
      "epoch": 2.83839529510607,
      "grad_norm": 1.078125,
      "learning_rate": 0.000419624153438352,
      "loss": 5.094,
      "mean_token_accuracy": 0.19641731530427933,
      "num_tokens": 62331959.0,
      "step": 33785
    },
    {
      "entropy": 5.454195594787597,
      "epoch": 2.838815374921235,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004196012114282557,
      "loss": 5.08,
      "mean_token_accuracy": 0.19552946239709854,
      "num_tokens": 62341578.0,
      "step": 33790
    },
    {
      "entropy": 5.444974708557129,
      "epoch": 2.8392354547364,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041957826685662343,
      "loss": 5.0904,
      "mean_token_accuracy": 0.2003238707780838,
      "num_tokens": 62350421.0,
      "step": 33795
    },
    {
      "entropy": 5.577842330932617,
      "epoch": 2.839655534551565,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004195553197238617,
      "loss": 5.1186,
      "mean_token_accuracy": 0.19164956659078597,
      "num_tokens": 62359528.0,
      "step": 33800
    },
    {
      "entropy": 5.453657817840576,
      "epoch": 2.8400756143667296,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004195323700303768,
      "loss": 5.119,
      "mean_token_accuracy": 0.19209021627902984,
      "num_tokens": 62369041.0,
      "step": 33805
    },
    {
      "entropy": 5.420483636856079,
      "epoch": 2.8404956941818944,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041950941777657547,
      "loss": 4.9644,
      "mean_token_accuracy": 0.199736787378788,
      "num_tokens": 62377536.0,
      "step": 33810
    },
    {
      "entropy": 5.516615486145019,
      "epoch": 2.840915773997059,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004194864629628642,
      "loss": 5.2365,
      "mean_token_accuracy": 0.18690713346004487,
      "num_tokens": 62388002.0,
      "step": 33815
    },
    {
      "entropy": 5.507530593872071,
      "epoch": 2.8413358538122244,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004194635055896496,
      "loss": 5.1123,
      "mean_token_accuracy": 0.1953244313597679,
      "num_tokens": 62396332.0,
      "step": 33820
    },
    {
      "entropy": 5.499201345443725,
      "epoch": 2.841755933627389,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041944054565733836,
      "loss": 5.0799,
      "mean_token_accuracy": 0.19576286971569062,
      "num_tokens": 62405137.0,
      "step": 33825
    },
    {
      "entropy": 5.500701761245727,
      "epoch": 2.842176013442554,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041941758316633727,
      "loss": 5.1861,
      "mean_token_accuracy": 0.19123488366603852,
      "num_tokens": 62413849.0,
      "step": 33830
    },
    {
      "entropy": 5.505924797058105,
      "epoch": 2.8425960932577192,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041939461811705306,
      "loss": 5.0848,
      "mean_token_accuracy": 0.19444913566112518,
      "num_tokens": 62422693.0,
      "step": 33835
    },
    {
      "entropy": 5.524555683135986,
      "epoch": 2.843016173072884,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041937165050989257,
      "loss": 5.1487,
      "mean_token_accuracy": 0.18680481761693954,
      "num_tokens": 62432960.0,
      "step": 33840
    },
    {
      "entropy": 5.386117553710937,
      "epoch": 2.843436252888049,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041934868034526265,
      "loss": 5.0188,
      "mean_token_accuracy": 0.19842917919158937,
      "num_tokens": 62441838.0,
      "step": 33845
    },
    {
      "entropy": 5.480356693267822,
      "epoch": 2.8438563327032136,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004193257076235701,
      "loss": 5.0894,
      "mean_token_accuracy": 0.19500146806240082,
      "num_tokens": 62451111.0,
      "step": 33850
    },
    {
      "entropy": 5.501254987716675,
      "epoch": 2.8442764125183784,
      "grad_norm": 1.203125,
      "learning_rate": 0.000419302732345222,
      "loss": 5.1727,
      "mean_token_accuracy": 0.19054460227489473,
      "num_tokens": 62459975.0,
      "step": 33855
    },
    {
      "entropy": 5.593841075897217,
      "epoch": 2.8446964923335436,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004192797545106252,
      "loss": 5.2137,
      "mean_token_accuracy": 0.1872469648718834,
      "num_tokens": 62469529.0,
      "step": 33860
    },
    {
      "entropy": 5.477652883529663,
      "epoch": 2.8451165721487084,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000419256774120187,
      "loss": 5.0762,
      "mean_token_accuracy": 0.1971086084842682,
      "num_tokens": 62479122.0,
      "step": 33865
    },
    {
      "entropy": 5.49901180267334,
      "epoch": 2.845536651963873,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004192337911743143,
      "loss": 5.1197,
      "mean_token_accuracy": 0.19378749877214432,
      "num_tokens": 62489457.0,
      "step": 33870
    },
    {
      "entropy": 5.4831109046936035,
      "epoch": 2.845956731779038,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004192108056734142,
      "loss": 5.0929,
      "mean_token_accuracy": 0.20581929087638856,
      "num_tokens": 62499063.0,
      "step": 33875
    },
    {
      "entropy": 5.538233280181885,
      "epoch": 2.8463768115942027,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004191878176178939,
      "loss": 5.0665,
      "mean_token_accuracy": 0.19780644327402114,
      "num_tokens": 62507728.0,
      "step": 33880
    },
    {
      "entropy": 5.577447700500488,
      "epoch": 2.846796891409368,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004191648270081608,
      "loss": 5.2488,
      "mean_token_accuracy": 0.18043371737003328,
      "num_tokens": 62517770.0,
      "step": 33885
    },
    {
      "entropy": 5.5524426937103275,
      "epoch": 2.8472169712245328,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000419141833844622,
      "loss": 5.1927,
      "mean_token_accuracy": 0.19034942835569382,
      "num_tokens": 62526988.0,
      "step": 33890
    },
    {
      "entropy": 5.467883777618408,
      "epoch": 2.8476370510396976,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004191188381276848,
      "loss": 5.1056,
      "mean_token_accuracy": 0.19561028182506562,
      "num_tokens": 62536096.0,
      "step": 33895
    },
    {
      "entropy": 5.504553890228271,
      "epoch": 2.8480571308548623,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004190958398577565,
      "loss": 5.1095,
      "mean_token_accuracy": 0.19537804424762725,
      "num_tokens": 62546316.0,
      "step": 33900
    },
    {
      "entropy": 5.4882114887237545,
      "epoch": 2.848477210670027,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004190728390352447,
      "loss": 4.9977,
      "mean_token_accuracy": 0.20013438165187836,
      "num_tokens": 62555141.0,
      "step": 33905
    },
    {
      "entropy": 5.401898908615112,
      "epoch": 2.8488972904851924,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004190498356605567,
      "loss": 5.1426,
      "mean_token_accuracy": 0.19038619995117187,
      "num_tokens": 62564459.0,
      "step": 33910
    },
    {
      "entropy": 5.566623306274414,
      "epoch": 2.849317370300357,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004190268297341001,
      "loss": 5.1453,
      "mean_token_accuracy": 0.1900817260146141,
      "num_tokens": 62573740.0,
      "step": 33915
    },
    {
      "entropy": 5.632818603515625,
      "epoch": 2.849737450115522,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004190038212562822,
      "loss": 5.1993,
      "mean_token_accuracy": 0.1850172460079193,
      "num_tokens": 62582727.0,
      "step": 33920
    },
    {
      "entropy": 5.4810295581817625,
      "epoch": 2.8501575299306867,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004189808102275109,
      "loss": 5.0728,
      "mean_token_accuracy": 0.19775340259075164,
      "num_tokens": 62591722.0,
      "step": 33925
    },
    {
      "entropy": 5.529133558273315,
      "epoch": 2.8505776097458515,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041895779664819367,
      "loss": 5.1744,
      "mean_token_accuracy": 0.18562400490045547,
      "num_tokens": 62600127.0,
      "step": 33930
    },
    {
      "entropy": 5.525624704360962,
      "epoch": 2.8509976895610167,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004189347805187381,
      "loss": 5.1796,
      "mean_token_accuracy": 0.18933746963739395,
      "num_tokens": 62609452.0,
      "step": 33935
    },
    {
      "entropy": 5.566415882110595,
      "epoch": 2.8514177693761815,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004189117618395522,
      "loss": 5.1582,
      "mean_token_accuracy": 0.19129081070423126,
      "num_tokens": 62618390.0,
      "step": 33940
    },
    {
      "entropy": 5.502596044540406,
      "epoch": 2.8518378491913463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004188887406110434,
      "loss": 5.1354,
      "mean_token_accuracy": 0.19130517840385436,
      "num_tokens": 62627088.0,
      "step": 33945
    },
    {
      "entropy": 5.46813473701477,
      "epoch": 2.852257929006511,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041886571683361973,
      "loss": 4.9921,
      "mean_token_accuracy": 0.19774828106164932,
      "num_tokens": 62635318.0,
      "step": 33950
    },
    {
      "entropy": 5.490778875350952,
      "epoch": 2.852678008821676,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004188426905076889,
      "loss": 5.1289,
      "mean_token_accuracy": 0.18993591219186784,
      "num_tokens": 62644226.0,
      "step": 33955
    },
    {
      "entropy": 5.507098531723022,
      "epoch": 2.853098088636841,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004188196616336589,
      "loss": 5.0878,
      "mean_token_accuracy": 0.197640360891819,
      "num_tokens": 62653403.0,
      "step": 33960
    },
    {
      "entropy": 5.521958208084106,
      "epoch": 2.853518168452006,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004187966302119377,
      "loss": 5.0774,
      "mean_token_accuracy": 0.1977795273065567,
      "num_tokens": 62662711.0,
      "step": 33965
    },
    {
      "entropy": 5.523420333862305,
      "epoch": 2.8539382482671707,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041877359624293317,
      "loss": 5.0595,
      "mean_token_accuracy": 0.19240635335445405,
      "num_tokens": 62671673.0,
      "step": 33970
    },
    {
      "entropy": 5.49688606262207,
      "epoch": 2.854358328082336,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041875055972705346,
      "loss": 5.0306,
      "mean_token_accuracy": 0.19570242315530778,
      "num_tokens": 62681479.0,
      "step": 33975
    },
    {
      "entropy": 5.3620696544647215,
      "epoch": 2.8547784078975003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004187275206647066,
      "loss": 5.0217,
      "mean_token_accuracy": 0.2006630688905716,
      "num_tokens": 62690487.0,
      "step": 33980
    },
    {
      "entropy": 5.483897733688354,
      "epoch": 2.8551984877126655,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004187044790563007,
      "loss": 5.02,
      "mean_token_accuracy": 0.1974349945783615,
      "num_tokens": 62699438.0,
      "step": 33985
    },
    {
      "entropy": 5.549653339385986,
      "epoch": 2.8556185675278303,
      "grad_norm": 0.984375,
      "learning_rate": 0.00041868143490224395,
      "loss": 5.1488,
      "mean_token_accuracy": 0.19411643743515014,
      "num_tokens": 62708856.0,
      "step": 33990
    },
    {
      "entropy": 5.495194959640503,
      "epoch": 2.856038647342995,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041865838820294454,
      "loss": 5.1116,
      "mean_token_accuracy": 0.19384364783763885,
      "num_tokens": 62718759.0,
      "step": 33995
    },
    {
      "entropy": 5.53157958984375,
      "epoch": 2.8564587271581603,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004186353389588108,
      "loss": 5.1656,
      "mean_token_accuracy": 0.1876309335231781,
      "num_tokens": 62729112.0,
      "step": 34000
    },
    {
      "entropy": 5.525336790084839,
      "epoch": 2.856878806973325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041861228717025097,
      "loss": 5.2037,
      "mean_token_accuracy": 0.1827524036169052,
      "num_tokens": 62738688.0,
      "step": 34005
    },
    {
      "entropy": 5.572087097167969,
      "epoch": 2.85729888678849,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041858923283767343,
      "loss": 5.2096,
      "mean_token_accuracy": 0.18601759672164916,
      "num_tokens": 62748818.0,
      "step": 34010
    },
    {
      "entropy": 5.465973520278931,
      "epoch": 2.8577189666036547,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041856617596148653,
      "loss": 5.0409,
      "mean_token_accuracy": 0.19844723045825957,
      "num_tokens": 62757739.0,
      "step": 34015
    },
    {
      "entropy": 5.475200366973877,
      "epoch": 2.8581390464188194,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004185431165420988,
      "loss": 5.1071,
      "mean_token_accuracy": 0.19714057743549346,
      "num_tokens": 62766050.0,
      "step": 34020
    },
    {
      "entropy": 5.45564374923706,
      "epoch": 2.8585591262339847,
      "grad_norm": 1.25,
      "learning_rate": 0.0004185200545799186,
      "loss": 5.1235,
      "mean_token_accuracy": 0.18994459062814711,
      "num_tokens": 62776803.0,
      "step": 34025
    },
    {
      "entropy": 5.565985298156738,
      "epoch": 2.8589792060491495,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041849699007535474,
      "loss": 5.1716,
      "mean_token_accuracy": 0.17921035140752792,
      "num_tokens": 62785846.0,
      "step": 34030
    },
    {
      "entropy": 5.449346208572388,
      "epoch": 2.8593992858643142,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004184739230288154,
      "loss": 4.9847,
      "mean_token_accuracy": 0.20307222455739976,
      "num_tokens": 62794656.0,
      "step": 34035
    },
    {
      "entropy": 5.460077524185181,
      "epoch": 2.859819365679479,
      "grad_norm": 1.0,
      "learning_rate": 0.00041845085344070944,
      "loss": 5.0312,
      "mean_token_accuracy": 0.19717591404914855,
      "num_tokens": 62804102.0,
      "step": 34040
    },
    {
      "entropy": 5.506455039978027,
      "epoch": 2.860239445494644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004184277813114455,
      "loss": 5.1663,
      "mean_token_accuracy": 0.18952234834432602,
      "num_tokens": 62813345.0,
      "step": 34045
    },
    {
      "entropy": 5.5227796077728275,
      "epoch": 2.860659525309809,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041840470664143226,
      "loss": 5.1552,
      "mean_token_accuracy": 0.19265616089105606,
      "num_tokens": 62822666.0,
      "step": 34050
    },
    {
      "entropy": 5.45999174118042,
      "epoch": 2.861079605124974,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041838162943107844,
      "loss": 5.0198,
      "mean_token_accuracy": 0.20829365849494935,
      "num_tokens": 62831313.0,
      "step": 34055
    },
    {
      "entropy": 5.466146516799927,
      "epoch": 2.8614996849401386,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004183585496807929,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19459480345249175,
      "num_tokens": 62840697.0,
      "step": 34060
    },
    {
      "entropy": 5.466827869415283,
      "epoch": 2.8619197647553034,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004183354673909845,
      "loss": 5.0898,
      "mean_token_accuracy": 0.20025592297315598,
      "num_tokens": 62850415.0,
      "step": 34065
    },
    {
      "entropy": 5.450450849533081,
      "epoch": 2.862339844570468,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041831238256206206,
      "loss": 5.0757,
      "mean_token_accuracy": 0.19228608012199402,
      "num_tokens": 62859470.0,
      "step": 34070
    },
    {
      "entropy": 5.530712509155274,
      "epoch": 2.8627599243856334,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004182892951944346,
      "loss": 5.1706,
      "mean_token_accuracy": 0.18324217796325684,
      "num_tokens": 62869925.0,
      "step": 34075
    },
    {
      "entropy": 5.510244941711425,
      "epoch": 2.863180004200798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000418266205288511,
      "loss": 5.1047,
      "mean_token_accuracy": 0.19874086380004882,
      "num_tokens": 62879848.0,
      "step": 34080
    },
    {
      "entropy": 5.4994710922241214,
      "epoch": 2.863600084015963,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00041824311284470044,
      "loss": 5.0387,
      "mean_token_accuracy": 0.19725942611694336,
      "num_tokens": 62889221.0,
      "step": 34085
    },
    {
      "entropy": 5.4581968784332275,
      "epoch": 2.864020163831128,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041822001786341183,
      "loss": 5.0532,
      "mean_token_accuracy": 0.193779256939888,
      "num_tokens": 62899257.0,
      "step": 34090
    },
    {
      "entropy": 5.469721508026123,
      "epoch": 2.8644402436462926,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041819692034505436,
      "loss": 5.1189,
      "mean_token_accuracy": 0.18273680210113524,
      "num_tokens": 62908467.0,
      "step": 34095
    },
    {
      "entropy": 5.407088851928711,
      "epoch": 2.864860323461458,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004181738202900372,
      "loss": 4.9692,
      "mean_token_accuracy": 0.20952391922473906,
      "num_tokens": 62918264.0,
      "step": 34100
    },
    {
      "entropy": 5.454645586013794,
      "epoch": 2.8652804032766226,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004181507176987695,
      "loss": 5.035,
      "mean_token_accuracy": 0.20323162376880646,
      "num_tokens": 62926208.0,
      "step": 34105
    },
    {
      "entropy": 5.522467994689942,
      "epoch": 2.8657004830917874,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004181276125716606,
      "loss": 5.179,
      "mean_token_accuracy": 0.18826130777597427,
      "num_tokens": 62934761.0,
      "step": 34110
    },
    {
      "entropy": 5.538998746871949,
      "epoch": 2.866120562906952,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004181045049091197,
      "loss": 5.1531,
      "mean_token_accuracy": 0.18768304586410522,
      "num_tokens": 62944653.0,
      "step": 34115
    },
    {
      "entropy": 5.5776145458221436,
      "epoch": 2.866540642722117,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041808139471155624,
      "loss": 5.0719,
      "mean_token_accuracy": 0.20196710973978044,
      "num_tokens": 62952829.0,
      "step": 34120
    },
    {
      "entropy": 5.41392502784729,
      "epoch": 2.866960722537282,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041805828197937953,
      "loss": 5.0415,
      "mean_token_accuracy": 0.1963067352771759,
      "num_tokens": 62961577.0,
      "step": 34125
    },
    {
      "entropy": 5.4747096538543705,
      "epoch": 2.867380802352447,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004180351667129989,
      "loss": 5.0954,
      "mean_token_accuracy": 0.19882970303297043,
      "num_tokens": 62970911.0,
      "step": 34130
    },
    {
      "entropy": 5.539982748031616,
      "epoch": 2.8678008821676118,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004180120489128241,
      "loss": 5.2249,
      "mean_token_accuracy": 0.18617857545614241,
      "num_tokens": 62980280.0,
      "step": 34135
    },
    {
      "entropy": 5.543603420257568,
      "epoch": 2.868220961982777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004179889285792644,
      "loss": 5.1088,
      "mean_token_accuracy": 0.19193854033946992,
      "num_tokens": 62989504.0,
      "step": 34140
    },
    {
      "entropy": 5.531171274185181,
      "epoch": 2.8686410417979418,
      "grad_norm": 1.125,
      "learning_rate": 0.00041796580571272946,
      "loss": 5.1464,
      "mean_token_accuracy": 0.19478082954883574,
      "num_tokens": 62998815.0,
      "step": 34145
    },
    {
      "entropy": 5.516097593307495,
      "epoch": 2.8690611216131066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004179426803136289,
      "loss": 5.1404,
      "mean_token_accuracy": 0.19125259518623353,
      "num_tokens": 63009280.0,
      "step": 34150
    },
    {
      "entropy": 5.582079935073852,
      "epoch": 2.8694812014282713,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004179195523823724,
      "loss": 5.1963,
      "mean_token_accuracy": 0.19064952284097672,
      "num_tokens": 63018732.0,
      "step": 34155
    },
    {
      "entropy": 5.4638721466064455,
      "epoch": 2.869901281243436,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004178964219193697,
      "loss": 5.0231,
      "mean_token_accuracy": 0.19585521668195724,
      "num_tokens": 63027761.0,
      "step": 34160
    },
    {
      "entropy": 5.476844930648804,
      "epoch": 2.8703213610586014,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041787328892503036,
      "loss": 5.1429,
      "mean_token_accuracy": 0.18987800776958466,
      "num_tokens": 63036562.0,
      "step": 34165
    },
    {
      "entropy": 5.564364910125732,
      "epoch": 2.870741440873766,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041785015339976424,
      "loss": 5.1479,
      "mean_token_accuracy": 0.19185880571603775,
      "num_tokens": 63045448.0,
      "step": 34170
    },
    {
      "entropy": 5.496568822860718,
      "epoch": 2.871161520688931,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004178270153439812,
      "loss": 5.0718,
      "mean_token_accuracy": 0.1995813399553299,
      "num_tokens": 63054677.0,
      "step": 34175
    },
    {
      "entropy": 5.4374183177947994,
      "epoch": 2.8715816005040957,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004178038747580912,
      "loss": 5.029,
      "mean_token_accuracy": 0.19805348068475723,
      "num_tokens": 63064498.0,
      "step": 34180
    },
    {
      "entropy": 5.347333335876465,
      "epoch": 2.8720016803192605,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041778073164250415,
      "loss": 5.0551,
      "mean_token_accuracy": 0.19753432124853135,
      "num_tokens": 63073252.0,
      "step": 34185
    },
    {
      "entropy": 5.494639873504639,
      "epoch": 2.8724217601344257,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041775758599762986,
      "loss": 5.0963,
      "mean_token_accuracy": 0.1977521449327469,
      "num_tokens": 63082397.0,
      "step": 34190
    },
    {
      "entropy": 5.5200756072998045,
      "epoch": 2.8728418399495905,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041773443782387845,
      "loss": 5.1307,
      "mean_token_accuracy": 0.18866875469684602,
      "num_tokens": 63091233.0,
      "step": 34195
    },
    {
      "entropy": 5.535387992858887,
      "epoch": 2.8732619197647553,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041771128712165997,
      "loss": 5.1588,
      "mean_token_accuracy": 0.18207337111234664,
      "num_tokens": 63101083.0,
      "step": 34200
    },
    {
      "entropy": 5.487075090408325,
      "epoch": 2.87368199957992,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004176881338913845,
      "loss": 5.1365,
      "mean_token_accuracy": 0.19463736265897752,
      "num_tokens": 63109956.0,
      "step": 34205
    },
    {
      "entropy": 5.482854986190796,
      "epoch": 2.874102079395085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041766497813346235,
      "loss": 5.1181,
      "mean_token_accuracy": 0.19050100743770598,
      "num_tokens": 63119357.0,
      "step": 34210
    },
    {
      "entropy": 5.620587539672852,
      "epoch": 2.87452215921025,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041764181984830346,
      "loss": 5.2193,
      "mean_token_accuracy": 0.18207336515188216,
      "num_tokens": 63128963.0,
      "step": 34215
    },
    {
      "entropy": 5.517769479751587,
      "epoch": 2.874942239025415,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041761865903631817,
      "loss": 5.1164,
      "mean_token_accuracy": 0.19392411708831786,
      "num_tokens": 63137817.0,
      "step": 34220
    },
    {
      "entropy": 5.386983060836792,
      "epoch": 2.8753623188405797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041759549569791684,
      "loss": 4.9405,
      "mean_token_accuracy": 0.2076481729745865,
      "num_tokens": 63146191.0,
      "step": 34225
    },
    {
      "entropy": 5.381958913803101,
      "epoch": 2.8757823986557445,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004175723298335097,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20345861911773683,
      "num_tokens": 63154834.0,
      "step": 34230
    },
    {
      "entropy": 5.331991481781006,
      "epoch": 2.8762024784709093,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004175491614435071,
      "loss": 4.9475,
      "mean_token_accuracy": 0.20790566951036454,
      "num_tokens": 63163388.0,
      "step": 34235
    },
    {
      "entropy": 5.490551710128784,
      "epoch": 2.8766225582860745,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041752599052831955,
      "loss": 5.1709,
      "mean_token_accuracy": 0.18775008469820023,
      "num_tokens": 63172749.0,
      "step": 34240
    },
    {
      "entropy": 5.547119617462158,
      "epoch": 2.8770426381012393,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041750281708835747,
      "loss": 5.139,
      "mean_token_accuracy": 0.1916719913482666,
      "num_tokens": 63182823.0,
      "step": 34245
    },
    {
      "entropy": 5.494550800323486,
      "epoch": 2.877462717916404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041747964112403137,
      "loss": 5.0847,
      "mean_token_accuracy": 0.19830551296472548,
      "num_tokens": 63191770.0,
      "step": 34250
    },
    {
      "entropy": 5.442732620239258,
      "epoch": 2.877882797731569,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041745646263575183,
      "loss": 5.1106,
      "mean_token_accuracy": 0.18805919736623763,
      "num_tokens": 63201343.0,
      "step": 34255
    },
    {
      "entropy": 5.455474424362182,
      "epoch": 2.8783028775467336,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041743328162392935,
      "loss": 5.0727,
      "mean_token_accuracy": 0.19564990252256392,
      "num_tokens": 63209673.0,
      "step": 34260
    },
    {
      "entropy": 5.530594444274902,
      "epoch": 2.878722957361899,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041741009808897455,
      "loss": 5.1599,
      "mean_token_accuracy": 0.19226163178682326,
      "num_tokens": 63218873.0,
      "step": 34265
    },
    {
      "entropy": 5.513440322875977,
      "epoch": 2.8791430371770637,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041738691203129836,
      "loss": 5.1403,
      "mean_token_accuracy": 0.1878921642899513,
      "num_tokens": 63228100.0,
      "step": 34270
    },
    {
      "entropy": 5.538701152801513,
      "epoch": 2.8795631169922284,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041736372345131125,
      "loss": 5.1277,
      "mean_token_accuracy": 0.19573871791362762,
      "num_tokens": 63237036.0,
      "step": 34275
    },
    {
      "entropy": 5.4435090065002445,
      "epoch": 2.8799831968073937,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004173405323494241,
      "loss": 5.0271,
      "mean_token_accuracy": 0.19678186029195785,
      "num_tokens": 63246982.0,
      "step": 34280
    },
    {
      "entropy": 5.645844411849976,
      "epoch": 2.880403276622558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004173173387260477,
      "loss": 5.2334,
      "mean_token_accuracy": 0.18543805032968522,
      "num_tokens": 63256380.0,
      "step": 34285
    },
    {
      "entropy": 5.515723466873169,
      "epoch": 2.8808233564377232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000417294142581593,
      "loss": 5.0153,
      "mean_token_accuracy": 0.2028910294175148,
      "num_tokens": 63265098.0,
      "step": 34290
    },
    {
      "entropy": 5.403167390823365,
      "epoch": 2.881243436252888,
      "grad_norm": 1.125,
      "learning_rate": 0.0004172709439164707,
      "loss": 5.1095,
      "mean_token_accuracy": 0.19849657416343688,
      "num_tokens": 63274083.0,
      "step": 34295
    },
    {
      "entropy": 5.475836563110351,
      "epoch": 2.881663516068053,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041724774273109195,
      "loss": 5.0386,
      "mean_token_accuracy": 0.19537715315818788,
      "num_tokens": 63283387.0,
      "step": 34300
    },
    {
      "entropy": 5.643346500396729,
      "epoch": 2.882083595883218,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004172245390258677,
      "loss": 5.2432,
      "mean_token_accuracy": 0.1836852416396141,
      "num_tokens": 63293583.0,
      "step": 34305
    },
    {
      "entropy": 5.55661096572876,
      "epoch": 2.882503675698383,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041720133280120903,
      "loss": 5.1694,
      "mean_token_accuracy": 0.19260716885328294,
      "num_tokens": 63303274.0,
      "step": 34310
    },
    {
      "entropy": 5.532939624786377,
      "epoch": 2.8829237555135476,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041717812405752693,
      "loss": 5.2045,
      "mean_token_accuracy": 0.1886974960565567,
      "num_tokens": 63312299.0,
      "step": 34315
    },
    {
      "entropy": 5.474243783950806,
      "epoch": 2.8833438353287124,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041715491279523256,
      "loss": 5.0968,
      "mean_token_accuracy": 0.19442351907491684,
      "num_tokens": 63321616.0,
      "step": 34320
    },
    {
      "entropy": 5.541963577270508,
      "epoch": 2.883763915143877,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004171316990147371,
      "loss": 5.1179,
      "mean_token_accuracy": 0.19765319526195527,
      "num_tokens": 63330777.0,
      "step": 34325
    },
    {
      "entropy": 5.528542232513428,
      "epoch": 2.8841839949590424,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041710848271645174,
      "loss": 5.1047,
      "mean_token_accuracy": 0.1912448912858963,
      "num_tokens": 63340852.0,
      "step": 34330
    },
    {
      "entropy": 5.49384708404541,
      "epoch": 2.884604074774207,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004170852639007878,
      "loss": 5.1736,
      "mean_token_accuracy": 0.18966876715421677,
      "num_tokens": 63349572.0,
      "step": 34335
    },
    {
      "entropy": 5.407116413116455,
      "epoch": 2.885024154589372,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004170620425681566,
      "loss": 4.9894,
      "mean_token_accuracy": 0.20322617441415786,
      "num_tokens": 63358250.0,
      "step": 34340
    },
    {
      "entropy": 5.469576072692871,
      "epoch": 2.885444234404537,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041703881871896947,
      "loss": 5.1013,
      "mean_token_accuracy": 0.1944442003965378,
      "num_tokens": 63366656.0,
      "step": 34345
    },
    {
      "entropy": 5.56414794921875,
      "epoch": 2.8858643142197016,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041701559235363784,
      "loss": 5.2949,
      "mean_token_accuracy": 0.18347270637750626,
      "num_tokens": 63376455.0,
      "step": 34350
    },
    {
      "entropy": 5.6064300537109375,
      "epoch": 2.886284394034867,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041699236347257315,
      "loss": 5.1134,
      "mean_token_accuracy": 0.19020037204027176,
      "num_tokens": 63385333.0,
      "step": 34355
    },
    {
      "entropy": 5.602957487106323,
      "epoch": 2.8867044738500316,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041696913207618674,
      "loss": 5.2235,
      "mean_token_accuracy": 0.18512357622385026,
      "num_tokens": 63396174.0,
      "step": 34360
    },
    {
      "entropy": 5.483866357803345,
      "epoch": 2.8871245536651964,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041694589816489034,
      "loss": 5.0116,
      "mean_token_accuracy": 0.20068084597587585,
      "num_tokens": 63404441.0,
      "step": 34365
    },
    {
      "entropy": 5.420630598068238,
      "epoch": 2.887544633480361,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041692266173909545,
      "loss": 5.1017,
      "mean_token_accuracy": 0.18708769828081132,
      "num_tokens": 63412884.0,
      "step": 34370
    },
    {
      "entropy": 5.476957321166992,
      "epoch": 2.887964713295526,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004168994227992136,
      "loss": 5.0556,
      "mean_token_accuracy": 0.19477604180574418,
      "num_tokens": 63422666.0,
      "step": 34375
    },
    {
      "entropy": 5.610987663269043,
      "epoch": 2.888384793110691,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004168761813456566,
      "loss": 5.1907,
      "mean_token_accuracy": 0.190963414311409,
      "num_tokens": 63432235.0,
      "step": 34380
    },
    {
      "entropy": 5.5044385433197025,
      "epoch": 2.888804872925856,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004168529373788361,
      "loss": 5.1056,
      "mean_token_accuracy": 0.19983265399932862,
      "num_tokens": 63441569.0,
      "step": 34385
    },
    {
      "entropy": 5.511942911148071,
      "epoch": 2.8892249527410208,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041682969089916385,
      "loss": 5.2084,
      "mean_token_accuracy": 0.1799772560596466,
      "num_tokens": 63451500.0,
      "step": 34390
    },
    {
      "entropy": 5.602361249923706,
      "epoch": 2.8896450325561855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004168064419070516,
      "loss": 5.2465,
      "mean_token_accuracy": 0.18117722868919373,
      "num_tokens": 63461384.0,
      "step": 34395
    },
    {
      "entropy": 5.57672610282898,
      "epoch": 2.8900651123713503,
      "grad_norm": 0.96875,
      "learning_rate": 0.00041678319040291137,
      "loss": 5.1971,
      "mean_token_accuracy": 0.18755096793174744,
      "num_tokens": 63471835.0,
      "step": 34400
    },
    {
      "entropy": 5.5090290069580075,
      "epoch": 2.8904851921865156,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004167599363871549,
      "loss": 5.1708,
      "mean_token_accuracy": 0.1860038086771965,
      "num_tokens": 63482066.0,
      "step": 34405
    },
    {
      "entropy": 5.537455606460571,
      "epoch": 2.8909052720016803,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004167366798601941,
      "loss": 5.1628,
      "mean_token_accuracy": 0.18771634846925736,
      "num_tokens": 63491554.0,
      "step": 34410
    },
    {
      "entropy": 5.531447696685791,
      "epoch": 2.891325351816845,
      "grad_norm": 1.15625,
      "learning_rate": 0.000416713420822441,
      "loss": 5.1951,
      "mean_token_accuracy": 0.19255711734294892,
      "num_tokens": 63500410.0,
      "step": 34415
    },
    {
      "entropy": 5.431568670272827,
      "epoch": 2.89174543163201,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041669015927430766,
      "loss": 5.0242,
      "mean_token_accuracy": 0.19689542055130005,
      "num_tokens": 63508831.0,
      "step": 34420
    },
    {
      "entropy": 5.551270389556885,
      "epoch": 2.8921655114471747,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004166668952162061,
      "loss": 5.1351,
      "mean_token_accuracy": 0.19440643340349198,
      "num_tokens": 63517696.0,
      "step": 34425
    },
    {
      "entropy": 5.4946410179138185,
      "epoch": 2.89258559126234,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004166436286485484,
      "loss": 5.1665,
      "mean_token_accuracy": 0.1872995302081108,
      "num_tokens": 63527071.0,
      "step": 34430
    },
    {
      "entropy": 5.512196445465088,
      "epoch": 2.8930056710775047,
      "grad_norm": 1.125,
      "learning_rate": 0.0004166203595717468,
      "loss": 5.1901,
      "mean_token_accuracy": 0.18700876832008362,
      "num_tokens": 63536458.0,
      "step": 34435
    },
    {
      "entropy": 5.493632650375366,
      "epoch": 2.8934257508926695,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004165970879862134,
      "loss": 5.0556,
      "mean_token_accuracy": 0.20096144825220108,
      "num_tokens": 63545718.0,
      "step": 34440
    },
    {
      "entropy": 5.587862920761109,
      "epoch": 2.8938458307078347,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004165738138923607,
      "loss": 5.1908,
      "mean_token_accuracy": 0.18798945844173431,
      "num_tokens": 63554416.0,
      "step": 34445
    },
    {
      "entropy": 5.494671726226807,
      "epoch": 2.8942659105229995,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004165505372906005,
      "loss": 5.1556,
      "mean_token_accuracy": 0.1857709676027298,
      "num_tokens": 63563448.0,
      "step": 34450
    },
    {
      "entropy": 5.509906864166259,
      "epoch": 2.8946859903381643,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041652725818134553,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19095098972320557,
      "num_tokens": 63572101.0,
      "step": 34455
    },
    {
      "entropy": 5.543903493881226,
      "epoch": 2.895106070153329,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004165039765650081,
      "loss": 5.1804,
      "mean_token_accuracy": 0.1950398027896881,
      "num_tokens": 63581122.0,
      "step": 34460
    },
    {
      "entropy": 5.555679082870483,
      "epoch": 2.895526149968494,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041648069244200055,
      "loss": 5.1402,
      "mean_token_accuracy": 0.1910608395934105,
      "num_tokens": 63590360.0,
      "step": 34465
    },
    {
      "entropy": 5.520509099960327,
      "epoch": 2.895946229783659,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041645740581273537,
      "loss": 5.024,
      "mean_token_accuracy": 0.19982933700084687,
      "num_tokens": 63599412.0,
      "step": 34470
    },
    {
      "entropy": 5.4539696216583256,
      "epoch": 2.896366309598824,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004164341166776251,
      "loss": 5.1337,
      "mean_token_accuracy": 0.1882551997900009,
      "num_tokens": 63609117.0,
      "step": 34475
    },
    {
      "entropy": 5.495347499847412,
      "epoch": 2.8967863894139887,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004164108250370823,
      "loss": 5.0744,
      "mean_token_accuracy": 0.1916407436132431,
      "num_tokens": 63618841.0,
      "step": 34480
    },
    {
      "entropy": 5.491547536849976,
      "epoch": 2.8972064692291535,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041638753089151954,
      "loss": 5.1586,
      "mean_token_accuracy": 0.186068195104599,
      "num_tokens": 63628152.0,
      "step": 34485
    },
    {
      "entropy": 5.5734625339508055,
      "epoch": 2.8976265490443183,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041636423424134953,
      "loss": 5.1496,
      "mean_token_accuracy": 0.18759823441505433,
      "num_tokens": 63637016.0,
      "step": 34490
    },
    {
      "entropy": 5.52823224067688,
      "epoch": 2.8980466288594835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041634093508698486,
      "loss": 5.1215,
      "mean_token_accuracy": 0.19227929413318634,
      "num_tokens": 63646842.0,
      "step": 34495
    },
    {
      "entropy": 5.533495664596558,
      "epoch": 2.8984667086746483,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004163176334288383,
      "loss": 5.1789,
      "mean_token_accuracy": 0.19071363359689714,
      "num_tokens": 63656229.0,
      "step": 34500
    },
    {
      "entropy": 5.507710361480713,
      "epoch": 2.898886788489813,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004162943292673226,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1978663131594658,
      "num_tokens": 63665149.0,
      "step": 34505
    },
    {
      "entropy": 5.520256805419922,
      "epoch": 2.899306868304978,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041627102260285065,
      "loss": 5.111,
      "mean_token_accuracy": 0.19047555029392244,
      "num_tokens": 63673896.0,
      "step": 34510
    },
    {
      "entropy": 5.417391967773438,
      "epoch": 2.8997269481201426,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041624771343583516,
      "loss": 4.9914,
      "mean_token_accuracy": 0.20021887123584747,
      "num_tokens": 63683219.0,
      "step": 34515
    },
    {
      "entropy": 5.478581333160401,
      "epoch": 2.900147027935308,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041622440176668934,
      "loss": 5.0703,
      "mean_token_accuracy": 0.1924726977944374,
      "num_tokens": 63692480.0,
      "step": 34520
    },
    {
      "entropy": 5.498868322372436,
      "epoch": 2.9005671077504727,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004162010875958259,
      "loss": 5.1354,
      "mean_token_accuracy": 0.1951663762331009,
      "num_tokens": 63700935.0,
      "step": 34525
    },
    {
      "entropy": 5.483849287033081,
      "epoch": 2.9009871875656374,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041617777092365786,
      "loss": 5.0668,
      "mean_token_accuracy": 0.20187804698944092,
      "num_tokens": 63709971.0,
      "step": 34530
    },
    {
      "entropy": 5.487220907211304,
      "epoch": 2.9014072673808022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004161544517505983,
      "loss": 5.1414,
      "mean_token_accuracy": 0.1879653036594391,
      "num_tokens": 63719340.0,
      "step": 34535
    },
    {
      "entropy": 5.452374744415283,
      "epoch": 2.901827347195967,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004161311300770603,
      "loss": 5.1111,
      "mean_token_accuracy": 0.19203499257564544,
      "num_tokens": 63728099.0,
      "step": 34540
    },
    {
      "entropy": 5.4804603099823,
      "epoch": 2.9022474270111323,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004161078059034572,
      "loss": 5.0582,
      "mean_token_accuracy": 0.19829224944114685,
      "num_tokens": 63737507.0,
      "step": 34545
    },
    {
      "entropy": 5.449522924423218,
      "epoch": 2.902667506826297,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041608447923020174,
      "loss": 5.0165,
      "mean_token_accuracy": 0.20455281883478166,
      "num_tokens": 63747007.0,
      "step": 34550
    },
    {
      "entropy": 5.467708492279053,
      "epoch": 2.903087586641462,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041606115005770743,
      "loss": 5.0775,
      "mean_token_accuracy": 0.19577114284038544,
      "num_tokens": 63755667.0,
      "step": 34555
    },
    {
      "entropy": 5.422799730300904,
      "epoch": 2.9035076664566266,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041603781838638754,
      "loss": 5.0673,
      "mean_token_accuracy": 0.19812458753585815,
      "num_tokens": 63764427.0,
      "step": 34560
    },
    {
      "entropy": 5.586203908920288,
      "epoch": 2.9039277462717914,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041601448421665536,
      "loss": 5.2388,
      "mean_token_accuracy": 0.18700022250413895,
      "num_tokens": 63775223.0,
      "step": 34565
    },
    {
      "entropy": 5.576774406433105,
      "epoch": 2.9043478260869566,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004159911475489241,
      "loss": 5.1488,
      "mean_token_accuracy": 0.19507619738578796,
      "num_tokens": 63784366.0,
      "step": 34570
    },
    {
      "entropy": 5.5482066631317135,
      "epoch": 2.9047679059021214,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041596780838360734,
      "loss": 5.1074,
      "mean_token_accuracy": 0.1955839902162552,
      "num_tokens": 63793175.0,
      "step": 34575
    },
    {
      "entropy": 5.506938552856445,
      "epoch": 2.905187985717286,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041594446672111836,
      "loss": 5.1107,
      "mean_token_accuracy": 0.1873517394065857,
      "num_tokens": 63802425.0,
      "step": 34580
    },
    {
      "entropy": 5.545834493637085,
      "epoch": 2.9056080655324514,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041592112256187073,
      "loss": 5.1257,
      "mean_token_accuracy": 0.19304557740688325,
      "num_tokens": 63811665.0,
      "step": 34585
    },
    {
      "entropy": 5.518129110336304,
      "epoch": 2.9060281453476158,
      "grad_norm": 1.078125,
      "learning_rate": 0.000415897775906278,
      "loss": 5.1736,
      "mean_token_accuracy": 0.20014474540948868,
      "num_tokens": 63821684.0,
      "step": 34590
    },
    {
      "entropy": 5.511342334747314,
      "epoch": 2.906448225162781,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041587442675475376,
      "loss": 5.1186,
      "mean_token_accuracy": 0.19491571485996245,
      "num_tokens": 63830641.0,
      "step": 34595
    },
    {
      "entropy": 5.536042404174805,
      "epoch": 2.906868304977946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041585107510771157,
      "loss": 5.044,
      "mean_token_accuracy": 0.19754016697406768,
      "num_tokens": 63838761.0,
      "step": 34600
    },
    {
      "entropy": 5.441779518127442,
      "epoch": 2.9072883847931106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004158277209655651,
      "loss": 5.0695,
      "mean_token_accuracy": 0.20062098950147628,
      "num_tokens": 63847140.0,
      "step": 34605
    },
    {
      "entropy": 5.499097728729248,
      "epoch": 2.907708464608276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004158043643287281,
      "loss": 5.1943,
      "mean_token_accuracy": 0.18884539753198623,
      "num_tokens": 63857598.0,
      "step": 34610
    },
    {
      "entropy": 5.548840284347534,
      "epoch": 2.9081285444234406,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041578100519761425,
      "loss": 5.0249,
      "mean_token_accuracy": 0.1990709960460663,
      "num_tokens": 63866382.0,
      "step": 34615
    },
    {
      "entropy": 5.450946092605591,
      "epoch": 2.9085486242386054,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004157576435726374,
      "loss": 5.0299,
      "mean_token_accuracy": 0.20373745262622833,
      "num_tokens": 63876047.0,
      "step": 34620
    },
    {
      "entropy": 5.439879941940307,
      "epoch": 2.90896870405377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041573427945421136,
      "loss": 5.0868,
      "mean_token_accuracy": 0.20099395960569383,
      "num_tokens": 63885867.0,
      "step": 34625
    },
    {
      "entropy": 5.508423328399658,
      "epoch": 2.909388783868935,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041571091284275003,
      "loss": 5.1283,
      "mean_token_accuracy": 0.1926201969385147,
      "num_tokens": 63895087.0,
      "step": 34630
    },
    {
      "entropy": 5.4401562213897705,
      "epoch": 2.9098088636841,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004156875437386674,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20549572259187698,
      "num_tokens": 63904006.0,
      "step": 34635
    },
    {
      "entropy": 5.516796493530274,
      "epoch": 2.910228943499265,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041566417214237733,
      "loss": 5.1783,
      "mean_token_accuracy": 0.18749792575836183,
      "num_tokens": 63913180.0,
      "step": 34640
    },
    {
      "entropy": 5.541478538513184,
      "epoch": 2.9106490233144298,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004156407980542938,
      "loss": 5.1969,
      "mean_token_accuracy": 0.19307056069374084,
      "num_tokens": 63922966.0,
      "step": 34645
    },
    {
      "entropy": 5.6027990818023685,
      "epoch": 2.9110691031295945,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004156174214748311,
      "loss": 5.2164,
      "mean_token_accuracy": 0.18287577480077744,
      "num_tokens": 63933069.0,
      "step": 34650
    },
    {
      "entropy": 5.528793096542358,
      "epoch": 2.9114891829447593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041559404240440305,
      "loss": 5.1429,
      "mean_token_accuracy": 0.1906629666686058,
      "num_tokens": 63942485.0,
      "step": 34655
    },
    {
      "entropy": 5.615823078155517,
      "epoch": 2.9119092627599246,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000415570660843424,
      "loss": 5.3113,
      "mean_token_accuracy": 0.17611884772777558,
      "num_tokens": 63953651.0,
      "step": 34660
    },
    {
      "entropy": 5.542328310012818,
      "epoch": 2.9123293425750894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041554727679230815,
      "loss": 5.1168,
      "mean_token_accuracy": 0.20198774039745332,
      "num_tokens": 63963694.0,
      "step": 34665
    },
    {
      "entropy": 5.412470531463623,
      "epoch": 2.912749422390254,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041552389025146955,
      "loss": 5.0404,
      "mean_token_accuracy": 0.19987600296735764,
      "num_tokens": 63972897.0,
      "step": 34670
    },
    {
      "entropy": 5.580990362167358,
      "epoch": 2.913169502205419,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041550050122132266,
      "loss": 5.2048,
      "mean_token_accuracy": 0.18558869063854216,
      "num_tokens": 63981977.0,
      "step": 34675
    },
    {
      "entropy": 5.517338228225708,
      "epoch": 2.9135895820205837,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041547710970228164,
      "loss": 5.1276,
      "mean_token_accuracy": 0.18696735799312592,
      "num_tokens": 63991713.0,
      "step": 34680
    },
    {
      "entropy": 5.535034084320069,
      "epoch": 2.914009661835749,
      "grad_norm": 1.0,
      "learning_rate": 0.00041545371569476103,
      "loss": 5.1822,
      "mean_token_accuracy": 0.1889250561594963,
      "num_tokens": 64000763.0,
      "step": 34685
    },
    {
      "entropy": 5.37685604095459,
      "epoch": 2.9144297416509137,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004154303191991751,
      "loss": 4.936,
      "mean_token_accuracy": 0.20909763723611832,
      "num_tokens": 64009889.0,
      "step": 34690
    },
    {
      "entropy": 5.378986549377442,
      "epoch": 2.9148498214660785,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004154069202159385,
      "loss": 5.0153,
      "mean_token_accuracy": 0.19716450273990632,
      "num_tokens": 64019441.0,
      "step": 34695
    },
    {
      "entropy": 5.403966665267944,
      "epoch": 2.9152699012812433,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004153835187454655,
      "loss": 5.0344,
      "mean_token_accuracy": 0.20090289711952208,
      "num_tokens": 64028953.0,
      "step": 34700
    },
    {
      "entropy": 5.53577675819397,
      "epoch": 2.915689981096408,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004153601147881708,
      "loss": 5.1208,
      "mean_token_accuracy": 0.1954495295882225,
      "num_tokens": 64038492.0,
      "step": 34705
    },
    {
      "entropy": 5.575509738922119,
      "epoch": 2.9161100609115733,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000415336708344469,
      "loss": 5.1782,
      "mean_token_accuracy": 0.19250085949897766,
      "num_tokens": 64046949.0,
      "step": 34710
    },
    {
      "entropy": 5.437145137786866,
      "epoch": 2.916530140726738,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004153132994147746,
      "loss": 4.95,
      "mean_token_accuracy": 0.2037343293428421,
      "num_tokens": 64055628.0,
      "step": 34715
    },
    {
      "entropy": 5.459938383102417,
      "epoch": 2.916950220541903,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041528988799950233,
      "loss": 5.0708,
      "mean_token_accuracy": 0.2033136159181595,
      "num_tokens": 64064625.0,
      "step": 34720
    },
    {
      "entropy": 5.529673480987549,
      "epoch": 2.9173703003570677,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000415266474099067,
      "loss": 5.1903,
      "mean_token_accuracy": 0.19473304748535156,
      "num_tokens": 64074870.0,
      "step": 34725
    },
    {
      "entropy": 5.433637475967407,
      "epoch": 2.9177903801722325,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041524305771388324,
      "loss": 4.992,
      "mean_token_accuracy": 0.19591658115386962,
      "num_tokens": 64083303.0,
      "step": 34730
    },
    {
      "entropy": 5.472391796112061,
      "epoch": 2.9182104599873977,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041521963884436595,
      "loss": 5.0482,
      "mean_token_accuracy": 0.1971310630440712,
      "num_tokens": 64092062.0,
      "step": 34735
    },
    {
      "entropy": 5.643289089202881,
      "epoch": 2.9186305398025625,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041519621749093,
      "loss": 5.3379,
      "mean_token_accuracy": 0.18304308205842973,
      "num_tokens": 64102340.0,
      "step": 34740
    },
    {
      "entropy": 5.42947359085083,
      "epoch": 2.9190506196177273,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004151727936539902,
      "loss": 5.0514,
      "mean_token_accuracy": 0.19897307008504866,
      "num_tokens": 64111678.0,
      "step": 34745
    },
    {
      "entropy": 5.5118323802948,
      "epoch": 2.9194706994328925,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004151493673339615,
      "loss": 5.1,
      "mean_token_accuracy": 0.19228249341249465,
      "num_tokens": 64121650.0,
      "step": 34750
    },
    {
      "entropy": 5.478714656829834,
      "epoch": 2.9198907792480573,
      "grad_norm": 1.171875,
      "learning_rate": 0.000415125938531259,
      "loss": 5.0825,
      "mean_token_accuracy": 0.20102402120828627,
      "num_tokens": 64130443.0,
      "step": 34755
    },
    {
      "entropy": 5.463911437988282,
      "epoch": 2.920310859063222,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004151025072462977,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19768149703741072,
      "num_tokens": 64139222.0,
      "step": 34760
    },
    {
      "entropy": 5.5571917533874515,
      "epoch": 2.920730938878387,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004150790734794925,
      "loss": 5.1981,
      "mean_token_accuracy": 0.1906736671924591,
      "num_tokens": 64148646.0,
      "step": 34765
    },
    {
      "entropy": 5.425016450881958,
      "epoch": 2.9211510186935516,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004150556372312587,
      "loss": 5.0797,
      "mean_token_accuracy": 0.19127792567014695,
      "num_tokens": 64157620.0,
      "step": 34770
    },
    {
      "entropy": 5.466651344299317,
      "epoch": 2.921571098508717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041503219850201145,
      "loss": 5.0385,
      "mean_token_accuracy": 0.18905141353607177,
      "num_tokens": 64166624.0,
      "step": 34775
    },
    {
      "entropy": 5.522683954238891,
      "epoch": 2.9219911783238817,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041500875729216585,
      "loss": 5.0558,
      "mean_token_accuracy": 0.19656521677970887,
      "num_tokens": 64174881.0,
      "step": 34780
    },
    {
      "entropy": 5.560286951065064,
      "epoch": 2.9224112581390465,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041498531360213724,
      "loss": 5.1593,
      "mean_token_accuracy": 0.1897697255015373,
      "num_tokens": 64184269.0,
      "step": 34785
    },
    {
      "entropy": 5.425645446777343,
      "epoch": 2.9228313379542112,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041496186743234095,
      "loss": 4.9973,
      "mean_token_accuracy": 0.20241705030202867,
      "num_tokens": 64192744.0,
      "step": 34790
    },
    {
      "entropy": 5.468153047561645,
      "epoch": 2.923251417769376,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004149384187831921,
      "loss": 5.1466,
      "mean_token_accuracy": 0.18848497420549393,
      "num_tokens": 64201367.0,
      "step": 34795
    },
    {
      "entropy": 5.570171976089478,
      "epoch": 2.9236714975845413,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004149149676551063,
      "loss": 5.2185,
      "mean_token_accuracy": 0.18719493448734284,
      "num_tokens": 64210354.0,
      "step": 34800
    },
    {
      "entropy": 5.606080198287964,
      "epoch": 2.924091577399706,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041489151404849883,
      "loss": 5.0996,
      "mean_token_accuracy": 0.1977371633052826,
      "num_tokens": 64219902.0,
      "step": 34805
    },
    {
      "entropy": 5.581861543655395,
      "epoch": 2.924511657214871,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041486805796378523,
      "loss": 5.1865,
      "mean_token_accuracy": 0.18955160677433014,
      "num_tokens": 64228952.0,
      "step": 34810
    },
    {
      "entropy": 5.537541055679322,
      "epoch": 2.9249317370300356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041484459940138116,
      "loss": 5.1483,
      "mean_token_accuracy": 0.19227136075496673,
      "num_tokens": 64238515.0,
      "step": 34815
    },
    {
      "entropy": 5.521598386764526,
      "epoch": 2.9253518168452004,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004148211383617018,
      "loss": 5.1197,
      "mean_token_accuracy": 0.19613443464040756,
      "num_tokens": 64248141.0,
      "step": 34820
    },
    {
      "entropy": 5.5121019840240475,
      "epoch": 2.9257718966603656,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004147976748451631,
      "loss": 5.0835,
      "mean_token_accuracy": 0.19514672458171844,
      "num_tokens": 64257376.0,
      "step": 34825
    },
    {
      "entropy": 5.457294273376465,
      "epoch": 2.9261919764755304,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004147742088521806,
      "loss": 5.0823,
      "mean_token_accuracy": 0.19946516156196595,
      "num_tokens": 64266848.0,
      "step": 34830
    },
    {
      "entropy": 5.479612731933594,
      "epoch": 2.926612056290695,
      "grad_norm": 1.125,
      "learning_rate": 0.0004147507403831699,
      "loss": 5.0706,
      "mean_token_accuracy": 0.1958560526371002,
      "num_tokens": 64276363.0,
      "step": 34835
    },
    {
      "entropy": 5.586691999435425,
      "epoch": 2.92703213610586,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041472726943854677,
      "loss": 5.214,
      "mean_token_accuracy": 0.18790175020694733,
      "num_tokens": 64286173.0,
      "step": 34840
    },
    {
      "entropy": 5.5091941356658936,
      "epoch": 2.927452215921025,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004147037960187271,
      "loss": 5.0861,
      "mean_token_accuracy": 0.20169589072465896,
      "num_tokens": 64294326.0,
      "step": 34845
    },
    {
      "entropy": 5.398253536224365,
      "epoch": 2.92787229573619,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041468032012412654,
      "loss": 4.9944,
      "mean_token_accuracy": 0.20395357459783553,
      "num_tokens": 64303657.0,
      "step": 34850
    },
    {
      "entropy": 5.491083860397339,
      "epoch": 2.928292375551355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000414656841755161,
      "loss": 5.083,
      "mean_token_accuracy": 0.1932200938463211,
      "num_tokens": 64312445.0,
      "step": 34855
    },
    {
      "entropy": 5.46591625213623,
      "epoch": 2.9287124553665196,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041463336091224657,
      "loss": 5.0168,
      "mean_token_accuracy": 0.19995124638080597,
      "num_tokens": 64321238.0,
      "step": 34860
    },
    {
      "entropy": 5.399024152755738,
      "epoch": 2.9291325351816844,
      "grad_norm": 1.15625,
      "learning_rate": 0.000414609877595799,
      "loss": 4.9711,
      "mean_token_accuracy": 0.20996550917625428,
      "num_tokens": 64330830.0,
      "step": 34865
    },
    {
      "entropy": 5.510783863067627,
      "epoch": 2.929552614996849,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041458639180623426,
      "loss": 5.2221,
      "mean_token_accuracy": 0.18319059163331985,
      "num_tokens": 64339944.0,
      "step": 34870
    },
    {
      "entropy": 5.569463920593262,
      "epoch": 2.9299726948120144,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004145629035439685,
      "loss": 5.0556,
      "mean_token_accuracy": 0.20398827344179155,
      "num_tokens": 64349417.0,
      "step": 34875
    },
    {
      "entropy": 5.537578439712524,
      "epoch": 2.930392774627179,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041453941280941774,
      "loss": 5.1375,
      "mean_token_accuracy": 0.19020186811685563,
      "num_tokens": 64358728.0,
      "step": 34880
    },
    {
      "entropy": 5.502933931350708,
      "epoch": 2.930812854442344,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00041451591960299815,
      "loss": 5.1624,
      "mean_token_accuracy": 0.19230347722768784,
      "num_tokens": 64368999.0,
      "step": 34885
    },
    {
      "entropy": 5.574857616424561,
      "epoch": 2.931232934257509,
      "grad_norm": 1.125,
      "learning_rate": 0.000414492423925126,
      "loss": 5.2165,
      "mean_token_accuracy": 0.18874310255050658,
      "num_tokens": 64380079.0,
      "step": 34890
    },
    {
      "entropy": 5.572251415252685,
      "epoch": 2.9316530140726735,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004144689257762173,
      "loss": 5.1199,
      "mean_token_accuracy": 0.192536561191082,
      "num_tokens": 64389980.0,
      "step": 34895
    },
    {
      "entropy": 5.428394603729248,
      "epoch": 2.9320730938878388,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041444542515668833,
      "loss": 5.0409,
      "mean_token_accuracy": 0.1970046728849411,
      "num_tokens": 64398852.0,
      "step": 34900
    },
    {
      "entropy": 5.454108953475952,
      "epoch": 2.9324931737030036,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004144219220669555,
      "loss": 5.0634,
      "mean_token_accuracy": 0.1956365242600441,
      "num_tokens": 64407857.0,
      "step": 34905
    },
    {
      "entropy": 5.44381046295166,
      "epoch": 2.9329132535181683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041439841650743516,
      "loss": 5.043,
      "mean_token_accuracy": 0.199790795147419,
      "num_tokens": 64417072.0,
      "step": 34910
    },
    {
      "entropy": 5.45609769821167,
      "epoch": 2.9333333333333336,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004143749084785436,
      "loss": 5.0771,
      "mean_token_accuracy": 0.1966608852148056,
      "num_tokens": 64426145.0,
      "step": 34915
    },
    {
      "entropy": 5.478054523468018,
      "epoch": 2.9337534131484984,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041435139798069737,
      "loss": 5.1009,
      "mean_token_accuracy": 0.19275179356336594,
      "num_tokens": 64435672.0,
      "step": 34920
    },
    {
      "entropy": 5.513485288619995,
      "epoch": 2.934173492963663,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041432788501431284,
      "loss": 5.1103,
      "mean_token_accuracy": 0.19523230493068694,
      "num_tokens": 64445401.0,
      "step": 34925
    },
    {
      "entropy": 5.4921605587005615,
      "epoch": 2.934593572778828,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004143043695798065,
      "loss": 5.143,
      "mean_token_accuracy": 0.19121236652135848,
      "num_tokens": 64455652.0,
      "step": 34930
    },
    {
      "entropy": 5.5191809177398685,
      "epoch": 2.9350136525939927,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004142808516775951,
      "loss": 5.1059,
      "mean_token_accuracy": 0.1906184062361717,
      "num_tokens": 64465058.0,
      "step": 34935
    },
    {
      "entropy": 5.485248327255249,
      "epoch": 2.935433732409158,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000414257331308095,
      "loss": 5.134,
      "mean_token_accuracy": 0.19602167010307311,
      "num_tokens": 64474534.0,
      "step": 34940
    },
    {
      "entropy": 5.467251491546631,
      "epoch": 2.9358538122243227,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041423380847172307,
      "loss": 5.1348,
      "mean_token_accuracy": 0.1963595062494278,
      "num_tokens": 64483849.0,
      "step": 34945
    },
    {
      "entropy": 5.549788284301758,
      "epoch": 2.9362738920394875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041421028316889597,
      "loss": 5.1793,
      "mean_token_accuracy": 0.19160843044519424,
      "num_tokens": 64492451.0,
      "step": 34950
    },
    {
      "entropy": 5.545271968841552,
      "epoch": 2.9366939718546523,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004141867554000303,
      "loss": 5.0963,
      "mean_token_accuracy": 0.2005492180585861,
      "num_tokens": 64502592.0,
      "step": 34955
    },
    {
      "entropy": 5.497558116912842,
      "epoch": 2.937114051669817,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041416322516554295,
      "loss": 5.0804,
      "mean_token_accuracy": 0.205788354575634,
      "num_tokens": 64511769.0,
      "step": 34960
    },
    {
      "entropy": 5.531216669082641,
      "epoch": 2.9375341314849823,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004141396924658507,
      "loss": 5.1637,
      "mean_token_accuracy": 0.19192230999469756,
      "num_tokens": 64521303.0,
      "step": 34965
    },
    {
      "entropy": 5.485566806793213,
      "epoch": 2.937954211300147,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004141161573013705,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19733950346708298,
      "num_tokens": 64530629.0,
      "step": 34970
    },
    {
      "entropy": 5.5298104763031,
      "epoch": 2.938374291115312,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041409261967251915,
      "loss": 5.1867,
      "mean_token_accuracy": 0.1920712798833847,
      "num_tokens": 64539350.0,
      "step": 34975
    },
    {
      "entropy": 5.457802248001099,
      "epoch": 2.9387943709304767,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041406907957971373,
      "loss": 5.0248,
      "mean_token_accuracy": 0.1972404330968857,
      "num_tokens": 64547662.0,
      "step": 34980
    },
    {
      "entropy": 5.4968993186950685,
      "epoch": 2.9392144507456415,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004140455370233711,
      "loss": 5.0767,
      "mean_token_accuracy": 0.19130657315254213,
      "num_tokens": 64558075.0,
      "step": 34985
    },
    {
      "entropy": 5.531182479858399,
      "epoch": 2.9396345305608067,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004140219920039085,
      "loss": 5.2031,
      "mean_token_accuracy": 0.1924383357167244,
      "num_tokens": 64567817.0,
      "step": 34990
    },
    {
      "entropy": 5.524786949157715,
      "epoch": 2.9400546103759715,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004139984445217427,
      "loss": 5.1241,
      "mean_token_accuracy": 0.19780758321285247,
      "num_tokens": 64576656.0,
      "step": 34995
    },
    {
      "entropy": 5.475639057159424,
      "epoch": 2.9404746901911363,
      "grad_norm": 1.125,
      "learning_rate": 0.0004139748945772912,
      "loss": 5.0502,
      "mean_token_accuracy": 0.196832475066185,
      "num_tokens": 64585537.0,
      "step": 35000
    },
    {
      "entropy": 5.512558221817017,
      "epoch": 2.940894770006301,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000413951342170971,
      "loss": 5.1366,
      "mean_token_accuracy": 0.19801500737667083,
      "num_tokens": 64595104.0,
      "step": 35005
    },
    {
      "entropy": 5.371467542648316,
      "epoch": 2.941314849821466,
      "grad_norm": 1.125,
      "learning_rate": 0.00041392778730319917,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2090962991118431,
      "num_tokens": 64602917.0,
      "step": 35010
    },
    {
      "entropy": 5.537664461135864,
      "epoch": 2.941734929636631,
      "grad_norm": 1.125,
      "learning_rate": 0.00041390422997439325,
      "loss": 5.1905,
      "mean_token_accuracy": 0.18306822776794435,
      "num_tokens": 64612383.0,
      "step": 35015
    },
    {
      "entropy": 5.553844833374024,
      "epoch": 2.942155009451796,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041388067018497044,
      "loss": 5.0727,
      "mean_token_accuracy": 0.20276571065187454,
      "num_tokens": 64621557.0,
      "step": 35020
    },
    {
      "entropy": 5.468598651885986,
      "epoch": 2.9425750892669607,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041385710793534806,
      "loss": 5.1019,
      "mean_token_accuracy": 0.19183942228555678,
      "num_tokens": 64631513.0,
      "step": 35025
    },
    {
      "entropy": 5.471777105331421,
      "epoch": 2.9429951690821254,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041383354322594346,
      "loss": 5.1319,
      "mean_token_accuracy": 0.19300687611103057,
      "num_tokens": 64641050.0,
      "step": 35030
    },
    {
      "entropy": 5.52652792930603,
      "epoch": 2.9434152488972902,
      "grad_norm": 1.125,
      "learning_rate": 0.00041380997605717417,
      "loss": 5.1374,
      "mean_token_accuracy": 0.19347133338451386,
      "num_tokens": 64650010.0,
      "step": 35035
    },
    {
      "entropy": 5.515239429473877,
      "epoch": 2.9438353287124555,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004137864064294576,
      "loss": 5.0471,
      "mean_token_accuracy": 0.19762482196092607,
      "num_tokens": 64659586.0,
      "step": 35040
    },
    {
      "entropy": 5.468162393569946,
      "epoch": 2.9442554085276202,
      "grad_norm": 1.125,
      "learning_rate": 0.0004137628343432113,
      "loss": 5.0652,
      "mean_token_accuracy": 0.1980970412492752,
      "num_tokens": 64669759.0,
      "step": 35045
    },
    {
      "entropy": 5.556880712509155,
      "epoch": 2.944675488342785,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041373925979885283,
      "loss": 5.2249,
      "mean_token_accuracy": 0.18662554621696473,
      "num_tokens": 64678614.0,
      "step": 35050
    },
    {
      "entropy": 5.485786628723145,
      "epoch": 2.9450955681579503,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004137156827967999,
      "loss": 5.1055,
      "mean_token_accuracy": 0.19952459633350372,
      "num_tokens": 64688253.0,
      "step": 35055
    },
    {
      "entropy": 5.564535093307495,
      "epoch": 2.945515647973115,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041369210333747005,
      "loss": 5.2089,
      "mean_token_accuracy": 0.1910799115896225,
      "num_tokens": 64697755.0,
      "step": 35060
    },
    {
      "entropy": 5.493434286117553,
      "epoch": 2.94593572778828,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004136685214212809,
      "loss": 5.0506,
      "mean_token_accuracy": 0.19564608186483384,
      "num_tokens": 64706668.0,
      "step": 35065
    },
    {
      "entropy": 5.41173243522644,
      "epoch": 2.9463558076034446,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041364493704865044,
      "loss": 5.0889,
      "mean_token_accuracy": 0.20179139375686644,
      "num_tokens": 64715770.0,
      "step": 35070
    },
    {
      "entropy": 5.557145452499389,
      "epoch": 2.9467758874186094,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004136213502199962,
      "loss": 5.1176,
      "mean_token_accuracy": 0.19435902386903764,
      "num_tokens": 64725912.0,
      "step": 35075
    },
    {
      "entropy": 5.530763244628906,
      "epoch": 2.9471959672337746,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041359776093573617,
      "loss": 5.0912,
      "mean_token_accuracy": 0.19258591085672377,
      "num_tokens": 64735567.0,
      "step": 35080
    },
    {
      "entropy": 5.485283088684082,
      "epoch": 2.9476160470489394,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041357416919628816,
      "loss": 5.0766,
      "mean_token_accuracy": 0.1928081914782524,
      "num_tokens": 64743972.0,
      "step": 35085
    },
    {
      "entropy": 5.471222448348999,
      "epoch": 2.948036126864104,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041355057500207004,
      "loss": 5.1066,
      "mean_token_accuracy": 0.19886018931865693,
      "num_tokens": 64753030.0,
      "step": 35090
    },
    {
      "entropy": 5.585369205474853,
      "epoch": 2.948456206679269,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041352697835349994,
      "loss": 5.1776,
      "mean_token_accuracy": 0.18708908706903457,
      "num_tokens": 64762717.0,
      "step": 35095
    },
    {
      "entropy": 5.489133405685425,
      "epoch": 2.948876286494434,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004135033792509957,
      "loss": 5.1177,
      "mean_token_accuracy": 0.19073213189840316,
      "num_tokens": 64771776.0,
      "step": 35100
    },
    {
      "entropy": 5.486712265014648,
      "epoch": 2.949296366309599,
      "grad_norm": 1.125,
      "learning_rate": 0.0004134797776949755,
      "loss": 5.1399,
      "mean_token_accuracy": 0.19518659114837647,
      "num_tokens": 64781418.0,
      "step": 35105
    },
    {
      "entropy": 5.487704753875732,
      "epoch": 2.949716446124764,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004134561736858572,
      "loss": 5.1263,
      "mean_token_accuracy": 0.1980529621243477,
      "num_tokens": 64790343.0,
      "step": 35110
    },
    {
      "entropy": 5.585173892974853,
      "epoch": 2.9501365259399286,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041343256722405915,
      "loss": 5.1296,
      "mean_token_accuracy": 0.19515352696180344,
      "num_tokens": 64799189.0,
      "step": 35115
    },
    {
      "entropy": 5.5763078212738035,
      "epoch": 2.9505566057550934,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004134089583099994,
      "loss": 5.1653,
      "mean_token_accuracy": 0.18971058130264282,
      "num_tokens": 64809276.0,
      "step": 35120
    },
    {
      "entropy": 5.471062850952149,
      "epoch": 2.950976685570258,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004133853469440963,
      "loss": 5.0961,
      "mean_token_accuracy": 0.19479290246963502,
      "num_tokens": 64818008.0,
      "step": 35125
    },
    {
      "entropy": 5.464027357101441,
      "epoch": 2.9513967653854234,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041336173312676795,
      "loss": 4.995,
      "mean_token_accuracy": 0.1920005351305008,
      "num_tokens": 64826576.0,
      "step": 35130
    },
    {
      "entropy": 5.533541917800903,
      "epoch": 2.951816845200588,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004133381168584329,
      "loss": 5.1132,
      "mean_token_accuracy": 0.19731767028570174,
      "num_tokens": 64835510.0,
      "step": 35135
    },
    {
      "entropy": 5.438952875137329,
      "epoch": 2.952236925015753,
      "grad_norm": 1.125,
      "learning_rate": 0.0004133144981395092,
      "loss": 5.0261,
      "mean_token_accuracy": 0.19949463605880738,
      "num_tokens": 64845031.0,
      "step": 35140
    },
    {
      "entropy": 5.459491014480591,
      "epoch": 2.9526570048309178,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004132908769704154,
      "loss": 5.0814,
      "mean_token_accuracy": 0.19393713921308517,
      "num_tokens": 64855365.0,
      "step": 35145
    },
    {
      "entropy": 5.538378953933716,
      "epoch": 2.9530770846460825,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004132672533515699,
      "loss": 5.1118,
      "mean_token_accuracy": 0.19609958827495574,
      "num_tokens": 64864014.0,
      "step": 35150
    },
    {
      "entropy": 5.562719631195068,
      "epoch": 2.9534971644612478,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004132436272833913,
      "loss": 5.1151,
      "mean_token_accuracy": 0.19528750479221343,
      "num_tokens": 64873127.0,
      "step": 35155
    },
    {
      "entropy": 5.475818729400634,
      "epoch": 2.9539172442764126,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041321999876629795,
      "loss": 5.0853,
      "mean_token_accuracy": 0.1944798544049263,
      "num_tokens": 64881549.0,
      "step": 35160
    },
    {
      "entropy": 5.48251314163208,
      "epoch": 2.9543373240915773,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004131963678007085,
      "loss": 5.0855,
      "mean_token_accuracy": 0.20222388058900834,
      "num_tokens": 64890204.0,
      "step": 35165
    },
    {
      "entropy": 5.518037509918213,
      "epoch": 2.954757403906742,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041317273438704164,
      "loss": 5.1361,
      "mean_token_accuracy": 0.18923134952783585,
      "num_tokens": 64899466.0,
      "step": 35170
    },
    {
      "entropy": 5.496178770065308,
      "epoch": 2.955177483721907,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004131490985257159,
      "loss": 5.1154,
      "mean_token_accuracy": 0.1985822930932045,
      "num_tokens": 64908917.0,
      "step": 35175
    },
    {
      "entropy": 5.551247215270996,
      "epoch": 2.955597563537072,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041312546021714995,
      "loss": 5.1529,
      "mean_token_accuracy": 0.18814297616481782,
      "num_tokens": 64918048.0,
      "step": 35180
    },
    {
      "entropy": 5.395475578308106,
      "epoch": 2.956017643352237,
      "grad_norm": 0.984375,
      "learning_rate": 0.00041310181946176263,
      "loss": 4.9143,
      "mean_token_accuracy": 0.20658757537603378,
      "num_tokens": 64926664.0,
      "step": 35185
    },
    {
      "entropy": 5.462780857086182,
      "epoch": 2.9564377231674017,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041307817625997275,
      "loss": 5.0556,
      "mean_token_accuracy": 0.19733170568943023,
      "num_tokens": 64935283.0,
      "step": 35190
    },
    {
      "entropy": 5.517564344406128,
      "epoch": 2.956857802982567,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041305453061219903,
      "loss": 5.1279,
      "mean_token_accuracy": 0.18825956881046296,
      "num_tokens": 64944432.0,
      "step": 35195
    },
    {
      "entropy": 5.540423011779785,
      "epoch": 2.9572778827977313,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004130308825188604,
      "loss": 5.0407,
      "mean_token_accuracy": 0.19262988418340682,
      "num_tokens": 64953565.0,
      "step": 35200
    },
    {
      "entropy": 5.523712491989135,
      "epoch": 2.9576979626128965,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041300723198037573,
      "loss": 5.1422,
      "mean_token_accuracy": 0.1856449291110039,
      "num_tokens": 64963387.0,
      "step": 35205
    },
    {
      "entropy": 5.462106370925904,
      "epoch": 2.9581180424280613,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004129835789971639,
      "loss": 5.1212,
      "mean_token_accuracy": 0.198987153172493,
      "num_tokens": 64973481.0,
      "step": 35210
    },
    {
      "entropy": 5.5743485450744625,
      "epoch": 2.958538122243226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041295992356964417,
      "loss": 5.2045,
      "mean_token_accuracy": 0.18686718940734864,
      "num_tokens": 64982614.0,
      "step": 35215
    },
    {
      "entropy": 5.546708106994629,
      "epoch": 2.9589582020583913,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041293626569823547,
      "loss": 5.0932,
      "mean_token_accuracy": 0.1929064691066742,
      "num_tokens": 64991981.0,
      "step": 35220
    },
    {
      "entropy": 5.534565448760986,
      "epoch": 2.959378281873556,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004129126053833568,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19420332312583924,
      "num_tokens": 65001527.0,
      "step": 35225
    },
    {
      "entropy": 5.388376379013062,
      "epoch": 2.959798361688721,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041288894262542724,
      "loss": 5.0094,
      "mean_token_accuracy": 0.20361241549253464,
      "num_tokens": 65011021.0,
      "step": 35230
    },
    {
      "entropy": 5.549781942367554,
      "epoch": 2.9602184415038857,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041286527742486617,
      "loss": 5.1326,
      "mean_token_accuracy": 0.1916443571448326,
      "num_tokens": 65020151.0,
      "step": 35235
    },
    {
      "entropy": 5.468728733062744,
      "epoch": 2.9606385213190505,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004128416097820927,
      "loss": 5.1173,
      "mean_token_accuracy": 0.20254102498292922,
      "num_tokens": 65028526.0,
      "step": 35240
    },
    {
      "entropy": 5.453743886947632,
      "epoch": 2.9610586011342157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041281793969752607,
      "loss": 5.0449,
      "mean_token_accuracy": 0.2010686993598938,
      "num_tokens": 65038552.0,
      "step": 35245
    },
    {
      "entropy": 5.562491846084595,
      "epoch": 2.9614786809493805,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004127942671715856,
      "loss": 5.1811,
      "mean_token_accuracy": 0.19069550037384034,
      "num_tokens": 65048321.0,
      "step": 35250
    },
    {
      "entropy": 5.44348030090332,
      "epoch": 2.9618987607645453,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041277059220469063,
      "loss": 4.9548,
      "mean_token_accuracy": 0.2042457401752472,
      "num_tokens": 65057239.0,
      "step": 35255
    },
    {
      "entropy": 5.490614509582519,
      "epoch": 2.96231884057971,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041274691479726064,
      "loss": 5.12,
      "mean_token_accuracy": 0.19062627851963043,
      "num_tokens": 65066248.0,
      "step": 35260
    },
    {
      "entropy": 5.500769233703613,
      "epoch": 2.962738920394875,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041272323494971494,
      "loss": 5.105,
      "mean_token_accuracy": 0.19500290602445602,
      "num_tokens": 65076174.0,
      "step": 35265
    },
    {
      "entropy": 5.455958318710327,
      "epoch": 2.96315900021004,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000412699552662473,
      "loss": 5.0275,
      "mean_token_accuracy": 0.19583927541971208,
      "num_tokens": 65085588.0,
      "step": 35270
    },
    {
      "entropy": 5.565438985824585,
      "epoch": 2.963579080025205,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041267586793595443,
      "loss": 5.1439,
      "mean_token_accuracy": 0.19070078283548356,
      "num_tokens": 65095997.0,
      "step": 35275
    },
    {
      "entropy": 5.476970434188843,
      "epoch": 2.9639991598403697,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004126521807705787,
      "loss": 4.9691,
      "mean_token_accuracy": 0.20514864176511766,
      "num_tokens": 65104709.0,
      "step": 35280
    },
    {
      "entropy": 5.502714204788208,
      "epoch": 2.9644192396555344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004126284911667655,
      "loss": 5.0851,
      "mean_token_accuracy": 0.19659290313720704,
      "num_tokens": 65113884.0,
      "step": 35285
    },
    {
      "entropy": 5.5578453063964846,
      "epoch": 2.9648393194706992,
      "grad_norm": 1.125,
      "learning_rate": 0.0004126047991249345,
      "loss": 5.0915,
      "mean_token_accuracy": 0.19269661158323287,
      "num_tokens": 65122629.0,
      "step": 35290
    },
    {
      "entropy": 5.488770198822022,
      "epoch": 2.9652593992858645,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041258110464550523,
      "loss": 5.0514,
      "mean_token_accuracy": 0.1915840581059456,
      "num_tokens": 65132037.0,
      "step": 35295
    },
    {
      "entropy": 5.6069553852081295,
      "epoch": 2.9656794791010292,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041255740772889766,
      "loss": 5.2688,
      "mean_token_accuracy": 0.18819421231746675,
      "num_tokens": 65141679.0,
      "step": 35300
    },
    {
      "entropy": 5.555350208282471,
      "epoch": 2.966099558916194,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004125337083755314,
      "loss": 5.0969,
      "mean_token_accuracy": 0.1960112512111664,
      "num_tokens": 65150503.0,
      "step": 35305
    },
    {
      "entropy": 5.460469913482666,
      "epoch": 2.966519638731359,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004125100065858263,
      "loss": 5.1053,
      "mean_token_accuracy": 0.1913669764995575,
      "num_tokens": 65159742.0,
      "step": 35310
    },
    {
      "entropy": 5.494222974777221,
      "epoch": 2.9669397185465236,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004124863023602022,
      "loss": 5.102,
      "mean_token_accuracy": 0.18663200438022615,
      "num_tokens": 65169189.0,
      "step": 35315
    },
    {
      "entropy": 5.533269357681275,
      "epoch": 2.967359798361689,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041246259569907914,
      "loss": 5.045,
      "mean_token_accuracy": 0.19515585899353027,
      "num_tokens": 65177904.0,
      "step": 35320
    },
    {
      "entropy": 5.518507957458496,
      "epoch": 2.9677798781768536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004124388866028769,
      "loss": 5.1627,
      "mean_token_accuracy": 0.19160544574260713,
      "num_tokens": 65187139.0,
      "step": 35325
    },
    {
      "entropy": 5.507043790817261,
      "epoch": 2.9681999579920184,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041241517507201555,
      "loss": 5.1152,
      "mean_token_accuracy": 0.1944026082754135,
      "num_tokens": 65195760.0,
      "step": 35330
    },
    {
      "entropy": 5.508194065093994,
      "epoch": 2.968620037807183,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041239146110691515,
      "loss": 5.0299,
      "mean_token_accuracy": 0.19730065912008285,
      "num_tokens": 65205039.0,
      "step": 35335
    },
    {
      "entropy": 5.490274143218994,
      "epoch": 2.969040117622348,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004123677447079958,
      "loss": 5.1731,
      "mean_token_accuracy": 0.18618528097867965,
      "num_tokens": 65214274.0,
      "step": 35340
    },
    {
      "entropy": 5.529817008972168,
      "epoch": 2.969460197437513,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041234402587567756,
      "loss": 5.0478,
      "mean_token_accuracy": 0.19722741097211838,
      "num_tokens": 65223205.0,
      "step": 35345
    },
    {
      "entropy": 5.405241870880127,
      "epoch": 2.969880277252678,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041232030461038063,
      "loss": 4.9959,
      "mean_token_accuracy": 0.2009130522608757,
      "num_tokens": 65232009.0,
      "step": 35350
    },
    {
      "entropy": 5.497226238250732,
      "epoch": 2.970300357067843,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041229658091252515,
      "loss": 5.1275,
      "mean_token_accuracy": 0.20185416489839553,
      "num_tokens": 65240578.0,
      "step": 35355
    },
    {
      "entropy": 5.493226528167725,
      "epoch": 2.970720436883008,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004122728547825316,
      "loss": 5.0825,
      "mean_token_accuracy": 0.2000383973121643,
      "num_tokens": 65250143.0,
      "step": 35360
    },
    {
      "entropy": 5.525571298599243,
      "epoch": 2.971140516698173,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041224912622081993,
      "loss": 5.1295,
      "mean_token_accuracy": 0.1960099771618843,
      "num_tokens": 65258556.0,
      "step": 35365
    },
    {
      "entropy": 5.484020566940307,
      "epoch": 2.9715605965133376,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041222539522781077,
      "loss": 5.0572,
      "mean_token_accuracy": 0.19338155686855316,
      "num_tokens": 65267326.0,
      "step": 35370
    },
    {
      "entropy": 5.5295243740081785,
      "epoch": 2.9719806763285024,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004122016618039244,
      "loss": 5.1281,
      "mean_token_accuracy": 0.20135017037391661,
      "num_tokens": 65276564.0,
      "step": 35375
    },
    {
      "entropy": 5.604083728790283,
      "epoch": 2.972400756143667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004121779259495811,
      "loss": 5.0486,
      "mean_token_accuracy": 0.19949811100959777,
      "num_tokens": 65286265.0,
      "step": 35380
    },
    {
      "entropy": 5.449265623092652,
      "epoch": 2.9728208359588324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041215418766520165,
      "loss": 5.0604,
      "mean_token_accuracy": 0.19961916655302048,
      "num_tokens": 65295682.0,
      "step": 35385
    },
    {
      "entropy": 5.5119603157043455,
      "epoch": 2.973240915773997,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004121304469512063,
      "loss": 5.1523,
      "mean_token_accuracy": 0.19542266875505448,
      "num_tokens": 65304417.0,
      "step": 35390
    },
    {
      "entropy": 5.499112224578857,
      "epoch": 2.973660995589162,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041210670380801575,
      "loss": 5.0575,
      "mean_token_accuracy": 0.20192180573940277,
      "num_tokens": 65313195.0,
      "step": 35395
    },
    {
      "entropy": 5.480107164382934,
      "epoch": 2.9740810754043268,
      "grad_norm": 1.0,
      "learning_rate": 0.0004120829582360505,
      "loss": 5.0096,
      "mean_token_accuracy": 0.20237478315830232,
      "num_tokens": 65321938.0,
      "step": 35400
    },
    {
      "entropy": 5.559457445144654,
      "epoch": 2.9745011552194915,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004120592102357312,
      "loss": 5.1564,
      "mean_token_accuracy": 0.18851232379674912,
      "num_tokens": 65331652.0,
      "step": 35405
    },
    {
      "entropy": 5.485412359237671,
      "epoch": 2.9749212350346568,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041203545980747865,
      "loss": 5.1326,
      "mean_token_accuracy": 0.18927058726549148,
      "num_tokens": 65340237.0,
      "step": 35410
    },
    {
      "entropy": 5.423387861251831,
      "epoch": 2.9753413148498216,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041201170695171337,
      "loss": 5.0576,
      "mean_token_accuracy": 0.2020630195736885,
      "num_tokens": 65349856.0,
      "step": 35415
    },
    {
      "entropy": 5.533485794067383,
      "epoch": 2.9757613946649863,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004119879516688564,
      "loss": 5.0404,
      "mean_token_accuracy": 0.19841411113739013,
      "num_tokens": 65358599.0,
      "step": 35420
    },
    {
      "entropy": 5.566175937652588,
      "epoch": 2.976181474480151,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004119641939593283,
      "loss": 5.1049,
      "mean_token_accuracy": 0.18841985017061233,
      "num_tokens": 65367875.0,
      "step": 35425
    },
    {
      "entropy": 5.547258949279785,
      "epoch": 2.976601554295316,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041194043382355007,
      "loss": 5.1525,
      "mean_token_accuracy": 0.19471598118543626,
      "num_tokens": 65377490.0,
      "step": 35430
    },
    {
      "entropy": 5.452721166610718,
      "epoch": 2.977021634110481,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041191667126194253,
      "loss": 5.1976,
      "mean_token_accuracy": 0.1906827434897423,
      "num_tokens": 65386903.0,
      "step": 35435
    },
    {
      "entropy": 5.439539670944214,
      "epoch": 2.977441713925646,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041189290627492676,
      "loss": 5.0342,
      "mean_token_accuracy": 0.19877443909645082,
      "num_tokens": 65396178.0,
      "step": 35440
    },
    {
      "entropy": 5.534272336959839,
      "epoch": 2.9778617937408107,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041186913886292353,
      "loss": 5.1097,
      "mean_token_accuracy": 0.19739274680614471,
      "num_tokens": 65404904.0,
      "step": 35445
    },
    {
      "entropy": 5.475108861923218,
      "epoch": 2.9782818735559755,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041184536902635406,
      "loss": 5.0523,
      "mean_token_accuracy": 0.19470528662204742,
      "num_tokens": 65413806.0,
      "step": 35450
    },
    {
      "entropy": 5.489691591262817,
      "epoch": 2.9787019533711403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004118215967656393,
      "loss": 5.0598,
      "mean_token_accuracy": 0.19683267623186113,
      "num_tokens": 65422734.0,
      "step": 35455
    },
    {
      "entropy": 5.544067001342773,
      "epoch": 2.9791220331863055,
      "grad_norm": 1.125,
      "learning_rate": 0.0004117978220812004,
      "loss": 5.1274,
      "mean_token_accuracy": 0.19271580129861832,
      "num_tokens": 65431150.0,
      "step": 35460
    },
    {
      "entropy": 5.4435686588287355,
      "epoch": 2.9795421130014703,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041177404497345856,
      "loss": 5.0206,
      "mean_token_accuracy": 0.19746517837047578,
      "num_tokens": 65440086.0,
      "step": 35465
    },
    {
      "entropy": 5.524831342697143,
      "epoch": 2.979962192816635,
      "grad_norm": 1.125,
      "learning_rate": 0.0004117502654428349,
      "loss": 5.1565,
      "mean_token_accuracy": 0.19487004578113556,
      "num_tokens": 65449212.0,
      "step": 35470
    },
    {
      "entropy": 5.484037303924561,
      "epoch": 2.9803822726318,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004117264834897508,
      "loss": 5.034,
      "mean_token_accuracy": 0.20275093168020247,
      "num_tokens": 65458109.0,
      "step": 35475
    },
    {
      "entropy": 5.565641736984253,
      "epoch": 2.9808023524469647,
      "grad_norm": 0.96875,
      "learning_rate": 0.00041170269911462726,
      "loss": 5.189,
      "mean_token_accuracy": 0.19254602044820784,
      "num_tokens": 65467630.0,
      "step": 35480
    },
    {
      "entropy": 5.525800037384033,
      "epoch": 2.98122243226213,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041167891231788596,
      "loss": 5.1659,
      "mean_token_accuracy": 0.1875656247138977,
      "num_tokens": 65476778.0,
      "step": 35485
    },
    {
      "entropy": 5.432040739059448,
      "epoch": 2.9816425120772947,
      "grad_norm": 1.0625,
      "learning_rate": 0.000411655123099948,
      "loss": 5.0473,
      "mean_token_accuracy": 0.2007401242852211,
      "num_tokens": 65486472.0,
      "step": 35490
    },
    {
      "entropy": 5.519916200637818,
      "epoch": 2.9820625918924595,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000411631331461235,
      "loss": 5.0819,
      "mean_token_accuracy": 0.1925846055150032,
      "num_tokens": 65496591.0,
      "step": 35495
    },
    {
      "entropy": 5.559891080856323,
      "epoch": 2.9824826717076247,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041160753740216826,
      "loss": 5.1339,
      "mean_token_accuracy": 0.19280914068222046,
      "num_tokens": 65506021.0,
      "step": 35500
    },
    {
      "entropy": 5.374431133270264,
      "epoch": 2.982902751522789,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041158374092316937,
      "loss": 4.961,
      "mean_token_accuracy": 0.20128210335969926,
      "num_tokens": 65514795.0,
      "step": 35505
    },
    {
      "entropy": 5.440941667556762,
      "epoch": 2.9833228313379543,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041155994202465986,
      "loss": 5.0577,
      "mean_token_accuracy": 0.1956911250948906,
      "num_tokens": 65523217.0,
      "step": 35510
    },
    {
      "entropy": 5.542503833770752,
      "epoch": 2.983742911153119,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041153614070706124,
      "loss": 5.0824,
      "mean_token_accuracy": 0.18855067044496537,
      "num_tokens": 65532526.0,
      "step": 35515
    },
    {
      "entropy": 5.4421955108642575,
      "epoch": 2.984162990968284,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041151233697079523,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20307016968727112,
      "num_tokens": 65541149.0,
      "step": 35520
    },
    {
      "entropy": 5.557836055755615,
      "epoch": 2.984583070783449,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004114885308162834,
      "loss": 5.1969,
      "mean_token_accuracy": 0.1871769607067108,
      "num_tokens": 65550785.0,
      "step": 35525
    },
    {
      "entropy": 5.500041055679321,
      "epoch": 2.985003150598614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004114647222439476,
      "loss": 5.174,
      "mean_token_accuracy": 0.19322681427001953,
      "num_tokens": 65561115.0,
      "step": 35530
    },
    {
      "entropy": 5.594686460494995,
      "epoch": 2.9854232304137787,
      "grad_norm": 1.125,
      "learning_rate": 0.0004114409112542095,
      "loss": 5.143,
      "mean_token_accuracy": 0.18948471546173096,
      "num_tokens": 65569641.0,
      "step": 35535
    },
    {
      "entropy": 5.483253097534179,
      "epoch": 2.9858433102289434,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041141709784749093,
      "loss": 4.9633,
      "mean_token_accuracy": 0.20430168956518174,
      "num_tokens": 65578553.0,
      "step": 35540
    },
    {
      "entropy": 5.626095104217529,
      "epoch": 2.9862633900441082,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004113932820242137,
      "loss": 5.2453,
      "mean_token_accuracy": 0.18763289898633956,
      "num_tokens": 65587401.0,
      "step": 35545
    },
    {
      "entropy": 5.460124254226685,
      "epoch": 2.9866834698592735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004113694637847996,
      "loss": 5.0062,
      "mean_token_accuracy": 0.2014205738902092,
      "num_tokens": 65595997.0,
      "step": 35550
    },
    {
      "entropy": 5.461797428131104,
      "epoch": 2.9871035496744383,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041134564312967075,
      "loss": 5.1037,
      "mean_token_accuracy": 0.18331654816865922,
      "num_tokens": 65605849.0,
      "step": 35555
    },
    {
      "entropy": 5.5357273578643795,
      "epoch": 2.987523629489603,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004113218200592491,
      "loss": 5.0789,
      "mean_token_accuracy": 0.1993962198495865,
      "num_tokens": 65614951.0,
      "step": 35560
    },
    {
      "entropy": 5.415885496139526,
      "epoch": 2.987943709304768,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041129799457395653,
      "loss": 5.0316,
      "mean_token_accuracy": 0.20742578506469728,
      "num_tokens": 65623535.0,
      "step": 35565
    },
    {
      "entropy": 5.496593904495239,
      "epoch": 2.9883637891199326,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041127416667421517,
      "loss": 5.1368,
      "mean_token_accuracy": 0.19159200191497802,
      "num_tokens": 65631761.0,
      "step": 35570
    },
    {
      "entropy": 5.47874903678894,
      "epoch": 2.988783868935098,
      "grad_norm": 1.125,
      "learning_rate": 0.00041125033636044725,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20143855810165406,
      "num_tokens": 65640859.0,
      "step": 35575
    },
    {
      "entropy": 5.5057793140411375,
      "epoch": 2.9892039487502626,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041122650363307464,
      "loss": 5.1014,
      "mean_token_accuracy": 0.1914748951792717,
      "num_tokens": 65649901.0,
      "step": 35580
    },
    {
      "entropy": 5.433684062957764,
      "epoch": 2.9896240285654274,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004112026684925197,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20467381477355956,
      "num_tokens": 65658294.0,
      "step": 35585
    },
    {
      "entropy": 5.562026309967041,
      "epoch": 2.990044108380592,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004111788309392046,
      "loss": 5.1707,
      "mean_token_accuracy": 0.19323484152555465,
      "num_tokens": 65667441.0,
      "step": 35590
    },
    {
      "entropy": 5.540914916992188,
      "epoch": 2.990464188195757,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004111549909735517,
      "loss": 5.1099,
      "mean_token_accuracy": 0.19462796151638032,
      "num_tokens": 65676898.0,
      "step": 35595
    },
    {
      "entropy": 5.457164669036866,
      "epoch": 2.990884268010922,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041113114859598314,
      "loss": 5.0342,
      "mean_token_accuracy": 0.2013632282614708,
      "num_tokens": 65685650.0,
      "step": 35600
    },
    {
      "entropy": 5.516831731796264,
      "epoch": 2.991304347826087,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041110730380692153,
      "loss": 5.0803,
      "mean_token_accuracy": 0.1929681807756424,
      "num_tokens": 65694197.0,
      "step": 35605
    },
    {
      "entropy": 5.481831836700439,
      "epoch": 2.991724427641252,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041108345660678897,
      "loss": 5.0756,
      "mean_token_accuracy": 0.19539926201105118,
      "num_tokens": 65702604.0,
      "step": 35610
    },
    {
      "entropy": 5.53784065246582,
      "epoch": 2.9921445074564166,
      "grad_norm": 1.125,
      "learning_rate": 0.0004110596069960081,
      "loss": 5.1598,
      "mean_token_accuracy": 0.191267491877079,
      "num_tokens": 65711855.0,
      "step": 35615
    },
    {
      "entropy": 5.477375125885009,
      "epoch": 2.9925645872715814,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004110357549750013,
      "loss": 5.0913,
      "mean_token_accuracy": 0.20176489800214767,
      "num_tokens": 65720731.0,
      "step": 35620
    },
    {
      "entropy": 5.553304243087768,
      "epoch": 2.9929846670867466,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004110119005441913,
      "loss": 5.108,
      "mean_token_accuracy": 0.19568962901830672,
      "num_tokens": 65729578.0,
      "step": 35625
    },
    {
      "entropy": 5.577563238143921,
      "epoch": 2.9934047469019114,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041098804370400037,
      "loss": 5.2151,
      "mean_token_accuracy": 0.18625173568725586,
      "num_tokens": 65739970.0,
      "step": 35630
    },
    {
      "entropy": 5.595309400558472,
      "epoch": 2.993824826717076,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041096418445485134,
      "loss": 5.1446,
      "mean_token_accuracy": 0.18626275062561035,
      "num_tokens": 65749733.0,
      "step": 35635
    },
    {
      "entropy": 5.5051905632019045,
      "epoch": 2.9942449065322414,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004109403227971668,
      "loss": 5.1227,
      "mean_token_accuracy": 0.19194432348012924,
      "num_tokens": 65760516.0,
      "step": 35640
    },
    {
      "entropy": 5.512326002120972,
      "epoch": 2.9946649863474057,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041091645873136934,
      "loss": 5.191,
      "mean_token_accuracy": 0.18821463882923126,
      "num_tokens": 65770731.0,
      "step": 35645
    },
    {
      "entropy": 5.541664361953735,
      "epoch": 2.995085066162571,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004108925922578819,
      "loss": 5.1362,
      "mean_token_accuracy": 0.1915773034095764,
      "num_tokens": 65780710.0,
      "step": 35650
    },
    {
      "entropy": 5.47835750579834,
      "epoch": 2.9955051459777358,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004108687233771272,
      "loss": 5.0254,
      "mean_token_accuracy": 0.19902457892894745,
      "num_tokens": 65788843.0,
      "step": 35655
    },
    {
      "entropy": 5.486465978622436,
      "epoch": 2.9959252257929005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004108448520895279,
      "loss": 5.0474,
      "mean_token_accuracy": 0.19575307220220567,
      "num_tokens": 65797451.0,
      "step": 35660
    },
    {
      "entropy": 5.408519983291626,
      "epoch": 2.996345305608066,
      "grad_norm": 1.125,
      "learning_rate": 0.00041082097839550714,
      "loss": 5.0643,
      "mean_token_accuracy": 0.19066828936338426,
      "num_tokens": 65806701.0,
      "step": 35665
    },
    {
      "entropy": 5.496346092224121,
      "epoch": 2.9967653854232306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004107971022954876,
      "loss": 5.0911,
      "mean_token_accuracy": 0.19013802111148834,
      "num_tokens": 65816384.0,
      "step": 35670
    },
    {
      "entropy": 5.583408689498901,
      "epoch": 2.9971854652383954,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041077322378989224,
      "loss": 5.1978,
      "mean_token_accuracy": 0.18999014496803285,
      "num_tokens": 65826595.0,
      "step": 35675
    },
    {
      "entropy": 5.498501396179199,
      "epoch": 2.99760554505356,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004107493428791442,
      "loss": 5.0656,
      "mean_token_accuracy": 0.19699027985334397,
      "num_tokens": 65836540.0,
      "step": 35680
    },
    {
      "entropy": 5.51586594581604,
      "epoch": 2.998025624868725,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041072545956366655,
      "loss": 5.1293,
      "mean_token_accuracy": 0.19856082946062087,
      "num_tokens": 65845594.0,
      "step": 35685
    },
    {
      "entropy": 5.452786779403686,
      "epoch": 2.99844570468389,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004107015738438821,
      "loss": 5.036,
      "mean_token_accuracy": 0.1997885525226593,
      "num_tokens": 65854634.0,
      "step": 35690
    },
    {
      "entropy": 5.451989650726318,
      "epoch": 2.998865784499055,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004106776857202143,
      "loss": 5.0367,
      "mean_token_accuracy": 0.1994938537478447,
      "num_tokens": 65863677.0,
      "step": 35695
    },
    {
      "entropy": 5.479758214950562,
      "epoch": 2.9992858643142197,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004106537951930861,
      "loss": 5.0684,
      "mean_token_accuracy": 0.20076475888490677,
      "num_tokens": 65873140.0,
      "step": 35700
    },
    {
      "entropy": 5.438322973251343,
      "epoch": 2.9997059441293845,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00041062990226292076,
      "loss": 5.0804,
      "mean_token_accuracy": 0.1886454463005066,
      "num_tokens": 65883269.0,
      "step": 35705
    },
    {
      "entropy": 5.485963503519694,
      "epoch": 3.000084015963033,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041060600693014157,
      "loss": 5.0576,
      "mean_token_accuracy": 0.19979129566086662,
      "num_tokens": 65891649.0,
      "step": 35710
    },
    {
      "entropy": 5.5296557426452635,
      "epoch": 3.0005040957781977,
      "grad_norm": 1.0,
      "learning_rate": 0.0004105821091951717,
      "loss": 5.1111,
      "mean_token_accuracy": 0.19324134290218353,
      "num_tokens": 65901044.0,
      "step": 35715
    },
    {
      "entropy": 5.487710905075073,
      "epoch": 3.000924175593363,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041055820905843473,
      "loss": 4.9913,
      "mean_token_accuracy": 0.20260340869426727,
      "num_tokens": 65910382.0,
      "step": 35720
    },
    {
      "entropy": 5.449863052368164,
      "epoch": 3.0013442554085277,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004105343065203538,
      "loss": 4.9997,
      "mean_token_accuracy": 0.1985916882753372,
      "num_tokens": 65919770.0,
      "step": 35725
    },
    {
      "entropy": 5.443807792663574,
      "epoch": 3.0017643352236925,
      "grad_norm": 1.125,
      "learning_rate": 0.00041051040158135237,
      "loss": 5.0036,
      "mean_token_accuracy": 0.19964202344417573,
      "num_tokens": 65928451.0,
      "step": 35730
    },
    {
      "entropy": 5.502658033370972,
      "epoch": 3.0021844150388572,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004104864942418541,
      "loss": 5.0026,
      "mean_token_accuracy": 0.2007636770606041,
      "num_tokens": 65938499.0,
      "step": 35735
    },
    {
      "entropy": 5.485469341278076,
      "epoch": 3.0026044948540225,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041046258450228224,
      "loss": 5.0248,
      "mean_token_accuracy": 0.20123015493154525,
      "num_tokens": 65946967.0,
      "step": 35740
    },
    {
      "entropy": 5.4917004108428955,
      "epoch": 3.0030245746691873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004104386723630604,
      "loss": 5.1242,
      "mean_token_accuracy": 0.19013231694698335,
      "num_tokens": 65956524.0,
      "step": 35745
    },
    {
      "entropy": 5.558381652832031,
      "epoch": 3.003444654484352,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004104147578246123,
      "loss": 5.06,
      "mean_token_accuracy": 0.19858537763357162,
      "num_tokens": 65965483.0,
      "step": 35750
    },
    {
      "entropy": 5.443302869796753,
      "epoch": 3.003864734299517,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004103908408873615,
      "loss": 4.9783,
      "mean_token_accuracy": 0.20572655647993088,
      "num_tokens": 65974603.0,
      "step": 35755
    },
    {
      "entropy": 5.519117832183838,
      "epoch": 3.0042848141146816,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004103669215517317,
      "loss": 5.0553,
      "mean_token_accuracy": 0.19730354994535446,
      "num_tokens": 65983504.0,
      "step": 35760
    },
    {
      "entropy": 5.4861527442932125,
      "epoch": 3.004704893929847,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041034299981814647,
      "loss": 5.0626,
      "mean_token_accuracy": 0.1980084776878357,
      "num_tokens": 65993078.0,
      "step": 35765
    },
    {
      "entropy": 5.482875490188599,
      "epoch": 3.0051249737450116,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041031907568702975,
      "loss": 5.0533,
      "mean_token_accuracy": 0.20096867233514787,
      "num_tokens": 66001371.0,
      "step": 35770
    },
    {
      "entropy": 5.4631791591644285,
      "epoch": 3.0055450535601764,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004102951491588054,
      "loss": 4.973,
      "mean_token_accuracy": 0.20448427349328996,
      "num_tokens": 66011066.0,
      "step": 35775
    },
    {
      "entropy": 5.485814189910888,
      "epoch": 3.005965133375341,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041027122023389696,
      "loss": 5.0767,
      "mean_token_accuracy": 0.191511832177639,
      "num_tokens": 66020342.0,
      "step": 35780
    },
    {
      "entropy": 5.568971824645996,
      "epoch": 3.006385213190506,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004102472889127286,
      "loss": 5.1397,
      "mean_token_accuracy": 0.19151001721620559,
      "num_tokens": 66029994.0,
      "step": 35785
    },
    {
      "entropy": 5.438309192657471,
      "epoch": 3.0068052930056712,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004102233551957241,
      "loss": 4.9173,
      "mean_token_accuracy": 0.20707933306694032,
      "num_tokens": 66038696.0,
      "step": 35790
    },
    {
      "entropy": 5.519137048721314,
      "epoch": 3.007225372820836,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004101994190833076,
      "loss": 5.1174,
      "mean_token_accuracy": 0.19275708347558976,
      "num_tokens": 66048032.0,
      "step": 35795
    },
    {
      "entropy": 5.476726531982422,
      "epoch": 3.007645452636001,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041017548057590295,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20051975250244142,
      "num_tokens": 66057494.0,
      "step": 35800
    },
    {
      "entropy": 5.513273096084594,
      "epoch": 3.0080655324511656,
      "grad_norm": 1.125,
      "learning_rate": 0.00041015153967393427,
      "loss": 5.0015,
      "mean_token_accuracy": 0.20037806332111358,
      "num_tokens": 66066525.0,
      "step": 35805
    },
    {
      "entropy": 5.5109189510345455,
      "epoch": 3.008485612266331,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004101275963778257,
      "loss": 5.0188,
      "mean_token_accuracy": 0.19393327981233596,
      "num_tokens": 66075204.0,
      "step": 35810
    },
    {
      "entropy": 5.49214243888855,
      "epoch": 3.0089056920814956,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041010365068800124,
      "loss": 5.0336,
      "mean_token_accuracy": 0.20342217087745668,
      "num_tokens": 66084070.0,
      "step": 35815
    },
    {
      "entropy": 5.466468811035156,
      "epoch": 3.0093257718966604,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004100797026048852,
      "loss": 5.0645,
      "mean_token_accuracy": 0.19535394608974457,
      "num_tokens": 66093891.0,
      "step": 35820
    },
    {
      "entropy": 5.4435924053192135,
      "epoch": 3.009745851711825,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041005575212890185,
      "loss": 5.0168,
      "mean_token_accuracy": 0.19035459458827972,
      "num_tokens": 66102605.0,
      "step": 35825
    },
    {
      "entropy": 5.542224264144897,
      "epoch": 3.01016593152699,
      "grad_norm": 1.125,
      "learning_rate": 0.00041003179926047534,
      "loss": 5.1434,
      "mean_token_accuracy": 0.18749296069145202,
      "num_tokens": 66111645.0,
      "step": 35830
    },
    {
      "entropy": 5.50278115272522,
      "epoch": 3.010586011342155,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004100078440000301,
      "loss": 5.0575,
      "mean_token_accuracy": 0.20439205914735795,
      "num_tokens": 66120444.0,
      "step": 35835
    },
    {
      "entropy": 5.532349538803101,
      "epoch": 3.01100609115732,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004099838863479904,
      "loss": 5.0464,
      "mean_token_accuracy": 0.19419250786304473,
      "num_tokens": 66129934.0,
      "step": 35840
    },
    {
      "entropy": 5.550216102600098,
      "epoch": 3.0114261709724848,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004099599263047806,
      "loss": 5.0514,
      "mean_token_accuracy": 0.20152912884950638,
      "num_tokens": 66139898.0,
      "step": 35845
    },
    {
      "entropy": 5.505129671096801,
      "epoch": 3.0118462507876496,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004099359638708253,
      "loss": 5.035,
      "mean_token_accuracy": 0.2014750525355339,
      "num_tokens": 66148036.0,
      "step": 35850
    },
    {
      "entropy": 5.394059848785401,
      "epoch": 3.0122663306028143,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004099119990465487,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20779180526733398,
      "num_tokens": 66157249.0,
      "step": 35855
    },
    {
      "entropy": 5.472588157653808,
      "epoch": 3.0126864104179796,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004098880318323757,
      "loss": 5.0486,
      "mean_token_accuracy": 0.19863729625940324,
      "num_tokens": 66167676.0,
      "step": 35860
    },
    {
      "entropy": 5.486320495605469,
      "epoch": 3.0131064902331444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040986406222873065,
      "loss": 5.0337,
      "mean_token_accuracy": 0.19985527098178862,
      "num_tokens": 66176925.0,
      "step": 35865
    },
    {
      "entropy": 5.482815408706665,
      "epoch": 3.013526570048309,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040984009023603806,
      "loss": 4.974,
      "mean_token_accuracy": 0.198090261220932,
      "num_tokens": 66186656.0,
      "step": 35870
    },
    {
      "entropy": 5.439781427383423,
      "epoch": 3.013946649863474,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040981611585472276,
      "loss": 5.0177,
      "mean_token_accuracy": 0.20676226764917374,
      "num_tokens": 66196240.0,
      "step": 35875
    },
    {
      "entropy": 5.499134540557861,
      "epoch": 3.0143667296786387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004097921390852094,
      "loss": 5.0222,
      "mean_token_accuracy": 0.19915864914655684,
      "num_tokens": 66205153.0,
      "step": 35880
    },
    {
      "entropy": 5.496306228637695,
      "epoch": 3.014786809493804,
      "grad_norm": 1.125,
      "learning_rate": 0.0004097681599279227,
      "loss": 5.0466,
      "mean_token_accuracy": 0.19405843317508698,
      "num_tokens": 66214063.0,
      "step": 35885
    },
    {
      "entropy": 5.450627279281616,
      "epoch": 3.0152068893089687,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004097441783832875,
      "loss": 5.0141,
      "mean_token_accuracy": 0.19936386346817017,
      "num_tokens": 66222647.0,
      "step": 35890
    },
    {
      "entropy": 5.46258020401001,
      "epoch": 3.0156269691241335,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004097201944517285,
      "loss": 5.0505,
      "mean_token_accuracy": 0.20301371216773986,
      "num_tokens": 66231701.0,
      "step": 35895
    },
    {
      "entropy": 5.505861568450928,
      "epoch": 3.0160470489392983,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040969620813367065,
      "loss": 5.0291,
      "mean_token_accuracy": 0.19977846294641494,
      "num_tokens": 66240927.0,
      "step": 35900
    },
    {
      "entropy": 5.51290454864502,
      "epoch": 3.0164671287544635,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004096722194295388,
      "loss": 5.0189,
      "mean_token_accuracy": 0.19790696650743483,
      "num_tokens": 66249688.0,
      "step": 35905
    },
    {
      "entropy": 5.468650913238525,
      "epoch": 3.0168872085696283,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004096482283397579,
      "loss": 5.0025,
      "mean_token_accuracy": 0.19544079154729843,
      "num_tokens": 66258981.0,
      "step": 35910
    },
    {
      "entropy": 5.491763353347778,
      "epoch": 3.017307288384793,
      "grad_norm": 1.03125,
      "learning_rate": 0.000409624234864753,
      "loss": 5.0554,
      "mean_token_accuracy": 0.19533132314682006,
      "num_tokens": 66268049.0,
      "step": 35915
    },
    {
      "entropy": 5.454478168487549,
      "epoch": 3.017727368199958,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004096002390049491,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20310468077659607,
      "num_tokens": 66277165.0,
      "step": 35920
    },
    {
      "entropy": 5.462166357040405,
      "epoch": 3.0181474480151227,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004095762407607714,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20003026127815246,
      "num_tokens": 66286717.0,
      "step": 35925
    },
    {
      "entropy": 5.492368698120117,
      "epoch": 3.018567527830288,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040955224013264476,
      "loss": 5.0634,
      "mean_token_accuracy": 0.2001332938671112,
      "num_tokens": 66296326.0,
      "step": 35930
    },
    {
      "entropy": 5.495102262496948,
      "epoch": 3.0189876076454527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004095282371209945,
      "loss": 5.0776,
      "mean_token_accuracy": 0.19623177498579025,
      "num_tokens": 66306362.0,
      "step": 35935
    },
    {
      "entropy": 5.542889738082886,
      "epoch": 3.0194076874606175,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040950423172624576,
      "loss": 5.0677,
      "mean_token_accuracy": 0.18984897285699845,
      "num_tokens": 66316281.0,
      "step": 35940
    },
    {
      "entropy": 5.523416805267334,
      "epoch": 3.0198277672757823,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040948022394882383,
      "loss": 5.0341,
      "mean_token_accuracy": 0.2006482884287834,
      "num_tokens": 66325415.0,
      "step": 35945
    },
    {
      "entropy": 5.424595880508423,
      "epoch": 3.020247847090947,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040945621378915396,
      "loss": 4.9401,
      "mean_token_accuracy": 0.20529822558164595,
      "num_tokens": 66333872.0,
      "step": 35950
    },
    {
      "entropy": 5.463702726364136,
      "epoch": 3.0206679269061123,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040943220124766154,
      "loss": 5.0588,
      "mean_token_accuracy": 0.20014411807060242,
      "num_tokens": 66343383.0,
      "step": 35955
    },
    {
      "entropy": 5.484094572067261,
      "epoch": 3.021088006721277,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040940818632477185,
      "loss": 5.0064,
      "mean_token_accuracy": 0.19478949755430222,
      "num_tokens": 66352191.0,
      "step": 35960
    },
    {
      "entropy": 5.361633157730102,
      "epoch": 3.021508086536442,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004093841690209105,
      "loss": 4.8507,
      "mean_token_accuracy": 0.21633446514606475,
      "num_tokens": 66360913.0,
      "step": 35965
    },
    {
      "entropy": 5.5373122215271,
      "epoch": 3.0219281663516067,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040936014933650263,
      "loss": 5.1356,
      "mean_token_accuracy": 0.20071977078914643,
      "num_tokens": 66369180.0,
      "step": 35970
    },
    {
      "entropy": 5.487418508529663,
      "epoch": 3.022348246166772,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040933612727197394,
      "loss": 5.0188,
      "mean_token_accuracy": 0.2048914536833763,
      "num_tokens": 66378538.0,
      "step": 35975
    },
    {
      "entropy": 5.52368426322937,
      "epoch": 3.0227683259819367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040931210282774995,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19048984050750734,
      "num_tokens": 66387993.0,
      "step": 35980
    },
    {
      "entropy": 5.452989387512207,
      "epoch": 3.0231884057971015,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040928807600425617,
      "loss": 4.9717,
      "mean_token_accuracy": 0.20261990278959274,
      "num_tokens": 66397218.0,
      "step": 35985
    },
    {
      "entropy": 5.371537685394287,
      "epoch": 3.0236084856122663,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004092640468019183,
      "loss": 4.9331,
      "mean_token_accuracy": 0.20737528800964355,
      "num_tokens": 66406633.0,
      "step": 35990
    },
    {
      "entropy": 5.385524702072144,
      "epoch": 3.024028565427431,
      "grad_norm": 1.015625,
      "learning_rate": 0.000409240015221162,
      "loss": 4.9466,
      "mean_token_accuracy": 0.20517672896385192,
      "num_tokens": 66415535.0,
      "step": 35995
    },
    {
      "entropy": 5.5251946449279785,
      "epoch": 3.0244486452425963,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004092159812624129,
      "loss": 5.0616,
      "mean_token_accuracy": 0.19322190582752227,
      "num_tokens": 66425574.0,
      "step": 36000
    },
    {
      "epoch": 3.0244486452425963,
      "eval_entropy": 5.3236203714296,
      "eval_loss": 5.162716388702393,
      "eval_mean_token_accuracy": 0.20078897676537313,
      "eval_num_tokens": 66425574.0,
      "eval_runtime": 27.4025,
      "eval_samples_per_second": 1363.598,
      "eval_steps_per_second": 170.459,
      "step": 36000
    },
    {
      "entropy": 5.455542945861817,
      "epoch": 3.024868725057761,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040919194492609683,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20410357117652894,
      "num_tokens": 66434385.0,
      "step": 36005
    },
    {
      "entropy": 5.468975925445557,
      "epoch": 3.025288804872926,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004091679062126396,
      "loss": 5.0341,
      "mean_token_accuracy": 0.1964765563607216,
      "num_tokens": 66443462.0,
      "step": 36010
    },
    {
      "entropy": 5.500638818740844,
      "epoch": 3.0257088846880906,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000409143865122467,
      "loss": 5.1309,
      "mean_token_accuracy": 0.19154869467020036,
      "num_tokens": 66452035.0,
      "step": 36015
    },
    {
      "entropy": 5.401268815994262,
      "epoch": 3.0261289645032554,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004091198216560048,
      "loss": 4.9357,
      "mean_token_accuracy": 0.2033479243516922,
      "num_tokens": 66460912.0,
      "step": 36020
    },
    {
      "entropy": 5.5332005500793455,
      "epoch": 3.0265490443184206,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004090957758136791,
      "loss": 5.0751,
      "mean_token_accuracy": 0.19753794223070145,
      "num_tokens": 66470096.0,
      "step": 36025
    },
    {
      "entropy": 5.594774341583252,
      "epoch": 3.0269691241335854,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040907172759591576,
      "loss": 5.1317,
      "mean_token_accuracy": 0.19477718323469162,
      "num_tokens": 66479297.0,
      "step": 36030
    },
    {
      "entropy": 5.5403180599212645,
      "epoch": 3.02738920394875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004090476770031408,
      "loss": 5.0914,
      "mean_token_accuracy": 0.1938771739602089,
      "num_tokens": 66488343.0,
      "step": 36035
    },
    {
      "entropy": 5.539695072174072,
      "epoch": 3.027809283763915,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040902362403578025,
      "loss": 5.0464,
      "mean_token_accuracy": 0.20832083225250245,
      "num_tokens": 66497692.0,
      "step": 36040
    },
    {
      "entropy": 5.494339370727539,
      "epoch": 3.0282293635790802,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040899956869426025,
      "loss": 5.036,
      "mean_token_accuracy": 0.19738152921199797,
      "num_tokens": 66507088.0,
      "step": 36045
    },
    {
      "entropy": 5.464199733734131,
      "epoch": 3.028649443394245,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040897551097900684,
      "loss": 5.0051,
      "mean_token_accuracy": 0.2026059329509735,
      "num_tokens": 66516577.0,
      "step": 36050
    },
    {
      "entropy": 5.451547336578369,
      "epoch": 3.02906952320941,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004089514508904463,
      "loss": 5.0429,
      "mean_token_accuracy": 0.1956268846988678,
      "num_tokens": 66526116.0,
      "step": 36055
    },
    {
      "entropy": 5.480962228775025,
      "epoch": 3.0294896030245746,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040892738842900474,
      "loss": 5.0109,
      "mean_token_accuracy": 0.19596570134162902,
      "num_tokens": 66535410.0,
      "step": 36060
    },
    {
      "entropy": 5.593154048919677,
      "epoch": 3.0299096828397394,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040890332359510846,
      "loss": 5.1838,
      "mean_token_accuracy": 0.1922274649143219,
      "num_tokens": 66544853.0,
      "step": 36065
    },
    {
      "entropy": 5.453614521026611,
      "epoch": 3.0303297626549046,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040887925638918377,
      "loss": 5.0018,
      "mean_token_accuracy": 0.2026774138212204,
      "num_tokens": 66553603.0,
      "step": 36070
    },
    {
      "entropy": 5.488261985778808,
      "epoch": 3.0307498424700694,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040885518681165703,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20541795939207078,
      "num_tokens": 66561971.0,
      "step": 36075
    },
    {
      "entropy": 5.433129215240479,
      "epoch": 3.031169922285234,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004088311148629545,
      "loss": 4.9827,
      "mean_token_accuracy": 0.19687819629907607,
      "num_tokens": 66571373.0,
      "step": 36080
    },
    {
      "entropy": 5.526107358932495,
      "epoch": 3.031590002100399,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040880704054350276,
      "loss": 5.1228,
      "mean_token_accuracy": 0.19249015897512436,
      "num_tokens": 66581372.0,
      "step": 36085
    },
    {
      "entropy": 5.574275636672974,
      "epoch": 3.0320100819155638,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004087829638537281,
      "loss": 5.109,
      "mean_token_accuracy": 0.19206047803163528,
      "num_tokens": 66590269.0,
      "step": 36090
    },
    {
      "entropy": 5.51594729423523,
      "epoch": 3.032430161730729,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004087588847940572,
      "loss": 5.0997,
      "mean_token_accuracy": 0.19232838302850724,
      "num_tokens": 66599746.0,
      "step": 36095
    },
    {
      "entropy": 5.397035932540893,
      "epoch": 3.0328502415458938,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040873480336491654,
      "loss": 4.877,
      "mean_token_accuracy": 0.21987744867801667,
      "num_tokens": 66609186.0,
      "step": 36100
    },
    {
      "entropy": 5.527581024169922,
      "epoch": 3.0332703213610586,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004087107195667326,
      "loss": 5.1191,
      "mean_token_accuracy": 0.19499116390943527,
      "num_tokens": 66619054.0,
      "step": 36105
    },
    {
      "entropy": 5.555025339126587,
      "epoch": 3.0336904011762234,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040868663339993223,
      "loss": 5.0771,
      "mean_token_accuracy": 0.1989346042275429,
      "num_tokens": 66628859.0,
      "step": 36110
    },
    {
      "entropy": 5.474628448486328,
      "epoch": 3.0341104809913886,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004086625448649419,
      "loss": 4.9873,
      "mean_token_accuracy": 0.20424953550100328,
      "num_tokens": 66637737.0,
      "step": 36115
    },
    {
      "entropy": 5.5028894424438475,
      "epoch": 3.0345305608065534,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040863845396218844,
      "loss": 5.0862,
      "mean_token_accuracy": 0.1954240009188652,
      "num_tokens": 66647363.0,
      "step": 36120
    },
    {
      "entropy": 5.5285218238830565,
      "epoch": 3.034950640621718,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004086143606920986,
      "loss": 5.0381,
      "mean_token_accuracy": 0.19733420312404631,
      "num_tokens": 66656947.0,
      "step": 36125
    },
    {
      "entropy": 5.456996631622315,
      "epoch": 3.035370720436883,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00040859026505509925,
      "loss": 5.0007,
      "mean_token_accuracy": 0.20412792563438414,
      "num_tokens": 66666891.0,
      "step": 36130
    },
    {
      "entropy": 5.4891808986663815,
      "epoch": 3.0357908002520477,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000408566167051617,
      "loss": 5.0497,
      "mean_token_accuracy": 0.19665799289941788,
      "num_tokens": 66676161.0,
      "step": 36135
    },
    {
      "entropy": 5.490340089797973,
      "epoch": 3.036210880067213,
      "grad_norm": 1.0,
      "learning_rate": 0.00040854206668207893,
      "loss": 5.1015,
      "mean_token_accuracy": 0.19213471859693526,
      "num_tokens": 66684750.0,
      "step": 36140
    },
    {
      "entropy": 5.476293563842773,
      "epoch": 3.0366309598823777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040851796394691193,
      "loss": 4.968,
      "mean_token_accuracy": 0.20605463832616805,
      "num_tokens": 66694153.0,
      "step": 36145
    },
    {
      "entropy": 5.4875670909881595,
      "epoch": 3.0370510396975425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040849385884654294,
      "loss": 5.0064,
      "mean_token_accuracy": 0.20521130859851838,
      "num_tokens": 66704311.0,
      "step": 36150
    },
    {
      "entropy": 5.487697410583496,
      "epoch": 3.0374711195127073,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040846975138139895,
      "loss": 5.1027,
      "mean_token_accuracy": 0.19297085404396058,
      "num_tokens": 66714280.0,
      "step": 36155
    },
    {
      "entropy": 5.472384643554688,
      "epoch": 3.037891199327872,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040844564155190704,
      "loss": 4.9761,
      "mean_token_accuracy": 0.2064717158675194,
      "num_tokens": 66723750.0,
      "step": 36160
    },
    {
      "entropy": 5.458390426635742,
      "epoch": 3.0383112791430373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004084215293584943,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20363476425409316,
      "num_tokens": 66733370.0,
      "step": 36165
    },
    {
      "entropy": 5.40551528930664,
      "epoch": 3.038731358958202,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040839741480158795,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20807172060012818,
      "num_tokens": 66743364.0,
      "step": 36170
    },
    {
      "entropy": 5.568520641326904,
      "epoch": 3.039151438773367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040837329788161506,
      "loss": 5.167,
      "mean_token_accuracy": 0.18733127266168595,
      "num_tokens": 66753383.0,
      "step": 36175
    },
    {
      "entropy": 5.581145334243774,
      "epoch": 3.0395715185885317,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040834917859900285,
      "loss": 5.0931,
      "mean_token_accuracy": 0.18787057995796203,
      "num_tokens": 66762563.0,
      "step": 36180
    },
    {
      "entropy": 5.508365345001221,
      "epoch": 3.039991598403697,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004083250569541786,
      "loss": 5.0695,
      "mean_token_accuracy": 0.2020334392786026,
      "num_tokens": 66771433.0,
      "step": 36185
    },
    {
      "entropy": 5.601531600952148,
      "epoch": 3.0404116782188617,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040830093294756957,
      "loss": 5.1339,
      "mean_token_accuracy": 0.1956314980983734,
      "num_tokens": 66780107.0,
      "step": 36190
    },
    {
      "entropy": 5.481285190582275,
      "epoch": 3.0408317580340265,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040827680657960324,
      "loss": 5.0393,
      "mean_token_accuracy": 0.20289119184017182,
      "num_tokens": 66789310.0,
      "step": 36195
    },
    {
      "entropy": 5.555394411087036,
      "epoch": 3.0412518378491913,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040825267785070683,
      "loss": 5.1084,
      "mean_token_accuracy": 0.20279776751995088,
      "num_tokens": 66798445.0,
      "step": 36200
    },
    {
      "entropy": 5.443537902832031,
      "epoch": 3.041671917664356,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040822854676130795,
      "loss": 5.0592,
      "mean_token_accuracy": 0.19347923547029494,
      "num_tokens": 66807329.0,
      "step": 36205
    },
    {
      "entropy": 5.468757677078247,
      "epoch": 3.0420919974795213,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004082044133118339,
      "loss": 5.0058,
      "mean_token_accuracy": 0.20401106178760528,
      "num_tokens": 66816568.0,
      "step": 36210
    },
    {
      "entropy": 5.497680187225342,
      "epoch": 3.042512077294686,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004081802775027123,
      "loss": 5.0524,
      "mean_token_accuracy": 0.19302351921796798,
      "num_tokens": 66826285.0,
      "step": 36215
    },
    {
      "entropy": 5.414137554168701,
      "epoch": 3.042932157109851,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040815613933437055,
      "loss": 4.9378,
      "mean_token_accuracy": 0.2057136192917824,
      "num_tokens": 66835358.0,
      "step": 36220
    },
    {
      "entropy": 5.430384540557862,
      "epoch": 3.0433522369250157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004081319988072365,
      "loss": 5.0193,
      "mean_token_accuracy": 0.19530170857906343,
      "num_tokens": 66844563.0,
      "step": 36225
    },
    {
      "entropy": 5.509450769424438,
      "epoch": 3.0437723167401805,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004081078559217375,
      "loss": 5.0943,
      "mean_token_accuracy": 0.20017892867326736,
      "num_tokens": 66854910.0,
      "step": 36230
    },
    {
      "entropy": 5.492685985565186,
      "epoch": 3.0441923965553457,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004080837106783015,
      "loss": 5.0815,
      "mean_token_accuracy": 0.1993740439414978,
      "num_tokens": 66863911.0,
      "step": 36235
    },
    {
      "entropy": 5.458597993850708,
      "epoch": 3.0446124763705105,
      "grad_norm": 1.046875,
      "learning_rate": 0.000408059563077356,
      "loss": 4.9301,
      "mean_token_accuracy": 0.2019940584897995,
      "num_tokens": 66872783.0,
      "step": 36240
    },
    {
      "entropy": 5.371376705169678,
      "epoch": 3.0450325561856753,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040803541311932895,
      "loss": 4.945,
      "mean_token_accuracy": 0.2028413623571396,
      "num_tokens": 66881605.0,
      "step": 36245
    },
    {
      "entropy": 5.4244225978851315,
      "epoch": 3.04545263600084,
      "grad_norm": 1.0625,
      "learning_rate": 0.000408011260804648,
      "loss": 4.9931,
      "mean_token_accuracy": 0.20151930004358293,
      "num_tokens": 66891553.0,
      "step": 36250
    },
    {
      "entropy": 5.5209822177886965,
      "epoch": 3.045872715816005,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004079871061337411,
      "loss": 5.1444,
      "mean_token_accuracy": 0.19304833859205245,
      "num_tokens": 66900152.0,
      "step": 36255
    },
    {
      "entropy": 5.4757325649261475,
      "epoch": 3.04629279563117,
      "grad_norm": 1.078125,
      "learning_rate": 0.000407962949107036,
      "loss": 4.9812,
      "mean_token_accuracy": 0.19509021639823915,
      "num_tokens": 66909028.0,
      "step": 36260
    },
    {
      "entropy": 5.488673686981201,
      "epoch": 3.046712875446335,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004079387897249608,
      "loss": 4.9743,
      "mean_token_accuracy": 0.2056784749031067,
      "num_tokens": 66918714.0,
      "step": 36265
    },
    {
      "entropy": 5.511315202713012,
      "epoch": 3.0471329552614996,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004079146279879434,
      "loss": 5.1071,
      "mean_token_accuracy": 0.19049689471721648,
      "num_tokens": 66928437.0,
      "step": 36270
    },
    {
      "entropy": 5.3597376346588135,
      "epoch": 3.0475530350766644,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004078904638964118,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21105151772499084,
      "num_tokens": 66937303.0,
      "step": 36275
    },
    {
      "entropy": 5.435036945343017,
      "epoch": 3.0479731148918296,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040786629745079403,
      "loss": 5.0132,
      "mean_token_accuracy": 0.1964663580060005,
      "num_tokens": 66947536.0,
      "step": 36280
    },
    {
      "entropy": 5.488442993164062,
      "epoch": 3.0483931947069944,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004078421286515182,
      "loss": 5.026,
      "mean_token_accuracy": 0.19340611696243287,
      "num_tokens": 66955970.0,
      "step": 36285
    },
    {
      "entropy": 5.37516222000122,
      "epoch": 3.0488132745221592,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004078179574990125,
      "loss": 4.9454,
      "mean_token_accuracy": 0.20353869795799256,
      "num_tokens": 66964407.0,
      "step": 36290
    },
    {
      "entropy": 5.535750865936279,
      "epoch": 3.049233354337324,
      "grad_norm": 1.125,
      "learning_rate": 0.000407793783993705,
      "loss": 5.1055,
      "mean_token_accuracy": 0.198799829185009,
      "num_tokens": 66973961.0,
      "step": 36295
    },
    {
      "entropy": 5.510569906234741,
      "epoch": 3.049653434152489,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004077696081360241,
      "loss": 5.0272,
      "mean_token_accuracy": 0.199089178442955,
      "num_tokens": 66982792.0,
      "step": 36300
    },
    {
      "entropy": 5.4068389415740965,
      "epoch": 3.050073513967654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004077454299263979,
      "loss": 4.9847,
      "mean_token_accuracy": 0.20138517022132874,
      "num_tokens": 66992081.0,
      "step": 36305
    },
    {
      "entropy": 5.425614833831787,
      "epoch": 3.050493593782819,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040772124936525484,
      "loss": 4.9721,
      "mean_token_accuracy": 0.2073587119579315,
      "num_tokens": 67001242.0,
      "step": 36310
    },
    {
      "entropy": 5.408051252365112,
      "epoch": 3.0509136735979836,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040769706645302315,
      "loss": 4.9581,
      "mean_token_accuracy": 0.198853100836277,
      "num_tokens": 67010541.0,
      "step": 36315
    },
    {
      "entropy": 5.403614568710327,
      "epoch": 3.0513337534131484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004076728811901312,
      "loss": 5.0057,
      "mean_token_accuracy": 0.20477484315633773,
      "num_tokens": 67019387.0,
      "step": 36320
    },
    {
      "entropy": 5.493142604827881,
      "epoch": 3.051753833228313,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004076486935770075,
      "loss": 5.0158,
      "mean_token_accuracy": 0.20030419975519181,
      "num_tokens": 67027902.0,
      "step": 36325
    },
    {
      "entropy": 5.476535034179688,
      "epoch": 3.0521739130434784,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004076245036140805,
      "loss": 5.0314,
      "mean_token_accuracy": 0.195711188018322,
      "num_tokens": 67036662.0,
      "step": 36330
    },
    {
      "entropy": 5.396395063400268,
      "epoch": 3.052593992858643,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004076003113017788,
      "loss": 4.9583,
      "mean_token_accuracy": 0.20590460896492005,
      "num_tokens": 67045960.0,
      "step": 36335
    },
    {
      "entropy": 5.425229835510254,
      "epoch": 3.053014072673808,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004075761166405307,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20269239097833633,
      "num_tokens": 67055470.0,
      "step": 36340
    },
    {
      "entropy": 5.498291301727295,
      "epoch": 3.0534341524889728,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040755191963076507,
      "loss": 5.0392,
      "mean_token_accuracy": 0.1996610701084137,
      "num_tokens": 67065133.0,
      "step": 36345
    },
    {
      "entropy": 5.4376890659332275,
      "epoch": 3.053854232304138,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004075277202729104,
      "loss": 5.0218,
      "mean_token_accuracy": 0.199187795817852,
      "num_tokens": 67075016.0,
      "step": 36350
    },
    {
      "entropy": 5.44080080986023,
      "epoch": 3.054274312119303,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040750351856739544,
      "loss": 4.9714,
      "mean_token_accuracy": 0.2010812669992447,
      "num_tokens": 67084789.0,
      "step": 36355
    },
    {
      "entropy": 5.376031160354614,
      "epoch": 3.0546943919344676,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004074793145146488,
      "loss": 4.9625,
      "mean_token_accuracy": 0.203558549284935,
      "num_tokens": 67094078.0,
      "step": 36360
    },
    {
      "entropy": 5.488125324249268,
      "epoch": 3.0551144717496324,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004074551081150994,
      "loss": 5.0227,
      "mean_token_accuracy": 0.1990969032049179,
      "num_tokens": 67103805.0,
      "step": 36365
    },
    {
      "entropy": 5.45218300819397,
      "epoch": 3.055534551564797,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040743089936917594,
      "loss": 5.0363,
      "mean_token_accuracy": 0.202442866563797,
      "num_tokens": 67113066.0,
      "step": 36370
    },
    {
      "entropy": 5.437224292755127,
      "epoch": 3.0559546313799624,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040740668827730724,
      "loss": 5.0145,
      "mean_token_accuracy": 0.19886055141687392,
      "num_tokens": 67122926.0,
      "step": 36375
    },
    {
      "entropy": 5.538024377822876,
      "epoch": 3.056374711195127,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040738247483992234,
      "loss": 5.0102,
      "mean_token_accuracy": 0.1927841931581497,
      "num_tokens": 67131813.0,
      "step": 36380
    },
    {
      "entropy": 5.470226860046386,
      "epoch": 3.056794791010292,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040735825905745,
      "loss": 5.0406,
      "mean_token_accuracy": 0.19794073700904846,
      "num_tokens": 67139991.0,
      "step": 36385
    },
    {
      "entropy": 5.519979047775268,
      "epoch": 3.0572148708254567,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004073340409303193,
      "loss": 5.1825,
      "mean_token_accuracy": 0.18844178020954133,
      "num_tokens": 67150389.0,
      "step": 36390
    },
    {
      "entropy": 5.389359664916992,
      "epoch": 3.0576349506406215,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040730982045895914,
      "loss": 4.9107,
      "mean_token_accuracy": 0.20329734981060027,
      "num_tokens": 67159222.0,
      "step": 36395
    },
    {
      "entropy": 5.42185001373291,
      "epoch": 3.0580550304557867,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040728559764379867,
      "loss": 5.007,
      "mean_token_accuracy": 0.20007191002368926,
      "num_tokens": 67168101.0,
      "step": 36400
    },
    {
      "entropy": 5.524215602874756,
      "epoch": 3.0584751102709515,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040726137248526704,
      "loss": 5.0995,
      "mean_token_accuracy": 0.1924113929271698,
      "num_tokens": 67177959.0,
      "step": 36405
    },
    {
      "entropy": 5.428938579559326,
      "epoch": 3.0588951900861163,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040723714498379316,
      "loss": 4.9905,
      "mean_token_accuracy": 0.2096765324473381,
      "num_tokens": 67187976.0,
      "step": 36410
    },
    {
      "entropy": 5.499889850616455,
      "epoch": 3.059315269901281,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004072129151398064,
      "loss": 5.0721,
      "mean_token_accuracy": 0.19412610083818435,
      "num_tokens": 67198781.0,
      "step": 36415
    },
    {
      "entropy": 5.56098165512085,
      "epoch": 3.0597353497164463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040718868295373594,
      "loss": 5.207,
      "mean_token_accuracy": 0.18327178210020065,
      "num_tokens": 67208153.0,
      "step": 36420
    },
    {
      "entropy": 5.494327163696289,
      "epoch": 3.060155429531611,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040716444842601107,
      "loss": 5.0954,
      "mean_token_accuracy": 0.19706928879022598,
      "num_tokens": 67217899.0,
      "step": 36425
    },
    {
      "entropy": 5.5180456161499025,
      "epoch": 3.060575509346776,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00040714021155706094,
      "loss": 5.0936,
      "mean_token_accuracy": 0.19380997121334076,
      "num_tokens": 67227172.0,
      "step": 36430
    },
    {
      "entropy": 5.482501459121704,
      "epoch": 3.0609955891619407,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040711597234731507,
      "loss": 5.027,
      "mean_token_accuracy": 0.20205966383218765,
      "num_tokens": 67236501.0,
      "step": 36435
    },
    {
      "entropy": 5.482695531845093,
      "epoch": 3.0614156689771055,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004070917307972028,
      "loss": 5.007,
      "mean_token_accuracy": 0.20034103095531464,
      "num_tokens": 67246227.0,
      "step": 36440
    },
    {
      "entropy": 5.507821130752563,
      "epoch": 3.0618357487922707,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004070674869071535,
      "loss": 5.0412,
      "mean_token_accuracy": 0.20018863826990127,
      "num_tokens": 67255051.0,
      "step": 36445
    },
    {
      "entropy": 5.458800554275513,
      "epoch": 3.0622558286074355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040704324067759667,
      "loss": 5.0442,
      "mean_token_accuracy": 0.19792485982179642,
      "num_tokens": 67264832.0,
      "step": 36450
    },
    {
      "entropy": 5.4975098133087155,
      "epoch": 3.0626759084226003,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040701899210896186,
      "loss": 5.0595,
      "mean_token_accuracy": 0.19657846093177794,
      "num_tokens": 67274212.0,
      "step": 36455
    },
    {
      "entropy": 5.492067956924439,
      "epoch": 3.063095988237765,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004069947412016786,
      "loss": 5.0032,
      "mean_token_accuracy": 0.19702027440071107,
      "num_tokens": 67282277.0,
      "step": 36460
    },
    {
      "entropy": 5.490545225143433,
      "epoch": 3.06351606805293,
      "grad_norm": 1.0,
      "learning_rate": 0.00040697048795617643,
      "loss": 5.0849,
      "mean_token_accuracy": 0.19205211400985717,
      "num_tokens": 67291768.0,
      "step": 36465
    },
    {
      "entropy": 5.428443813323975,
      "epoch": 3.063936147868095,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040694623237288503,
      "loss": 4.9963,
      "mean_token_accuracy": 0.19990718811750413,
      "num_tokens": 67300026.0,
      "step": 36470
    },
    {
      "entropy": 5.420484638214111,
      "epoch": 3.06435622768326,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040692197445223405,
      "loss": 5.0436,
      "mean_token_accuracy": 0.1991005063056946,
      "num_tokens": 67308924.0,
      "step": 36475
    },
    {
      "entropy": 5.512647438049316,
      "epoch": 3.0647763074984247,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004068977141946532,
      "loss": 5.0726,
      "mean_token_accuracy": 0.19748273193836213,
      "num_tokens": 67317938.0,
      "step": 36480
    },
    {
      "entropy": 5.527362251281739,
      "epoch": 3.0651963873135895,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00040687345160057235,
      "loss": 5.0836,
      "mean_token_accuracy": 0.18659288436174393,
      "num_tokens": 67328194.0,
      "step": 36485
    },
    {
      "entropy": 5.506193017959594,
      "epoch": 3.0656164671287547,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040684918667042107,
      "loss": 5.0436,
      "mean_token_accuracy": 0.197349613904953,
      "num_tokens": 67337824.0,
      "step": 36490
    },
    {
      "entropy": 5.4259544849395756,
      "epoch": 3.0660365469439195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040682491940462944,
      "loss": 4.9917,
      "mean_token_accuracy": 0.19825389236211777,
      "num_tokens": 67347366.0,
      "step": 36495
    },
    {
      "entropy": 5.433342599868775,
      "epoch": 3.0664566267590843,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040680064980362727,
      "loss": 4.9797,
      "mean_token_accuracy": 0.19954679012298585,
      "num_tokens": 67356348.0,
      "step": 36500
    },
    {
      "entropy": 5.432817792892456,
      "epoch": 3.066876706574249,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040677637786784435,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2113758623600006,
      "num_tokens": 67364524.0,
      "step": 36505
    },
    {
      "entropy": 5.411489772796631,
      "epoch": 3.067296786389414,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040675210359771084,
      "loss": 4.8921,
      "mean_token_accuracy": 0.2113875165581703,
      "num_tokens": 67373652.0,
      "step": 36510
    },
    {
      "entropy": 5.350401782989502,
      "epoch": 3.067716866204579,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004067278269936566,
      "loss": 4.9389,
      "mean_token_accuracy": 0.20976604223251344,
      "num_tokens": 67381939.0,
      "step": 36515
    },
    {
      "entropy": 5.3994612216949465,
      "epoch": 3.068136946019744,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004067035480561118,
      "loss": 5.1284,
      "mean_token_accuracy": 0.18460632115602493,
      "num_tokens": 67391140.0,
      "step": 36520
    },
    {
      "entropy": 5.468710899353027,
      "epoch": 3.0685570258349086,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004066792667855064,
      "loss": 5.037,
      "mean_token_accuracy": 0.20228205025196075,
      "num_tokens": 67400189.0,
      "step": 36525
    },
    {
      "entropy": 5.460496997833252,
      "epoch": 3.0689771056500734,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004066549831822706,
      "loss": 4.9431,
      "mean_token_accuracy": 0.20483127981424332,
      "num_tokens": 67409250.0,
      "step": 36530
    },
    {
      "entropy": 5.457351160049439,
      "epoch": 3.069397185465238,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004066306972468346,
      "loss": 5.0362,
      "mean_token_accuracy": 0.20288028717041015,
      "num_tokens": 67418233.0,
      "step": 36535
    },
    {
      "entropy": 5.585244464874267,
      "epoch": 3.0698172652804034,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004066064089796286,
      "loss": 5.1255,
      "mean_token_accuracy": 0.20367612093687057,
      "num_tokens": 67427580.0,
      "step": 36540
    },
    {
      "entropy": 5.431986093521118,
      "epoch": 3.0702373450955682,
      "grad_norm": 1.125,
      "learning_rate": 0.0004065821183810828,
      "loss": 5.022,
      "mean_token_accuracy": 0.20153724104166032,
      "num_tokens": 67436032.0,
      "step": 36545
    },
    {
      "entropy": 5.498827981948852,
      "epoch": 3.070657424910733,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004065578254516275,
      "loss": 5.1049,
      "mean_token_accuracy": 0.19885776191949844,
      "num_tokens": 67445073.0,
      "step": 36550
    },
    {
      "entropy": 5.493912696838379,
      "epoch": 3.071077504725898,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004065335301916931,
      "loss": 5.1073,
      "mean_token_accuracy": 0.19390051513910295,
      "num_tokens": 67453801.0,
      "step": 36555
    },
    {
      "entropy": 5.549066257476807,
      "epoch": 3.0714975845410626,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004065092326017099,
      "loss": 5.1122,
      "mean_token_accuracy": 0.2007450744509697,
      "num_tokens": 67462613.0,
      "step": 36560
    },
    {
      "entropy": 5.572985076904297,
      "epoch": 3.071917664356228,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004064849326821084,
      "loss": 5.0667,
      "mean_token_accuracy": 0.18840353190898895,
      "num_tokens": 67472079.0,
      "step": 36565
    },
    {
      "entropy": 5.508934116363525,
      "epoch": 3.0723377441713926,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000406460630433319,
      "loss": 5.0174,
      "mean_token_accuracy": 0.1933622181415558,
      "num_tokens": 67481407.0,
      "step": 36570
    },
    {
      "entropy": 5.498258209228515,
      "epoch": 3.0727578239865574,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040643632585577225,
      "loss": 5.0772,
      "mean_token_accuracy": 0.1918393611907959,
      "num_tokens": 67490893.0,
      "step": 36575
    },
    {
      "entropy": 5.421078252792358,
      "epoch": 3.073177903801722,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040641201894989865,
      "loss": 4.9874,
      "mean_token_accuracy": 0.19505516290664673,
      "num_tokens": 67499624.0,
      "step": 36580
    },
    {
      "entropy": 5.509522151947022,
      "epoch": 3.0735979836168874,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004063877097161288,
      "loss": 5.0936,
      "mean_token_accuracy": 0.20624389350414277,
      "num_tokens": 67508353.0,
      "step": 36585
    },
    {
      "entropy": 5.487695550918579,
      "epoch": 3.074018063432052,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004063633981548933,
      "loss": 5.0267,
      "mean_token_accuracy": 0.201602965593338,
      "num_tokens": 67517498.0,
      "step": 36590
    },
    {
      "entropy": 5.520168161392212,
      "epoch": 3.074438143247217,
      "grad_norm": 1.125,
      "learning_rate": 0.0004063390842666229,
      "loss": 5.0415,
      "mean_token_accuracy": 0.20227625370025634,
      "num_tokens": 67526246.0,
      "step": 36595
    },
    {
      "entropy": 5.51905574798584,
      "epoch": 3.0748582230623818,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004063147680517483,
      "loss": 5.0527,
      "mean_token_accuracy": 0.19741144329309462,
      "num_tokens": 67536283.0,
      "step": 36600
    },
    {
      "entropy": 5.454864406585694,
      "epoch": 3.0752783028775466,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040629044951070005,
      "loss": 5.0293,
      "mean_token_accuracy": 0.19936929792165756,
      "num_tokens": 67545797.0,
      "step": 36605
    },
    {
      "entropy": 5.505782794952393,
      "epoch": 3.075698382692712,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040626612864390924,
      "loss": 5.1517,
      "mean_token_accuracy": 0.19405818730592728,
      "num_tokens": 67555257.0,
      "step": 36610
    },
    {
      "entropy": 5.428848075866699,
      "epoch": 3.0761184625078766,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040624180545180644,
      "loss": 4.9,
      "mean_token_accuracy": 0.2056122377514839,
      "num_tokens": 67563490.0,
      "step": 36615
    },
    {
      "entropy": 5.395373725891114,
      "epoch": 3.0765385423230414,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040621747993482274,
      "loss": 4.921,
      "mean_token_accuracy": 0.2090475857257843,
      "num_tokens": 67571742.0,
      "step": 36620
    },
    {
      "entropy": 5.454728460311889,
      "epoch": 3.076958622138206,
      "grad_norm": 1.0,
      "learning_rate": 0.0004061931520933889,
      "loss": 5.028,
      "mean_token_accuracy": 0.20148227363824844,
      "num_tokens": 67581577.0,
      "step": 36625
    },
    {
      "entropy": 5.5149893283844,
      "epoch": 3.077378701953371,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040616882192793586,
      "loss": 5.0783,
      "mean_token_accuracy": 0.193037049472332,
      "num_tokens": 67590403.0,
      "step": 36630
    },
    {
      "entropy": 5.503889083862305,
      "epoch": 3.077798781768536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004061444894388947,
      "loss": 5.0005,
      "mean_token_accuracy": 0.2021706596016884,
      "num_tokens": 67599342.0,
      "step": 36635
    },
    {
      "entropy": 5.433884239196777,
      "epoch": 3.078218861583701,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004061201546266965,
      "loss": 4.983,
      "mean_token_accuracy": 0.199322646856308,
      "num_tokens": 67608813.0,
      "step": 36640
    },
    {
      "entropy": 5.424668073654175,
      "epoch": 3.0786389413988657,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004060958174917722,
      "loss": 5.0443,
      "mean_token_accuracy": 0.19747580587863922,
      "num_tokens": 67617996.0,
      "step": 36645
    },
    {
      "entropy": 5.454495477676391,
      "epoch": 3.0790590212140305,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004060714780345531,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20476406961679458,
      "num_tokens": 67626995.0,
      "step": 36650
    },
    {
      "entropy": 5.4262535572052,
      "epoch": 3.0794791010291958,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004060471362554702,
      "loss": 4.9194,
      "mean_token_accuracy": 0.20546222627162933,
      "num_tokens": 67636308.0,
      "step": 36655
    },
    {
      "entropy": 5.5008683681488035,
      "epoch": 3.0798991808443605,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004060227921549548,
      "loss": 5.0493,
      "mean_token_accuracy": 0.19389349222183228,
      "num_tokens": 67644420.0,
      "step": 36660
    },
    {
      "entropy": 5.5201128959655765,
      "epoch": 3.0803192606595253,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040599844573343803,
      "loss": 5.1336,
      "mean_token_accuracy": 0.18817951679229736,
      "num_tokens": 67654003.0,
      "step": 36665
    },
    {
      "entropy": 5.461288642883301,
      "epoch": 3.08073934047469,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004059740969913513,
      "loss": 4.9934,
      "mean_token_accuracy": 0.2016213893890381,
      "num_tokens": 67664295.0,
      "step": 36670
    },
    {
      "entropy": 5.433550071716309,
      "epoch": 3.081159420289855,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004059497459291259,
      "loss": 5.0011,
      "mean_token_accuracy": 0.19869534522295,
      "num_tokens": 67673494.0,
      "step": 36675
    },
    {
      "entropy": 5.489875316619873,
      "epoch": 3.08157950010502,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040592539254719317,
      "loss": 5.0797,
      "mean_token_accuracy": 0.1955941155552864,
      "num_tokens": 67682868.0,
      "step": 36680
    },
    {
      "entropy": 5.5282491683959964,
      "epoch": 3.081999579920185,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004059010368459845,
      "loss": 4.9985,
      "mean_token_accuracy": 0.2084253266453743,
      "num_tokens": 67692990.0,
      "step": 36685
    },
    {
      "entropy": 5.512255048751831,
      "epoch": 3.0824196597353497,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004058766788259314,
      "loss": 5.0762,
      "mean_token_accuracy": 0.1932537168264389,
      "num_tokens": 67702896.0,
      "step": 36690
    },
    {
      "entropy": 5.433644485473633,
      "epoch": 3.0828397395505145,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004058523184874653,
      "loss": 5.0218,
      "mean_token_accuracy": 0.21145516782999038,
      "num_tokens": 67712261.0,
      "step": 36695
    },
    {
      "entropy": 5.422098827362061,
      "epoch": 3.0832598193656793,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040582795583101773,
      "loss": 4.9382,
      "mean_token_accuracy": 0.2036696493625641,
      "num_tokens": 67721159.0,
      "step": 36700
    },
    {
      "entropy": 5.417815685272217,
      "epoch": 3.0836798991808445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040580359085702036,
      "loss": 4.927,
      "mean_token_accuracy": 0.21298697739839553,
      "num_tokens": 67731093.0,
      "step": 36705
    },
    {
      "entropy": 5.521150922775268,
      "epoch": 3.0840999789960093,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004057792235659046,
      "loss": 5.0572,
      "mean_token_accuracy": 0.19680135399103166,
      "num_tokens": 67739900.0,
      "step": 36710
    },
    {
      "entropy": 5.59428129196167,
      "epoch": 3.084520058811174,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004057548539581023,
      "loss": 5.1003,
      "mean_token_accuracy": 0.18999252766370772,
      "num_tokens": 67748303.0,
      "step": 36715
    },
    {
      "entropy": 5.46265664100647,
      "epoch": 3.084940138626339,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004057304820340452,
      "loss": 5.0225,
      "mean_token_accuracy": 0.1973123997449875,
      "num_tokens": 67758055.0,
      "step": 36720
    },
    {
      "entropy": 5.479275798797607,
      "epoch": 3.085360218441504,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040570610779416485,
      "loss": 5.0887,
      "mean_token_accuracy": 0.19064458459615707,
      "num_tokens": 67767340.0,
      "step": 36725
    },
    {
      "entropy": 5.479873609542847,
      "epoch": 3.085780298256669,
      "grad_norm": 1.0,
      "learning_rate": 0.0004056817312388931,
      "loss": 5.032,
      "mean_token_accuracy": 0.1959862694144249,
      "num_tokens": 67776395.0,
      "step": 36730
    },
    {
      "entropy": 5.517026329040528,
      "epoch": 3.0862003780718337,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004056573523686617,
      "loss": 4.9938,
      "mean_token_accuracy": 0.19871689677238463,
      "num_tokens": 67785158.0,
      "step": 36735
    },
    {
      "entropy": 5.50283374786377,
      "epoch": 3.0866204578869985,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004056329711839027,
      "loss": 5.0153,
      "mean_token_accuracy": 0.205469910800457,
      "num_tokens": 67794137.0,
      "step": 36740
    },
    {
      "entropy": 5.46326642036438,
      "epoch": 3.0870405377021632,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004056085876850479,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20295283645391465,
      "num_tokens": 67802767.0,
      "step": 36745
    },
    {
      "entropy": 5.433736562728882,
      "epoch": 3.0874606175173285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040558420187252907,
      "loss": 4.9832,
      "mean_token_accuracy": 0.2078685626387596,
      "num_tokens": 67811684.0,
      "step": 36750
    },
    {
      "entropy": 5.471851873397827,
      "epoch": 3.0878806973324933,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040555981374677845,
      "loss": 5.0186,
      "mean_token_accuracy": 0.19890465289354325,
      "num_tokens": 67821022.0,
      "step": 36755
    },
    {
      "entropy": 5.396272134780884,
      "epoch": 3.088300777147658,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004055354233082279,
      "loss": 4.9485,
      "mean_token_accuracy": 0.20528804808855056,
      "num_tokens": 67830070.0,
      "step": 36760
    },
    {
      "entropy": 5.434425115585327,
      "epoch": 3.088720856962823,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040551103055730967,
      "loss": 4.9095,
      "mean_token_accuracy": 0.20695864260196686,
      "num_tokens": 67839872.0,
      "step": 36765
    },
    {
      "entropy": 5.5838385105133055,
      "epoch": 3.0891409367779876,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004054866354944556,
      "loss": 5.1883,
      "mean_token_accuracy": 0.18785411715507508,
      "num_tokens": 67848824.0,
      "step": 36770
    },
    {
      "entropy": 5.491209173202515,
      "epoch": 3.089561016593153,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000405462238120098,
      "loss": 4.955,
      "mean_token_accuracy": 0.20725100338459015,
      "num_tokens": 67857157.0,
      "step": 36775
    },
    {
      "entropy": 5.42017617225647,
      "epoch": 3.0899810964083176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040543783843466905,
      "loss": 5.0512,
      "mean_token_accuracy": 0.19489056020975112,
      "num_tokens": 67866969.0,
      "step": 36780
    },
    {
      "entropy": 5.478717851638794,
      "epoch": 3.0904011762234824,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040541343643860095,
      "loss": 5.0013,
      "mean_token_accuracy": 0.2033000499010086,
      "num_tokens": 67876955.0,
      "step": 36785
    },
    {
      "entropy": 5.475678396224976,
      "epoch": 3.090821256038647,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040538903213232597,
      "loss": 4.9291,
      "mean_token_accuracy": 0.20990059971809388,
      "num_tokens": 67885503.0,
      "step": 36790
    },
    {
      "entropy": 5.501703929901123,
      "epoch": 3.0912413358538124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004053646255162765,
      "loss": 5.1152,
      "mean_token_accuracy": 0.19123511463403703,
      "num_tokens": 67894382.0,
      "step": 36795
    },
    {
      "entropy": 5.436781740188598,
      "epoch": 3.0916614156689772,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040534021659088473,
      "loss": 5.0495,
      "mean_token_accuracy": 0.2059767350554466,
      "num_tokens": 67904185.0,
      "step": 36800
    },
    {
      "entropy": 5.547846937179566,
      "epoch": 3.092081495484142,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040531580535658313,
      "loss": 5.0836,
      "mean_token_accuracy": 0.20210716277360916,
      "num_tokens": 67913464.0,
      "step": 36805
    },
    {
      "entropy": 5.442916345596314,
      "epoch": 3.092501575299307,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004052913918138043,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20878523588180542,
      "num_tokens": 67921669.0,
      "step": 36810
    },
    {
      "entropy": 5.413583421707154,
      "epoch": 3.0929216551144716,
      "grad_norm": 1.125,
      "learning_rate": 0.00040526697596298035,
      "loss": 5.0131,
      "mean_token_accuracy": 0.19982743114233018,
      "num_tokens": 67931283.0,
      "step": 36815
    },
    {
      "entropy": 5.490193939208984,
      "epoch": 3.093341734929637,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004052425578045441,
      "loss": 5.0017,
      "mean_token_accuracy": 0.19446865767240523,
      "num_tokens": 67940173.0,
      "step": 36820
    },
    {
      "entropy": 5.444202041625976,
      "epoch": 3.0937618147448016,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040521813733892793,
      "loss": 5.0743,
      "mean_token_accuracy": 0.1939355030655861,
      "num_tokens": 67949652.0,
      "step": 36825
    },
    {
      "entropy": 5.43759069442749,
      "epoch": 3.0941818945599664,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040519371456656454,
      "loss": 4.9791,
      "mean_token_accuracy": 0.2112655445933342,
      "num_tokens": 67959082.0,
      "step": 36830
    },
    {
      "entropy": 5.469271802902222,
      "epoch": 3.094601974375131,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004051692894878866,
      "loss": 4.9989,
      "mean_token_accuracy": 0.20284271687269212,
      "num_tokens": 67970190.0,
      "step": 36835
    },
    {
      "entropy": 5.531145858764648,
      "epoch": 3.095022054190296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004051448621033267,
      "loss": 5.0955,
      "mean_token_accuracy": 0.1957096830010414,
      "num_tokens": 67979305.0,
      "step": 36840
    },
    {
      "entropy": 5.49684247970581,
      "epoch": 3.095442134005461,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004051204324133175,
      "loss": 5.0893,
      "mean_token_accuracy": 0.19373988956212998,
      "num_tokens": 67988980.0,
      "step": 36845
    },
    {
      "entropy": 5.428503561019897,
      "epoch": 3.095862213820626,
      "grad_norm": 1.125,
      "learning_rate": 0.000405096000418292,
      "loss": 5.081,
      "mean_token_accuracy": 0.19369404762983322,
      "num_tokens": 67998354.0,
      "step": 36850
    },
    {
      "entropy": 5.461464357376099,
      "epoch": 3.0962822936357908,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040507156611868265,
      "loss": 5.008,
      "mean_token_accuracy": 0.20232483148574829,
      "num_tokens": 68006659.0,
      "step": 36855
    },
    {
      "entropy": 5.539222240447998,
      "epoch": 3.0967023734509556,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004050471295149226,
      "loss": 5.0517,
      "mean_token_accuracy": 0.19998451918363572,
      "num_tokens": 68015282.0,
      "step": 36860
    },
    {
      "entropy": 5.523220062255859,
      "epoch": 3.0971224532661203,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004050226906074446,
      "loss": 5.1466,
      "mean_token_accuracy": 0.19015444368124007,
      "num_tokens": 68024343.0,
      "step": 36865
    },
    {
      "entropy": 5.453747177124024,
      "epoch": 3.0975425330812856,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040499824939668164,
      "loss": 4.9931,
      "mean_token_accuracy": 0.20797598659992217,
      "num_tokens": 68033342.0,
      "step": 36870
    },
    {
      "entropy": 5.458126211166382,
      "epoch": 3.0979626128964504,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040497380588306654,
      "loss": 5.0283,
      "mean_token_accuracy": 0.19600499868392945,
      "num_tokens": 68042522.0,
      "step": 36875
    },
    {
      "entropy": 5.448766374588013,
      "epoch": 3.098382692711615,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040494936006703244,
      "loss": 5.012,
      "mean_token_accuracy": 0.19825417101383208,
      "num_tokens": 68051831.0,
      "step": 36880
    },
    {
      "entropy": 5.451537179946899,
      "epoch": 3.09880277252678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004049249119490123,
      "loss": 5.0056,
      "mean_token_accuracy": 0.20350633859634398,
      "num_tokens": 68060490.0,
      "step": 36885
    },
    {
      "entropy": 5.42790207862854,
      "epoch": 3.099222852341945,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040490046152943926,
      "loss": 4.9556,
      "mean_token_accuracy": 0.2029791221022606,
      "num_tokens": 68069644.0,
      "step": 36890
    },
    {
      "entropy": 5.535838174819946,
      "epoch": 3.09964293215711,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040487600880874646,
      "loss": 5.1036,
      "mean_token_accuracy": 0.1949576884508133,
      "num_tokens": 68079310.0,
      "step": 36895
    },
    {
      "entropy": 5.51855354309082,
      "epoch": 3.1000630119722747,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040485155378736706,
      "loss": 5.0957,
      "mean_token_accuracy": 0.1946578338742256,
      "num_tokens": 68088362.0,
      "step": 36900
    },
    {
      "entropy": 5.473817491531372,
      "epoch": 3.1004830917874395,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040482709646573424,
      "loss": 5.0548,
      "mean_token_accuracy": 0.19503826647996902,
      "num_tokens": 68097219.0,
      "step": 36905
    },
    {
      "entropy": 5.490108966827393,
      "epoch": 3.1009031716026043,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040480263684428114,
      "loss": 5.0839,
      "mean_token_accuracy": 0.1978638455271721,
      "num_tokens": 68106492.0,
      "step": 36910
    },
    {
      "entropy": 5.54800271987915,
      "epoch": 3.1013232514177695,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004047781749234413,
      "loss": 5.0124,
      "mean_token_accuracy": 0.2023507073521614,
      "num_tokens": 68116899.0,
      "step": 36915
    },
    {
      "entropy": 5.408131456375122,
      "epoch": 3.1017433312329343,
      "grad_norm": 1.125,
      "learning_rate": 0.0004047537107036479,
      "loss": 4.9572,
      "mean_token_accuracy": 0.2055579975247383,
      "num_tokens": 68126183.0,
      "step": 36920
    },
    {
      "entropy": 5.477376890182495,
      "epoch": 3.102163411048099,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040472924418533427,
      "loss": 5.123,
      "mean_token_accuracy": 0.19339732676744462,
      "num_tokens": 68136023.0,
      "step": 36925
    },
    {
      "entropy": 5.52962965965271,
      "epoch": 3.102583490863264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040470477536893397,
      "loss": 5.0689,
      "mean_token_accuracy": 0.1963701695203781,
      "num_tokens": 68145794.0,
      "step": 36930
    },
    {
      "entropy": 5.4156592845916744,
      "epoch": 3.1030035706784287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004046803042548803,
      "loss": 4.952,
      "mean_token_accuracy": 0.20447930842638015,
      "num_tokens": 68154972.0,
      "step": 36935
    },
    {
      "entropy": 5.478919315338135,
      "epoch": 3.103423650493594,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040465583084360677,
      "loss": 5.1023,
      "mean_token_accuracy": 0.20143062025308608,
      "num_tokens": 68164728.0,
      "step": 36940
    },
    {
      "entropy": 5.591201686859131,
      "epoch": 3.1038437303087587,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004046313551355471,
      "loss": 5.1416,
      "mean_token_accuracy": 0.19673374891281128,
      "num_tokens": 68175071.0,
      "step": 36945
    },
    {
      "entropy": 5.53394250869751,
      "epoch": 3.1042638101239235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004046068771311346,
      "loss": 5.0547,
      "mean_token_accuracy": 0.19023548662662507,
      "num_tokens": 68184722.0,
      "step": 36950
    },
    {
      "entropy": 5.406537199020386,
      "epoch": 3.1046838899390883,
      "grad_norm": 1.046875,
      "learning_rate": 0.000404582396830803,
      "loss": 4.8853,
      "mean_token_accuracy": 0.20604415982961655,
      "num_tokens": 68193394.0,
      "step": 36955
    },
    {
      "entropy": 5.422529840469361,
      "epoch": 3.1051039697542535,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040455791423498605,
      "loss": 4.9922,
      "mean_token_accuracy": 0.19094113409519195,
      "num_tokens": 68202614.0,
      "step": 36960
    },
    {
      "entropy": 5.509254837036133,
      "epoch": 3.1055240495694183,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040453342934411733,
      "loss": 5.0213,
      "mean_token_accuracy": 0.20474166721105574,
      "num_tokens": 68211819.0,
      "step": 36965
    },
    {
      "entropy": 5.511447477340698,
      "epoch": 3.105944129384583,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004045089421586306,
      "loss": 5.0782,
      "mean_token_accuracy": 0.1906043544411659,
      "num_tokens": 68221689.0,
      "step": 36970
    },
    {
      "entropy": 5.406260871887207,
      "epoch": 3.106364209199748,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004044844526789596,
      "loss": 4.9427,
      "mean_token_accuracy": 0.20269598960876464,
      "num_tokens": 68231160.0,
      "step": 36975
    },
    {
      "entropy": 5.4710877418518065,
      "epoch": 3.1067842890149127,
      "grad_norm": 1.125,
      "learning_rate": 0.00040445996090553827,
      "loss": 5.0615,
      "mean_token_accuracy": 0.20406401306390762,
      "num_tokens": 68240240.0,
      "step": 36980
    },
    {
      "entropy": 5.412181711196899,
      "epoch": 3.107204368830078,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040443546683880034,
      "loss": 4.9286,
      "mean_token_accuracy": 0.2099050059914589,
      "num_tokens": 68248966.0,
      "step": 36985
    },
    {
      "entropy": 5.432730388641358,
      "epoch": 3.1076244486452427,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040441097047917983,
      "loss": 4.9996,
      "mean_token_accuracy": 0.19996614307165145,
      "num_tokens": 68258340.0,
      "step": 36990
    },
    {
      "entropy": 5.410147094726563,
      "epoch": 3.1080445284604075,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004043864718271106,
      "loss": 5.0375,
      "mean_token_accuracy": 0.20380311012268065,
      "num_tokens": 68266295.0,
      "step": 36995
    },
    {
      "entropy": 5.442214775085449,
      "epoch": 3.1084646082755722,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040436197088302664,
      "loss": 5.0167,
      "mean_token_accuracy": 0.2050403580069542,
      "num_tokens": 68275636.0,
      "step": 37000
    },
    {
      "entropy": 5.584439039230347,
      "epoch": 3.108884688090737,
      "grad_norm": 1.125,
      "learning_rate": 0.00040433746764736197,
      "loss": 5.246,
      "mean_token_accuracy": 0.18254782557487487,
      "num_tokens": 68285310.0,
      "step": 37005
    },
    {
      "entropy": 5.533596658706665,
      "epoch": 3.1093047679059023,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040431296212055064,
      "loss": 5.0604,
      "mean_token_accuracy": 0.19661663025617598,
      "num_tokens": 68295600.0,
      "step": 37010
    },
    {
      "entropy": 5.450110387802124,
      "epoch": 3.109724847721067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004042884543030268,
      "loss": 4.9261,
      "mean_token_accuracy": 0.20669189244508743,
      "num_tokens": 68304165.0,
      "step": 37015
    },
    {
      "entropy": 5.420079469680786,
      "epoch": 3.110144927536232,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004042639441952245,
      "loss": 5.0332,
      "mean_token_accuracy": 0.20174043476581574,
      "num_tokens": 68313078.0,
      "step": 37020
    },
    {
      "entropy": 5.543984508514404,
      "epoch": 3.1105650073513966,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040423943179757803,
      "loss": 5.0882,
      "mean_token_accuracy": 0.19403094947338104,
      "num_tokens": 68322346.0,
      "step": 37025
    },
    {
      "entropy": 5.465675020217896,
      "epoch": 3.110985087166562,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004042149171105216,
      "loss": 4.9853,
      "mean_token_accuracy": 0.19885222762823104,
      "num_tokens": 68331872.0,
      "step": 37030
    },
    {
      "entropy": 5.624642276763916,
      "epoch": 3.1114051669817266,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004041904001344894,
      "loss": 5.1592,
      "mean_token_accuracy": 0.19019172787666322,
      "num_tokens": 68340861.0,
      "step": 37035
    },
    {
      "entropy": 5.448081684112549,
      "epoch": 3.1118252467968914,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040416588086991586,
      "loss": 4.9831,
      "mean_token_accuracy": 0.20846639424562455,
      "num_tokens": 68349873.0,
      "step": 37040
    },
    {
      "entropy": 5.447339010238648,
      "epoch": 3.112245326612056,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040414135931723515,
      "loss": 5.0689,
      "mean_token_accuracy": 0.19653595238924026,
      "num_tokens": 68360052.0,
      "step": 37045
    },
    {
      "entropy": 5.395950078964233,
      "epoch": 3.112665406427221,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040411683547688184,
      "loss": 4.9177,
      "mean_token_accuracy": 0.20893668830394746,
      "num_tokens": 68369464.0,
      "step": 37050
    },
    {
      "entropy": 5.5215537548065186,
      "epoch": 3.1130854862423862,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040409230934929026,
      "loss": 5.1355,
      "mean_token_accuracy": 0.19260660260915757,
      "num_tokens": 68378945.0,
      "step": 37055
    },
    {
      "entropy": 5.498246955871582,
      "epoch": 3.113505566057551,
      "grad_norm": 1.125,
      "learning_rate": 0.00040406778093489485,
      "loss": 5.0978,
      "mean_token_accuracy": 0.19650754630565642,
      "num_tokens": 68388460.0,
      "step": 37060
    },
    {
      "entropy": 5.527061176300049,
      "epoch": 3.113925645872716,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040404325023413016,
      "loss": 5.1508,
      "mean_token_accuracy": 0.18525511175394058,
      "num_tokens": 68397570.0,
      "step": 37065
    },
    {
      "entropy": 5.557853317260742,
      "epoch": 3.1143457256878806,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040401871724743085,
      "loss": 5.0637,
      "mean_token_accuracy": 0.1912556990981102,
      "num_tokens": 68407073.0,
      "step": 37070
    },
    {
      "entropy": 5.475820541381836,
      "epoch": 3.1147658055030454,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040399418197523135,
      "loss": 5.0587,
      "mean_token_accuracy": 0.19238010793924332,
      "num_tokens": 68416711.0,
      "step": 37075
    },
    {
      "entropy": 5.4995192527771,
      "epoch": 3.1151858853182106,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004039696444179663,
      "loss": 5.1373,
      "mean_token_accuracy": 0.19729681462049484,
      "num_tokens": 68426272.0,
      "step": 37080
    },
    {
      "entropy": 5.49731411933899,
      "epoch": 3.1156059651333754,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040394510457607036,
      "loss": 4.9737,
      "mean_token_accuracy": 0.2063100293278694,
      "num_tokens": 68435581.0,
      "step": 37085
    },
    {
      "entropy": 5.5413929462432865,
      "epoch": 3.11602604494854,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004039205624499784,
      "loss": 5.1268,
      "mean_token_accuracy": 0.20337017327547074,
      "num_tokens": 68444419.0,
      "step": 37090
    },
    {
      "entropy": 5.433540868759155,
      "epoch": 3.116446124763705,
      "grad_norm": 1.109375,
      "learning_rate": 0.000403896018040125,
      "loss": 4.9927,
      "mean_token_accuracy": 0.19803901314735411,
      "num_tokens": 68453777.0,
      "step": 37095
    },
    {
      "entropy": 5.532762098312378,
      "epoch": 3.11686620457887,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000403871471346945,
      "loss": 5.1141,
      "mean_token_accuracy": 0.1913565769791603,
      "num_tokens": 68462827.0,
      "step": 37100
    },
    {
      "entropy": 5.5130500316619875,
      "epoch": 3.117286284394035,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004038469223708733,
      "loss": 5.0682,
      "mean_token_accuracy": 0.20346516519784927,
      "num_tokens": 68472431.0,
      "step": 37105
    },
    {
      "entropy": 5.454656457901001,
      "epoch": 3.1177063642091998,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040382237111234466,
      "loss": 4.974,
      "mean_token_accuracy": 0.20463022887706755,
      "num_tokens": 68481405.0,
      "step": 37110
    },
    {
      "entropy": 5.543313837051391,
      "epoch": 3.1181264440243646,
      "grad_norm": 1.09375,
      "learning_rate": 0.000403797817571794,
      "loss": 5.046,
      "mean_token_accuracy": 0.19794756174087524,
      "num_tokens": 68490233.0,
      "step": 37115
    },
    {
      "entropy": 5.473341941833496,
      "epoch": 3.1185465238395293,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040377326174965647,
      "loss": 5.0447,
      "mean_token_accuracy": 0.19759988635778428,
      "num_tokens": 68499767.0,
      "step": 37120
    },
    {
      "entropy": 5.462639331817627,
      "epoch": 3.1189666036546946,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004037487036463668,
      "loss": 4.9741,
      "mean_token_accuracy": 0.20605372339487077,
      "num_tokens": 68509220.0,
      "step": 37125
    },
    {
      "entropy": 5.461697387695312,
      "epoch": 3.1193866834698594,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004037241432623602,
      "loss": 5.0077,
      "mean_token_accuracy": 0.20254812091588975,
      "num_tokens": 68518802.0,
      "step": 37130
    },
    {
      "entropy": 5.479873561859131,
      "epoch": 3.119806763285024,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040369958059807167,
      "loss": 5.0576,
      "mean_token_accuracy": 0.19892704039812087,
      "num_tokens": 68528588.0,
      "step": 37135
    },
    {
      "entropy": 5.417446231842041,
      "epoch": 3.120226843100189,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004036750156539363,
      "loss": 4.9728,
      "mean_token_accuracy": 0.2033772811293602,
      "num_tokens": 68537824.0,
      "step": 37140
    },
    {
      "entropy": 5.452915525436401,
      "epoch": 3.1206469229153537,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040365044843038937,
      "loss": 5.001,
      "mean_token_accuracy": 0.20527392625808716,
      "num_tokens": 68547500.0,
      "step": 37145
    },
    {
      "entropy": 5.506550931930542,
      "epoch": 3.121067002730519,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004036258789278659,
      "loss": 5.0891,
      "mean_token_accuracy": 0.20159732699394226,
      "num_tokens": 68557416.0,
      "step": 37150
    },
    {
      "entropy": 5.467936086654663,
      "epoch": 3.1214870825456837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004036013071468013,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20207619071006774,
      "num_tokens": 68565857.0,
      "step": 37155
    },
    {
      "entropy": 5.510158491134644,
      "epoch": 3.1219071623608485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004035767330876307,
      "loss": 5.104,
      "mean_token_accuracy": 0.198785263299942,
      "num_tokens": 68576093.0,
      "step": 37160
    },
    {
      "entropy": 5.454570722579956,
      "epoch": 3.1223272421760133,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004035521567507895,
      "loss": 5.0274,
      "mean_token_accuracy": 0.19911776930093766,
      "num_tokens": 68585925.0,
      "step": 37165
    },
    {
      "entropy": 5.399524736404419,
      "epoch": 3.122747321991178,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000403527578136713,
      "loss": 4.9979,
      "mean_token_accuracy": 0.19447092562913895,
      "num_tokens": 68594832.0,
      "step": 37170
    },
    {
      "entropy": 5.482207632064819,
      "epoch": 3.1231674018063433,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004035029972458367,
      "loss": 5.1365,
      "mean_token_accuracy": 0.19588345736265184,
      "num_tokens": 68603901.0,
      "step": 37175
    },
    {
      "entropy": 5.452970504760742,
      "epoch": 3.123587481621508,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000403478414078596,
      "loss": 4.9685,
      "mean_token_accuracy": 0.20110944360494615,
      "num_tokens": 68612622.0,
      "step": 37180
    },
    {
      "entropy": 5.536918354034424,
      "epoch": 3.124007561436673,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004034538286354263,
      "loss": 5.1299,
      "mean_token_accuracy": 0.19125060886144638,
      "num_tokens": 68621896.0,
      "step": 37185
    },
    {
      "entropy": 5.591430521011352,
      "epoch": 3.1244276412518377,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040342924091676306,
      "loss": 5.0748,
      "mean_token_accuracy": 0.19949036538600923,
      "num_tokens": 68630886.0,
      "step": 37190
    },
    {
      "entropy": 5.398139476776123,
      "epoch": 3.124847721067003,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040340465092304214,
      "loss": 5.012,
      "mean_token_accuracy": 0.19966126084327698,
      "num_tokens": 68640621.0,
      "step": 37195
    },
    {
      "entropy": 5.491727638244629,
      "epoch": 3.1252678008821677,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040338005865469883,
      "loss": 5.1622,
      "mean_token_accuracy": 0.19160334169864654,
      "num_tokens": 68650209.0,
      "step": 37200
    },
    {
      "entropy": 5.561231088638306,
      "epoch": 3.1256878806973325,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040335546411216893,
      "loss": 5.0968,
      "mean_token_accuracy": 0.18960471004247664,
      "num_tokens": 68659728.0,
      "step": 37205
    },
    {
      "entropy": 5.526576662063599,
      "epoch": 3.1261079605124973,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004033308672958881,
      "loss": 5.0585,
      "mean_token_accuracy": 0.20247759968042373,
      "num_tokens": 68668691.0,
      "step": 37210
    },
    {
      "entropy": 5.334374570846558,
      "epoch": 3.126528040327662,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000403306268206292,
      "loss": 4.9589,
      "mean_token_accuracy": 0.19726037979125977,
      "num_tokens": 68678006.0,
      "step": 37215
    },
    {
      "entropy": 5.476800441741943,
      "epoch": 3.1269481201428273,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040328166684381646,
      "loss": 5.1626,
      "mean_token_accuracy": 0.1975613996386528,
      "num_tokens": 68687836.0,
      "step": 37220
    },
    {
      "entropy": 5.583286094665527,
      "epoch": 3.127368199957992,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004032570632088972,
      "loss": 5.0364,
      "mean_token_accuracy": 0.19824420660734177,
      "num_tokens": 68697374.0,
      "step": 37225
    },
    {
      "entropy": 5.523312187194824,
      "epoch": 3.127788279773157,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040323245730197017,
      "loss": 5.0847,
      "mean_token_accuracy": 0.19852295368909836,
      "num_tokens": 68707331.0,
      "step": 37230
    },
    {
      "entropy": 5.477325296401977,
      "epoch": 3.1282083595883217,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004032078491234711,
      "loss": 4.9705,
      "mean_token_accuracy": 0.20415402352809905,
      "num_tokens": 68716073.0,
      "step": 37235
    },
    {
      "entropy": 5.4976249694824215,
      "epoch": 3.128628439403487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040318323867383606,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20608970075845717,
      "num_tokens": 68725074.0,
      "step": 37240
    },
    {
      "entropy": 5.437854194641114,
      "epoch": 3.1290485192186517,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040315862595350087,
      "loss": 5.0601,
      "mean_token_accuracy": 0.19146383851766585,
      "num_tokens": 68734099.0,
      "step": 37245
    },
    {
      "entropy": 5.4403074264526365,
      "epoch": 3.1294685990338165,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004031340109629017,
      "loss": 5.0127,
      "mean_token_accuracy": 0.2069549173116684,
      "num_tokens": 68743352.0,
      "step": 37250
    },
    {
      "entropy": 5.5049034595489506,
      "epoch": 3.1298886788489813,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040310939370247445,
      "loss": 5.0654,
      "mean_token_accuracy": 0.19824328273534775,
      "num_tokens": 68753468.0,
      "step": 37255
    },
    {
      "entropy": 5.494055461883545,
      "epoch": 3.130308758664146,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004030847741726553,
      "loss": 4.9836,
      "mean_token_accuracy": 0.20309716910123826,
      "num_tokens": 68761805.0,
      "step": 37260
    },
    {
      "entropy": 5.465588903427124,
      "epoch": 3.1307288384793113,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040306015237388034,
      "loss": 5.0043,
      "mean_token_accuracy": 0.20025774538517,
      "num_tokens": 68771074.0,
      "step": 37265
    },
    {
      "entropy": 5.456435823440552,
      "epoch": 3.131148918294476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040303552830658575,
      "loss": 5.0461,
      "mean_token_accuracy": 0.19194353371858597,
      "num_tokens": 68780302.0,
      "step": 37270
    },
    {
      "entropy": 5.457486009597778,
      "epoch": 3.131568998109641,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004030109019712076,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20150897949934005,
      "num_tokens": 68789501.0,
      "step": 37275
    },
    {
      "entropy": 5.434183740615845,
      "epoch": 3.1319890779248056,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040298627336818235,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2073083773255348,
      "num_tokens": 68798425.0,
      "step": 37280
    },
    {
      "entropy": 5.537635612487793,
      "epoch": 3.1324091577399704,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040296164249794614,
      "loss": 5.1698,
      "mean_token_accuracy": 0.1891252353787422,
      "num_tokens": 68808425.0,
      "step": 37285
    },
    {
      "entropy": 5.408549451828003,
      "epoch": 3.1328292375551356,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004029370093609352,
      "loss": 5.0274,
      "mean_token_accuracy": 0.18948356807231903,
      "num_tokens": 68817394.0,
      "step": 37290
    },
    {
      "entropy": 5.484293270111084,
      "epoch": 3.1332493173703004,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040291237395758615,
      "loss": 5.0726,
      "mean_token_accuracy": 0.1966031476855278,
      "num_tokens": 68826675.0,
      "step": 37295
    },
    {
      "entropy": 5.506428050994873,
      "epoch": 3.133669397185465,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040288773628833525,
      "loss": 5.048,
      "mean_token_accuracy": 0.1991967424750328,
      "num_tokens": 68835503.0,
      "step": 37300
    },
    {
      "entropy": 5.520005130767823,
      "epoch": 3.13408947700063,
      "grad_norm": 1.0,
      "learning_rate": 0.00040286309635361887,
      "loss": 5.0776,
      "mean_token_accuracy": 0.19701679348945617,
      "num_tokens": 68845280.0,
      "step": 37305
    },
    {
      "entropy": 5.478207159042358,
      "epoch": 3.134509556815795,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040283845415387367,
      "loss": 5.0701,
      "mean_token_accuracy": 0.19765228480100633,
      "num_tokens": 68854549.0,
      "step": 37310
    },
    {
      "entropy": 5.44771933555603,
      "epoch": 3.13492963663096,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040281380968953603,
      "loss": 5.0063,
      "mean_token_accuracy": 0.21046670377254487,
      "num_tokens": 68863890.0,
      "step": 37315
    },
    {
      "entropy": 5.600480413436889,
      "epoch": 3.135349716446125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040278916296104263,
      "loss": 5.1984,
      "mean_token_accuracy": 0.18909226208925248,
      "num_tokens": 68873279.0,
      "step": 37320
    },
    {
      "entropy": 5.514921236038208,
      "epoch": 3.1357697962612896,
      "grad_norm": 1.125,
      "learning_rate": 0.0004027645139688299,
      "loss": 4.9407,
      "mean_token_accuracy": 0.20383819192647934,
      "num_tokens": 68881027.0,
      "step": 37325
    },
    {
      "entropy": 5.399106216430664,
      "epoch": 3.1361898760764544,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004027398627133347,
      "loss": 4.9671,
      "mean_token_accuracy": 0.19745221734046936,
      "num_tokens": 68889730.0,
      "step": 37330
    },
    {
      "entropy": 5.496010971069336,
      "epoch": 3.136609955891619,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004027152091949936,
      "loss": 5.0455,
      "mean_token_accuracy": 0.20406874418258666,
      "num_tokens": 68898833.0,
      "step": 37335
    },
    {
      "entropy": 5.484404706954956,
      "epoch": 3.1370300357067844,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004026905534142433,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20091877728700638,
      "num_tokens": 68907701.0,
      "step": 37340
    },
    {
      "entropy": 5.49317193031311,
      "epoch": 3.137450115521949,
      "grad_norm": 1.0,
      "learning_rate": 0.00040266589537152055,
      "loss": 5.119,
      "mean_token_accuracy": 0.1946178585290909,
      "num_tokens": 68917190.0,
      "step": 37345
    },
    {
      "entropy": 5.435900545120239,
      "epoch": 3.137870195337114,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004026412350672623,
      "loss": 4.9832,
      "mean_token_accuracy": 0.1999549850821495,
      "num_tokens": 68926591.0,
      "step": 37350
    },
    {
      "entropy": 5.477269077301026,
      "epoch": 3.1382902751522788,
      "grad_norm": 1.0,
      "learning_rate": 0.0004026165725019052,
      "loss": 5.0177,
      "mean_token_accuracy": 0.19568361192941666,
      "num_tokens": 68936007.0,
      "step": 37355
    },
    {
      "entropy": 5.471881771087647,
      "epoch": 3.138710354967444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004025919076758863,
      "loss": 5.0227,
      "mean_token_accuracy": 0.19522252976894378,
      "num_tokens": 68945548.0,
      "step": 37360
    },
    {
      "entropy": 5.526436853408813,
      "epoch": 3.139130434782609,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040256724058964244,
      "loss": 5.0839,
      "mean_token_accuracy": 0.2001351982355118,
      "num_tokens": 68954837.0,
      "step": 37365
    },
    {
      "entropy": 5.571110391616822,
      "epoch": 3.1395505145977736,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004025425712436106,
      "loss": 5.1335,
      "mean_token_accuracy": 0.19879732877016068,
      "num_tokens": 68964169.0,
      "step": 37370
    },
    {
      "entropy": 5.484325170516968,
      "epoch": 3.1399705944129384,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004025178996382277,
      "loss": 5.062,
      "mean_token_accuracy": 0.19935344457626342,
      "num_tokens": 68974157.0,
      "step": 37375
    },
    {
      "entropy": 5.530482387542724,
      "epoch": 3.140390674228103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040249322577393093,
      "loss": 5.0571,
      "mean_token_accuracy": 0.20433970540761948,
      "num_tokens": 68983617.0,
      "step": 37380
    },
    {
      "entropy": 5.468642377853394,
      "epoch": 3.1408107540432684,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004024685496511574,
      "loss": 4.9765,
      "mean_token_accuracy": 0.1983997106552124,
      "num_tokens": 68992683.0,
      "step": 37385
    },
    {
      "entropy": 5.369307565689087,
      "epoch": 3.141230833858433,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040244387127034406,
      "loss": 4.9976,
      "mean_token_accuracy": 0.19984220415353776,
      "num_tokens": 69002915.0,
      "step": 37390
    },
    {
      "entropy": 5.527750635147095,
      "epoch": 3.141650913673598,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040241919063192815,
      "loss": 5.1347,
      "mean_token_accuracy": 0.19065427780151367,
      "num_tokens": 69012557.0,
      "step": 37395
    },
    {
      "entropy": 5.582179546356201,
      "epoch": 3.1420709934887627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040239450773634693,
      "loss": 5.0765,
      "mean_token_accuracy": 0.19930254817008972,
      "num_tokens": 69021734.0,
      "step": 37400
    },
    {
      "entropy": 5.585943174362183,
      "epoch": 3.142491073303928,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004023698225840377,
      "loss": 5.1001,
      "mean_token_accuracy": 0.19622798413038253,
      "num_tokens": 69029964.0,
      "step": 37405
    },
    {
      "entropy": 5.473897647857666,
      "epoch": 3.1429111531190927,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004023451351754375,
      "loss": 5.0471,
      "mean_token_accuracy": 0.1907884106040001,
      "num_tokens": 69039190.0,
      "step": 37410
    },
    {
      "entropy": 5.3875706672668455,
      "epoch": 3.1433312329342575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004023204455109839,
      "loss": 5.0088,
      "mean_token_accuracy": 0.20021602362394333,
      "num_tokens": 69048241.0,
      "step": 37415
    },
    {
      "entropy": 5.458818960189819,
      "epoch": 3.1437513127494223,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004022957535911141,
      "loss": 5.0223,
      "mean_token_accuracy": 0.19930376410484313,
      "num_tokens": 69057723.0,
      "step": 37420
    },
    {
      "entropy": 5.522338962554931,
      "epoch": 3.144171392564587,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004022710594162656,
      "loss": 5.0706,
      "mean_token_accuracy": 0.19626810401678085,
      "num_tokens": 69066605.0,
      "step": 37425
    },
    {
      "entropy": 5.427432060241699,
      "epoch": 3.1445914723797523,
      "grad_norm": 1.125,
      "learning_rate": 0.00040224636298687573,
      "loss": 4.9505,
      "mean_token_accuracy": 0.2042725905776024,
      "num_tokens": 69075212.0,
      "step": 37430
    },
    {
      "entropy": 5.418927574157715,
      "epoch": 3.145011552194917,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004022216643033822,
      "loss": 5.0452,
      "mean_token_accuracy": 0.19952599108219146,
      "num_tokens": 69083957.0,
      "step": 37435
    },
    {
      "entropy": 5.468562030792237,
      "epoch": 3.145431632010082,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040219696336622236,
      "loss": 5.0809,
      "mean_token_accuracy": 0.1937153548002243,
      "num_tokens": 69093588.0,
      "step": 37440
    },
    {
      "entropy": 5.440897607803345,
      "epoch": 3.1458517118252467,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004021722601758337,
      "loss": 4.9717,
      "mean_token_accuracy": 0.21194516122341156,
      "num_tokens": 69101992.0,
      "step": 37445
    },
    {
      "entropy": 5.517005729675293,
      "epoch": 3.1462717916404115,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040214755473265407,
      "loss": 5.0278,
      "mean_token_accuracy": 0.20211887508630752,
      "num_tokens": 69111151.0,
      "step": 37450
    },
    {
      "entropy": 5.496988248825073,
      "epoch": 3.1466918714555767,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040212284703712093,
      "loss": 5.0979,
      "mean_token_accuracy": 0.196604622900486,
      "num_tokens": 69120191.0,
      "step": 37455
    },
    {
      "entropy": 5.480171394348145,
      "epoch": 3.1471119512707415,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040209813708967196,
      "loss": 4.95,
      "mean_token_accuracy": 0.2024596318602562,
      "num_tokens": 69129144.0,
      "step": 37460
    },
    {
      "entropy": 5.499921226501465,
      "epoch": 3.1475320310859063,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040207342489074506,
      "loss": 5.0309,
      "mean_token_accuracy": 0.1975363090634346,
      "num_tokens": 69138530.0,
      "step": 37465
    },
    {
      "entropy": 5.465929651260376,
      "epoch": 3.147952110901071,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004020487104407778,
      "loss": 5.0815,
      "mean_token_accuracy": 0.19613071531057358,
      "num_tokens": 69148119.0,
      "step": 37470
    },
    {
      "entropy": 5.4517247676849365,
      "epoch": 3.148372190716236,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040202399374020807,
      "loss": 4.9952,
      "mean_token_accuracy": 0.19939846396446229,
      "num_tokens": 69157248.0,
      "step": 37475
    },
    {
      "entropy": 5.5413764953613285,
      "epoch": 3.148792270531401,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004019992747894736,
      "loss": 5.0684,
      "mean_token_accuracy": 0.1957733988761902,
      "num_tokens": 69166990.0,
      "step": 37480
    },
    {
      "entropy": 5.4703981399536135,
      "epoch": 3.149212350346566,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004019745535890125,
      "loss": 4.9574,
      "mean_token_accuracy": 0.20437587648630143,
      "num_tokens": 69175689.0,
      "step": 37485
    },
    {
      "entropy": 5.371008682250976,
      "epoch": 3.1496324301617307,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004019498301392626,
      "loss": 4.8686,
      "mean_token_accuracy": 0.20202555507421494,
      "num_tokens": 69184433.0,
      "step": 37490
    },
    {
      "entropy": 5.432024908065796,
      "epoch": 3.1500525099768955,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004019251044406617,
      "loss": 5.0884,
      "mean_token_accuracy": 0.19411247372627258,
      "num_tokens": 69193908.0,
      "step": 37495
    },
    {
      "entropy": 5.477922964096069,
      "epoch": 3.1504725897920607,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040190037649364796,
      "loss": 5.0451,
      "mean_token_accuracy": 0.1968008279800415,
      "num_tokens": 69202276.0,
      "step": 37500
    },
    {
      "entropy": 5.477943992614746,
      "epoch": 3.1508926696072255,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040187564629865944,
      "loss": 5.1127,
      "mean_token_accuracy": 0.18930143713951111,
      "num_tokens": 69211519.0,
      "step": 37505
    },
    {
      "entropy": 5.5186258316040036,
      "epoch": 3.1513127494223903,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040185091385613414,
      "loss": 5.049,
      "mean_token_accuracy": 0.19481094628572465,
      "num_tokens": 69221514.0,
      "step": 37510
    },
    {
      "entropy": 5.490809345245362,
      "epoch": 3.151732829237555,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004018261791665102,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2076988324522972,
      "num_tokens": 69230075.0,
      "step": 37515
    },
    {
      "entropy": 5.435186719894409,
      "epoch": 3.15215290905272,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004018014422302259,
      "loss": 5.0189,
      "mean_token_accuracy": 0.19860631674528123,
      "num_tokens": 69239912.0,
      "step": 37520
    },
    {
      "entropy": 5.537674617767334,
      "epoch": 3.152572988867885,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040177670304771923,
      "loss": 5.2108,
      "mean_token_accuracy": 0.18960022628307344,
      "num_tokens": 69249797.0,
      "step": 37525
    },
    {
      "entropy": 5.510783529281616,
      "epoch": 3.15299306868305,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004017519616194286,
      "loss": 4.9809,
      "mean_token_accuracy": 0.2093310162425041,
      "num_tokens": 69259156.0,
      "step": 37530
    },
    {
      "entropy": 5.465088701248169,
      "epoch": 3.1534131484982146,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040172721794579227,
      "loss": 5.0453,
      "mean_token_accuracy": 0.19838739335536956,
      "num_tokens": 69268802.0,
      "step": 37535
    },
    {
      "entropy": 5.494611167907715,
      "epoch": 3.1538332283133794,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004017024720272485,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20049381703138353,
      "num_tokens": 69277087.0,
      "step": 37540
    },
    {
      "entropy": 5.488122177124024,
      "epoch": 3.1542533081285447,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004016777238642356,
      "loss": 5.0171,
      "mean_token_accuracy": 0.1963404521346092,
      "num_tokens": 69285505.0,
      "step": 37545
    },
    {
      "entropy": 5.494830751419068,
      "epoch": 3.1546733879437094,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040165297345719214,
      "loss": 4.9525,
      "mean_token_accuracy": 0.20461241602897645,
      "num_tokens": 69294922.0,
      "step": 37550
    },
    {
      "entropy": 5.504093074798584,
      "epoch": 3.1550934677588742,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004016282208065565,
      "loss": 5.075,
      "mean_token_accuracy": 0.1999457761645317,
      "num_tokens": 69303391.0,
      "step": 37555
    },
    {
      "entropy": 5.3903467655181885,
      "epoch": 3.155513547574039,
      "grad_norm": 1.109375,
      "learning_rate": 0.000401603465912767,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20239628255367278,
      "num_tokens": 69313376.0,
      "step": 37560
    },
    {
      "entropy": 5.464016628265381,
      "epoch": 3.155933627389204,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004015787087762625,
      "loss": 4.9949,
      "mean_token_accuracy": 0.19979466795921325,
      "num_tokens": 69322420.0,
      "step": 37565
    },
    {
      "entropy": 5.482305955886841,
      "epoch": 3.156353707204369,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040155394939748117,
      "loss": 5.1254,
      "mean_token_accuracy": 0.1859981521964073,
      "num_tokens": 69332220.0,
      "step": 37570
    },
    {
      "entropy": 5.561050462722778,
      "epoch": 3.156773787019534,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040152918777686185,
      "loss": 5.1174,
      "mean_token_accuracy": 0.1977437987923622,
      "num_tokens": 69341644.0,
      "step": 37575
    },
    {
      "entropy": 5.544359159469605,
      "epoch": 3.1571938668346986,
      "grad_norm": 1.125,
      "learning_rate": 0.00040150442391484307,
      "loss": 5.136,
      "mean_token_accuracy": 0.19638286679983138,
      "num_tokens": 69350987.0,
      "step": 37580
    },
    {
      "entropy": 5.312761402130127,
      "epoch": 3.1576139466498634,
      "grad_norm": 1.125,
      "learning_rate": 0.0004014796578118636,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21759833991527558,
      "num_tokens": 69359961.0,
      "step": 37585
    },
    {
      "entropy": 5.439679145812988,
      "epoch": 3.158034026465028,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040145488946836223,
      "loss": 4.9796,
      "mean_token_accuracy": 0.2110493704676628,
      "num_tokens": 69369310.0,
      "step": 37590
    },
    {
      "entropy": 5.428500509262085,
      "epoch": 3.1584541062801934,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004014301188847775,
      "loss": 4.977,
      "mean_token_accuracy": 0.20203733891248704,
      "num_tokens": 69378277.0,
      "step": 37595
    },
    {
      "entropy": 5.5675922393798825,
      "epoch": 3.158874186095358,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00040140534606154826,
      "loss": 5.1007,
      "mean_token_accuracy": 0.19520897120237352,
      "num_tokens": 69387754.0,
      "step": 37600
    },
    {
      "entropy": 5.440639877319336,
      "epoch": 3.159294265910523,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004013805709991135,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20701155960559844,
      "num_tokens": 69396634.0,
      "step": 37605
    },
    {
      "entropy": 5.4854936599731445,
      "epoch": 3.1597143457256878,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000401355793697912,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19584751576185228,
      "num_tokens": 69406276.0,
      "step": 37610
    },
    {
      "entropy": 5.441031789779663,
      "epoch": 3.1601344255408526,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040133101415838263,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20119614005088807,
      "num_tokens": 69414625.0,
      "step": 37615
    },
    {
      "entropy": 5.4657610893249515,
      "epoch": 3.160554505356018,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040130623238096444,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20342430621385574,
      "num_tokens": 69423451.0,
      "step": 37620
    },
    {
      "entropy": 5.556706666946411,
      "epoch": 3.1609745851711826,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004012814483660964,
      "loss": 5.0906,
      "mean_token_accuracy": 0.20054681301116944,
      "num_tokens": 69433167.0,
      "step": 37625
    },
    {
      "entropy": 5.491298818588257,
      "epoch": 3.1613946649863474,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040125666211421747,
      "loss": 5.0506,
      "mean_token_accuracy": 0.2032269850373268,
      "num_tokens": 69442167.0,
      "step": 37630
    },
    {
      "entropy": 5.502869701385498,
      "epoch": 3.161814744801512,
      "grad_norm": 1.0,
      "learning_rate": 0.0004012318736257667,
      "loss": 5.0669,
      "mean_token_accuracy": 0.2013854667544365,
      "num_tokens": 69451875.0,
      "step": 37635
    },
    {
      "entropy": 5.48580379486084,
      "epoch": 3.1622348246166774,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004012070829011835,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20550106465816498,
      "num_tokens": 69460758.0,
      "step": 37640
    },
    {
      "entropy": 5.518317317962646,
      "epoch": 3.162654904431842,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004011822899409067,
      "loss": 5.1077,
      "mean_token_accuracy": 0.20214879214763642,
      "num_tokens": 69469856.0,
      "step": 37645
    },
    {
      "entropy": 5.460502481460571,
      "epoch": 3.163074984247007,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004011574947453755,
      "loss": 5.0045,
      "mean_token_accuracy": 0.20045301467180252,
      "num_tokens": 69479058.0,
      "step": 37650
    },
    {
      "entropy": 5.459316921234131,
      "epoch": 3.1634950640621717,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004011326973150294,
      "loss": 5.0867,
      "mean_token_accuracy": 0.1934477612376213,
      "num_tokens": 69488183.0,
      "step": 37655
    },
    {
      "entropy": 5.3921386241912845,
      "epoch": 3.1639151438773365,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040110789765030754,
      "loss": 4.9672,
      "mean_token_accuracy": 0.2014950171113014,
      "num_tokens": 69496282.0,
      "step": 37660
    },
    {
      "entropy": 5.441311550140381,
      "epoch": 3.1643352236925018,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040108309575164906,
      "loss": 5.0005,
      "mean_token_accuracy": 0.19739663898944854,
      "num_tokens": 69504448.0,
      "step": 37665
    },
    {
      "entropy": 5.500400543212891,
      "epoch": 3.1647553035076665,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004010582916194935,
      "loss": 5.1007,
      "mean_token_accuracy": 0.1941326841711998,
      "num_tokens": 69514496.0,
      "step": 37670
    },
    {
      "entropy": 5.441692543029785,
      "epoch": 3.1651753833228313,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004010334852542803,
      "loss": 4.9478,
      "mean_token_accuracy": 0.19845906049013137,
      "num_tokens": 69524255.0,
      "step": 37675
    },
    {
      "entropy": 5.477640390396118,
      "epoch": 3.165595463137996,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040100867665644876,
      "loss": 5.0275,
      "mean_token_accuracy": 0.20761329978704451,
      "num_tokens": 69534209.0,
      "step": 37680
    },
    {
      "entropy": 5.497815799713135,
      "epoch": 3.166015542953161,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004009838658264384,
      "loss": 5.0118,
      "mean_token_accuracy": 0.19697001129388808,
      "num_tokens": 69543875.0,
      "step": 37685
    },
    {
      "entropy": 5.524178886413575,
      "epoch": 3.166435622768326,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004009590527646887,
      "loss": 5.0604,
      "mean_token_accuracy": 0.2001832276582718,
      "num_tokens": 69553311.0,
      "step": 37690
    },
    {
      "entropy": 5.479667139053345,
      "epoch": 3.166855702583491,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004009342374716393,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20137991160154342,
      "num_tokens": 69562885.0,
      "step": 37695
    },
    {
      "entropy": 5.465568876266479,
      "epoch": 3.1672757823986557,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004009094199477297,
      "loss": 4.9746,
      "mean_token_accuracy": 0.19681413918733598,
      "num_tokens": 69572956.0,
      "step": 37700
    },
    {
      "entropy": 5.3897112846374515,
      "epoch": 3.1676958622138205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040088460019339954,
      "loss": 5.0225,
      "mean_token_accuracy": 0.20174658596515654,
      "num_tokens": 69581969.0,
      "step": 37705
    },
    {
      "entropy": 5.485685920715332,
      "epoch": 3.1681159420289857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004008597782090886,
      "loss": 5.0838,
      "mean_token_accuracy": 0.19718752950429916,
      "num_tokens": 69592271.0,
      "step": 37710
    },
    {
      "entropy": 5.503107213973999,
      "epoch": 3.1685360218441505,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040083495399523636,
      "loss": 5.084,
      "mean_token_accuracy": 0.20155852884054185,
      "num_tokens": 69602685.0,
      "step": 37715
    },
    {
      "entropy": 5.522109413146973,
      "epoch": 3.1689561016593153,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004008101275522827,
      "loss": 5.0496,
      "mean_token_accuracy": 0.19981058090925216,
      "num_tokens": 69612572.0,
      "step": 37720
    },
    {
      "entropy": 5.432215023040771,
      "epoch": 3.16937618147448,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004007852988806675,
      "loss": 4.9801,
      "mean_token_accuracy": 0.20320723354816436,
      "num_tokens": 69621661.0,
      "step": 37725
    },
    {
      "entropy": 5.38215913772583,
      "epoch": 3.169796261289645,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040076046798083047,
      "loss": 4.9717,
      "mean_token_accuracy": 0.20405869781970978,
      "num_tokens": 69630803.0,
      "step": 37730
    },
    {
      "entropy": 5.504638004302978,
      "epoch": 3.17021634110481,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004007356348532115,
      "loss": 5.0907,
      "mean_token_accuracy": 0.20152545124292373,
      "num_tokens": 69639992.0,
      "step": 37735
    },
    {
      "entropy": 5.5313231468200685,
      "epoch": 3.170636420919975,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004007107994982505,
      "loss": 5.0097,
      "mean_token_accuracy": 0.20337896347045897,
      "num_tokens": 69649156.0,
      "step": 37740
    },
    {
      "entropy": 5.475124025344849,
      "epoch": 3.1710565007351397,
      "grad_norm": 1.125,
      "learning_rate": 0.0004006859619163875,
      "loss": 4.9697,
      "mean_token_accuracy": 0.19661925435066224,
      "num_tokens": 69659129.0,
      "step": 37745
    },
    {
      "entropy": 5.426178979873657,
      "epoch": 3.1714765805503045,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040066112210806234,
      "loss": 5.039,
      "mean_token_accuracy": 0.1983688533306122,
      "num_tokens": 69668103.0,
      "step": 37750
    },
    {
      "entropy": 5.518638849258423,
      "epoch": 3.1718966603654692,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040063628007371504,
      "loss": 5.1474,
      "mean_token_accuracy": 0.18887291252613067,
      "num_tokens": 69678185.0,
      "step": 37755
    },
    {
      "entropy": 5.481713581085205,
      "epoch": 3.1723167401806345,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004006114358137858,
      "loss": 4.9763,
      "mean_token_accuracy": 0.20361330807209016,
      "num_tokens": 69687283.0,
      "step": 37760
    },
    {
      "entropy": 5.516620254516601,
      "epoch": 3.1727368199957993,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004005865893287147,
      "loss": 5.1074,
      "mean_token_accuracy": 0.1871713489294052,
      "num_tokens": 69697121.0,
      "step": 37765
    },
    {
      "entropy": 5.472781991958618,
      "epoch": 3.173156899810964,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004005617406189418,
      "loss": 5.0112,
      "mean_token_accuracy": 0.20440395325422286,
      "num_tokens": 69706270.0,
      "step": 37770
    },
    {
      "entropy": 5.559930801391602,
      "epoch": 3.173576979626129,
      "grad_norm": 1.125,
      "learning_rate": 0.0004005368896849073,
      "loss": 5.136,
      "mean_token_accuracy": 0.19211809039115907,
      "num_tokens": 69717099.0,
      "step": 37775
    },
    {
      "entropy": 5.510134553909301,
      "epoch": 3.1739970594412936,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004005120365270514,
      "loss": 5.0141,
      "mean_token_accuracy": 0.19691647589206696,
      "num_tokens": 69727719.0,
      "step": 37780
    },
    {
      "entropy": 5.500736999511719,
      "epoch": 3.174417139256459,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004004871811458145,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21011366695165634,
      "num_tokens": 69736898.0,
      "step": 37785
    },
    {
      "entropy": 5.497056245803833,
      "epoch": 3.1748372190716236,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004004623235416367,
      "loss": 5.0239,
      "mean_token_accuracy": 0.2027146264910698,
      "num_tokens": 69746520.0,
      "step": 37790
    },
    {
      "entropy": 5.469923830032348,
      "epoch": 3.1752572988867884,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040043746371495854,
      "loss": 5.0608,
      "mean_token_accuracy": 0.19891727566719056,
      "num_tokens": 69755125.0,
      "step": 37795
    },
    {
      "entropy": 5.44061279296875,
      "epoch": 3.175677378701953,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004004126016662202,
      "loss": 5.0049,
      "mean_token_accuracy": 0.19357188493013383,
      "num_tokens": 69763965.0,
      "step": 37800
    },
    {
      "entropy": 5.422125959396363,
      "epoch": 3.1760974585171184,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040038773739586224,
      "loss": 4.9619,
      "mean_token_accuracy": 0.19878497421741487,
      "num_tokens": 69773193.0,
      "step": 37805
    },
    {
      "entropy": 5.452244424819947,
      "epoch": 3.1765175383322832,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004003628709043252,
      "loss": 4.9943,
      "mean_token_accuracy": 0.20481649786233902,
      "num_tokens": 69782485.0,
      "step": 37810
    },
    {
      "entropy": 5.432509708404541,
      "epoch": 3.176937618147448,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004003380021920493,
      "loss": 4.9813,
      "mean_token_accuracy": 0.20218124240636826,
      "num_tokens": 69791304.0,
      "step": 37815
    },
    {
      "entropy": 5.456427717208863,
      "epoch": 3.177357697962613,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004003131312594753,
      "loss": 4.9924,
      "mean_token_accuracy": 0.20298028588294983,
      "num_tokens": 69800887.0,
      "step": 37820
    },
    {
      "entropy": 5.5331809520721436,
      "epoch": 3.1777777777777776,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040028825810704366,
      "loss": 5.0618,
      "mean_token_accuracy": 0.18891056329011918,
      "num_tokens": 69811062.0,
      "step": 37825
    },
    {
      "entropy": 5.542026662826538,
      "epoch": 3.178197857592943,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040026338273519514,
      "loss": 5.0828,
      "mean_token_accuracy": 0.19872816801071166,
      "num_tokens": 69819901.0,
      "step": 37830
    },
    {
      "entropy": 5.515076160430908,
      "epoch": 3.1786179374081076,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040023850514437015,
      "loss": 5.0763,
      "mean_token_accuracy": 0.19581133127212524,
      "num_tokens": 69830554.0,
      "step": 37835
    },
    {
      "entropy": 5.515093421936035,
      "epoch": 3.1790380172232724,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040021362533500964,
      "loss": 5.0476,
      "mean_token_accuracy": 0.19497296661138536,
      "num_tokens": 69840461.0,
      "step": 37840
    },
    {
      "entropy": 5.484316730499268,
      "epoch": 3.179458097038437,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040018874330755423,
      "loss": 5.0899,
      "mean_token_accuracy": 0.19578086584806442,
      "num_tokens": 69849546.0,
      "step": 37845
    },
    {
      "entropy": 5.412982940673828,
      "epoch": 3.1798781768536024,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040016385906244464,
      "loss": 4.9165,
      "mean_token_accuracy": 0.21507821828126908,
      "num_tokens": 69858875.0,
      "step": 37850
    },
    {
      "entropy": 5.462613487243653,
      "epoch": 3.180298256668767,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004001389726001217,
      "loss": 5.0217,
      "mean_token_accuracy": 0.19560861140489577,
      "num_tokens": 69867440.0,
      "step": 37855
    },
    {
      "entropy": 5.409283399581909,
      "epoch": 3.180718336483932,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040011408392102635,
      "loss": 4.9072,
      "mean_token_accuracy": 0.20924327671527862,
      "num_tokens": 69876167.0,
      "step": 37860
    },
    {
      "entropy": 5.5085841655731205,
      "epoch": 3.1811384162990968,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004000891930255994,
      "loss": 5.0759,
      "mean_token_accuracy": 0.1947849541902542,
      "num_tokens": 69886548.0,
      "step": 37865
    },
    {
      "entropy": 5.454615545272827,
      "epoch": 3.1815584961142616,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004000642999142819,
      "loss": 5.0159,
      "mean_token_accuracy": 0.20463456958532333,
      "num_tokens": 69894969.0,
      "step": 37870
    },
    {
      "entropy": 5.458572816848755,
      "epoch": 3.181978575929427,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040003940458751457,
      "loss": 4.9987,
      "mean_token_accuracy": 0.195869317650795,
      "num_tokens": 69903649.0,
      "step": 37875
    },
    {
      "entropy": 5.470513486862183,
      "epoch": 3.1823986557445916,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040001450704573875,
      "loss": 5.0495,
      "mean_token_accuracy": 0.19584746211767196,
      "num_tokens": 69913266.0,
      "step": 37880
    },
    {
      "entropy": 5.453602695465088,
      "epoch": 3.1828187355597564,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003999896072893952,
      "loss": 5.0082,
      "mean_token_accuracy": 0.19878625720739365,
      "num_tokens": 69922821.0,
      "step": 37885
    },
    {
      "entropy": 5.5880499362945555,
      "epoch": 3.183238815374921,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003999647053189252,
      "loss": 5.1284,
      "mean_token_accuracy": 0.19759187400341033,
      "num_tokens": 69932234.0,
      "step": 37890
    },
    {
      "entropy": 5.509851789474487,
      "epoch": 3.183658895190086,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039993980113476967,
      "loss": 5.0235,
      "mean_token_accuracy": 0.19589988738298417,
      "num_tokens": 69941588.0,
      "step": 37895
    },
    {
      "entropy": 5.4445935726165775,
      "epoch": 3.184078975005251,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039991489473737,
      "loss": 4.9322,
      "mean_token_accuracy": 0.2124770849943161,
      "num_tokens": 69950676.0,
      "step": 37900
    },
    {
      "entropy": 5.468519258499145,
      "epoch": 3.184499054820416,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039988998612716723,
      "loss": 4.9533,
      "mean_token_accuracy": 0.21135903596878053,
      "num_tokens": 69958616.0,
      "step": 37905
    },
    {
      "entropy": 5.473688793182373,
      "epoch": 3.1849191346355807,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039986507530460273,
      "loss": 5.094,
      "mean_token_accuracy": 0.1956064522266388,
      "num_tokens": 69967818.0,
      "step": 37910
    },
    {
      "entropy": 5.523324728012085,
      "epoch": 3.1853392144507455,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039984016227011764,
      "loss": 5.066,
      "mean_token_accuracy": 0.20044741481542588,
      "num_tokens": 69976996.0,
      "step": 37915
    },
    {
      "entropy": 5.470655345916748,
      "epoch": 3.1857592942659103,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003998152470241534,
      "loss": 5.0614,
      "mean_token_accuracy": 0.1969575345516205,
      "num_tokens": 69986337.0,
      "step": 37920
    },
    {
      "entropy": 5.404251050949097,
      "epoch": 3.1861793740810755,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003997903295671513,
      "loss": 4.9642,
      "mean_token_accuracy": 0.1994946777820587,
      "num_tokens": 69994980.0,
      "step": 37925
    },
    {
      "entropy": 5.422129774093628,
      "epoch": 3.1865994538962403,
      "grad_norm": 1.125,
      "learning_rate": 0.00039976540989955284,
      "loss": 4.9709,
      "mean_token_accuracy": 0.2006675511598587,
      "num_tokens": 70002842.0,
      "step": 37930
    },
    {
      "entropy": 5.39985294342041,
      "epoch": 3.187019533711405,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039974048802179934,
      "loss": 4.9051,
      "mean_token_accuracy": 0.21041079312562944,
      "num_tokens": 70012148.0,
      "step": 37935
    },
    {
      "entropy": 5.503354930877686,
      "epoch": 3.18743961352657,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003997155639343323,
      "loss": 5.0615,
      "mean_token_accuracy": 0.2030509188771248,
      "num_tokens": 70021408.0,
      "step": 37940
    },
    {
      "entropy": 5.465933322906494,
      "epoch": 3.187859693341735,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003996906376375933,
      "loss": 4.9597,
      "mean_token_accuracy": 0.2061482384800911,
      "num_tokens": 70030484.0,
      "step": 37945
    },
    {
      "entropy": 5.354239559173584,
      "epoch": 3.1882797731569,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003996657091320239,
      "loss": 4.8786,
      "mean_token_accuracy": 0.21138615310192108,
      "num_tokens": 70039061.0,
      "step": 37950
    },
    {
      "entropy": 5.349978256225586,
      "epoch": 3.1886998529720647,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003996407784180656,
      "loss": 4.9489,
      "mean_token_accuracy": 0.20569489598274232,
      "num_tokens": 70048905.0,
      "step": 37955
    },
    {
      "entropy": 5.466477108001709,
      "epoch": 3.1891199327872295,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003996158454961601,
      "loss": 5.0372,
      "mean_token_accuracy": 0.19740619361400605,
      "num_tokens": 70057831.0,
      "step": 37960
    },
    {
      "entropy": 5.472411108016968,
      "epoch": 3.1895400126023943,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039959091036674907,
      "loss": 5.0161,
      "mean_token_accuracy": 0.2061769187450409,
      "num_tokens": 70066799.0,
      "step": 37965
    },
    {
      "entropy": 5.4235272884368895,
      "epoch": 3.1899600924175595,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039956597303027414,
      "loss": 4.9145,
      "mean_token_accuracy": 0.20260728299617767,
      "num_tokens": 70075535.0,
      "step": 37970
    },
    {
      "entropy": 5.368491840362549,
      "epoch": 3.1903801722327243,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003995410334871773,
      "loss": 4.9018,
      "mean_token_accuracy": 0.21422181129455567,
      "num_tokens": 70083991.0,
      "step": 37975
    },
    {
      "entropy": 5.433748197555542,
      "epoch": 3.190800252047889,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039951609173790003,
      "loss": 5.0496,
      "mean_token_accuracy": 0.19999650567770005,
      "num_tokens": 70093936.0,
      "step": 37980
    },
    {
      "entropy": 5.417809104919433,
      "epoch": 3.191220331863054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003994911477828844,
      "loss": 4.9012,
      "mean_token_accuracy": 0.21390611678361893,
      "num_tokens": 70102501.0,
      "step": 37985
    },
    {
      "entropy": 5.491446876525879,
      "epoch": 3.1916404116782187,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039946620162257206,
      "loss": 5.0293,
      "mean_token_accuracy": 0.20092049688100816,
      "num_tokens": 70111074.0,
      "step": 37990
    },
    {
      "entropy": 5.476159334182739,
      "epoch": 3.192060491493384,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003994412532574053,
      "loss": 4.9966,
      "mean_token_accuracy": 0.20942515432834624,
      "num_tokens": 70120232.0,
      "step": 37995
    },
    {
      "entropy": 5.515239524841308,
      "epoch": 3.1924805713085487,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003994163026878256,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19532779455184937,
      "num_tokens": 70129776.0,
      "step": 38000
    },
    {
      "entropy": 5.480040454864502,
      "epoch": 3.1929006511237135,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003993913499142752,
      "loss": 5.1026,
      "mean_token_accuracy": 0.1975361153483391,
      "num_tokens": 70139223.0,
      "step": 38005
    },
    {
      "entropy": 5.478199815750122,
      "epoch": 3.1933207309388782,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003993663949371962,
      "loss": 4.9917,
      "mean_token_accuracy": 0.2040783077478409,
      "num_tokens": 70148455.0,
      "step": 38010
    },
    {
      "entropy": 5.413830184936524,
      "epoch": 3.1937408107540435,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039934143775703044,
      "loss": 4.9645,
      "mean_token_accuracy": 0.20376754254102708,
      "num_tokens": 70157924.0,
      "step": 38015
    },
    {
      "entropy": 5.47748122215271,
      "epoch": 3.1941608905692083,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039931647837422016,
      "loss": 5.0492,
      "mean_token_accuracy": 0.20274214595556259,
      "num_tokens": 70167191.0,
      "step": 38020
    },
    {
      "entropy": 5.563238620758057,
      "epoch": 3.194580970384373,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003992915167892075,
      "loss": 5.0914,
      "mean_token_accuracy": 0.19817918092012404,
      "num_tokens": 70177426.0,
      "step": 38025
    },
    {
      "entropy": 5.492373275756836,
      "epoch": 3.195001050199538,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003992665530024346,
      "loss": 5.0417,
      "mean_token_accuracy": 0.19574618935585023,
      "num_tokens": 70186464.0,
      "step": 38030
    },
    {
      "entropy": 5.5137567043304445,
      "epoch": 3.1954211300147026,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003992415870143437,
      "loss": 5.0668,
      "mean_token_accuracy": 0.18701160699129105,
      "num_tokens": 70195528.0,
      "step": 38035
    },
    {
      "entropy": 5.602050876617431,
      "epoch": 3.195841209829868,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003992166188253771,
      "loss": 5.2338,
      "mean_token_accuracy": 0.18800810873508453,
      "num_tokens": 70206450.0,
      "step": 38040
    },
    {
      "entropy": 5.467767810821533,
      "epoch": 3.1962612896450326,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039919164843597697,
      "loss": 5.0119,
      "mean_token_accuracy": 0.20355167984962463,
      "num_tokens": 70214729.0,
      "step": 38045
    },
    {
      "entropy": 5.508221387863159,
      "epoch": 3.1966813694601974,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039916667584658584,
      "loss": 5.0381,
      "mean_token_accuracy": 0.2002306267619133,
      "num_tokens": 70223473.0,
      "step": 38050
    },
    {
      "entropy": 5.404289388656617,
      "epoch": 3.197101449275362,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003991417010576459,
      "loss": 4.9803,
      "mean_token_accuracy": 0.21372398287057875,
      "num_tokens": 70232970.0,
      "step": 38055
    },
    {
      "entropy": 5.4409499168396,
      "epoch": 3.197521529090527,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003991167240695997,
      "loss": 5.0468,
      "mean_token_accuracy": 0.19464128613471984,
      "num_tokens": 70242734.0,
      "step": 38060
    },
    {
      "entropy": 5.448752737045288,
      "epoch": 3.1979416089056922,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003990917448828897,
      "loss": 5.001,
      "mean_token_accuracy": 0.2004161983728409,
      "num_tokens": 70251606.0,
      "step": 38065
    },
    {
      "entropy": 5.466198396682739,
      "epoch": 3.198361688720857,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039906676349795817,
      "loss": 4.952,
      "mean_token_accuracy": 0.20351530164480208,
      "num_tokens": 70260228.0,
      "step": 38070
    },
    {
      "entropy": 5.405320882797241,
      "epoch": 3.198781768536022,
      "grad_norm": 1.109375,
      "learning_rate": 0.000399041779915248,
      "loss": 5.0594,
      "mean_token_accuracy": 0.20029732286930085,
      "num_tokens": 70268819.0,
      "step": 38075
    },
    {
      "entropy": 5.490515995025635,
      "epoch": 3.1992018483511866,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003990167941352014,
      "loss": 5.1731,
      "mean_token_accuracy": 0.19130388647317886,
      "num_tokens": 70278081.0,
      "step": 38080
    },
    {
      "entropy": 5.499806499481201,
      "epoch": 3.1996219281663514,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003989918061582613,
      "loss": 4.9851,
      "mean_token_accuracy": 0.19651320427656174,
      "num_tokens": 70287128.0,
      "step": 38085
    },
    {
      "entropy": 5.507339906692505,
      "epoch": 3.2000420079815166,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003989668159848701,
      "loss": 4.9608,
      "mean_token_accuracy": 0.20097176879644393,
      "num_tokens": 70296163.0,
      "step": 38090
    },
    {
      "entropy": 5.414227724075317,
      "epoch": 3.2004620877966814,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003989418236154707,
      "loss": 4.9955,
      "mean_token_accuracy": 0.20146827399730682,
      "num_tokens": 70305084.0,
      "step": 38095
    },
    {
      "entropy": 5.409955072402954,
      "epoch": 3.200882167611846,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003989168290505056,
      "loss": 4.96,
      "mean_token_accuracy": 0.1992536649107933,
      "num_tokens": 70314634.0,
      "step": 38100
    },
    {
      "entropy": 5.446388053894043,
      "epoch": 3.201302247427011,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003988918322904177,
      "loss": 4.9991,
      "mean_token_accuracy": 0.2028224915266037,
      "num_tokens": 70324377.0,
      "step": 38105
    },
    {
      "entropy": 5.469159936904907,
      "epoch": 3.201722327242176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039886683333564983,
      "loss": 4.9765,
      "mean_token_accuracy": 0.2056470289826393,
      "num_tokens": 70333146.0,
      "step": 38110
    },
    {
      "entropy": 5.517913055419922,
      "epoch": 3.202142407057341,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003988418321866447,
      "loss": 5.0634,
      "mean_token_accuracy": 0.19329778403043746,
      "num_tokens": 70342896.0,
      "step": 38115
    },
    {
      "entropy": 5.377479410171508,
      "epoch": 3.2025624868725058,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039881682884384536,
      "loss": 4.9685,
      "mean_token_accuracy": 0.2046784982085228,
      "num_tokens": 70351993.0,
      "step": 38120
    },
    {
      "entropy": 5.440989780426025,
      "epoch": 3.2029825666876706,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039879182330769465,
      "loss": 5.0074,
      "mean_token_accuracy": 0.2039107635617256,
      "num_tokens": 70362009.0,
      "step": 38125
    },
    {
      "entropy": 5.49866623878479,
      "epoch": 3.2034026465028353,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003987668155786355,
      "loss": 5.028,
      "mean_token_accuracy": 0.20368460714817047,
      "num_tokens": 70370724.0,
      "step": 38130
    },
    {
      "entropy": 5.419697427749634,
      "epoch": 3.2038227263180006,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00039874180565711094,
      "loss": 4.8729,
      "mean_token_accuracy": 0.21098068356513977,
      "num_tokens": 70379789.0,
      "step": 38135
    },
    {
      "entropy": 5.386115074157715,
      "epoch": 3.2042428061331654,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039871679354356405,
      "loss": 4.9217,
      "mean_token_accuracy": 0.20000389516353606,
      "num_tokens": 70388481.0,
      "step": 38140
    },
    {
      "entropy": 5.404332494735717,
      "epoch": 3.20466288594833,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039869177923843784,
      "loss": 4.9559,
      "mean_token_accuracy": 0.20690224170684815,
      "num_tokens": 70397199.0,
      "step": 38145
    },
    {
      "entropy": 5.4566761493682865,
      "epoch": 3.205082965763495,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003986667627421755,
      "loss": 4.9267,
      "mean_token_accuracy": 0.207894204556942,
      "num_tokens": 70406484.0,
      "step": 38150
    },
    {
      "entropy": 5.395777082443237,
      "epoch": 3.20550304557866,
      "grad_norm": 1.125,
      "learning_rate": 0.0003986417440552201,
      "loss": 4.98,
      "mean_token_accuracy": 0.20897393077611923,
      "num_tokens": 70415023.0,
      "step": 38155
    },
    {
      "entropy": 5.400672769546508,
      "epoch": 3.205923125393825,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039861672317801483,
      "loss": 4.9211,
      "mean_token_accuracy": 0.19815942645072937,
      "num_tokens": 70424882.0,
      "step": 38160
    },
    {
      "entropy": 5.525783491134644,
      "epoch": 3.2063432052089897,
      "grad_norm": 1.140625,
      "learning_rate": 0.000398591700111003,
      "loss": 5.111,
      "mean_token_accuracy": 0.19585951566696166,
      "num_tokens": 70434024.0,
      "step": 38165
    },
    {
      "entropy": 5.479860067367554,
      "epoch": 3.2067632850241545,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039856667485462785,
      "loss": 4.9961,
      "mean_token_accuracy": 0.20868973582983016,
      "num_tokens": 70442976.0,
      "step": 38170
    },
    {
      "entropy": 5.473452472686768,
      "epoch": 3.2071833648393193,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039854164740933266,
      "loss": 5.0533,
      "mean_token_accuracy": 0.1997311532497406,
      "num_tokens": 70452820.0,
      "step": 38175
    },
    {
      "entropy": 5.341928768157959,
      "epoch": 3.2076034446544845,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039851661777556087,
      "loss": 4.8432,
      "mean_token_accuracy": 0.21004477888345718,
      "num_tokens": 70460547.0,
      "step": 38180
    },
    {
      "entropy": 5.4476062774658205,
      "epoch": 3.2080235244696493,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039849158595375573,
      "loss": 5.0255,
      "mean_token_accuracy": 0.19405293315649033,
      "num_tokens": 70469550.0,
      "step": 38185
    },
    {
      "entropy": 5.4291136264801025,
      "epoch": 3.208443604284814,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003984665519443607,
      "loss": 4.9367,
      "mean_token_accuracy": 0.2087012156844139,
      "num_tokens": 70478239.0,
      "step": 38190
    },
    {
      "entropy": 5.460344314575195,
      "epoch": 3.208863684099979,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039844151574781937,
      "loss": 5.0183,
      "mean_token_accuracy": 0.19263792932033538,
      "num_tokens": 70487712.0,
      "step": 38195
    },
    {
      "entropy": 5.485580968856811,
      "epoch": 3.2092837639151437,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039841647736457513,
      "loss": 5.0365,
      "mean_token_accuracy": 0.20145193487405777,
      "num_tokens": 70497118.0,
      "step": 38200
    },
    {
      "entropy": 5.414409828186035,
      "epoch": 3.209703843730309,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003983914367950715,
      "loss": 4.9055,
      "mean_token_accuracy": 0.20282009541988372,
      "num_tokens": 70505501.0,
      "step": 38205
    },
    {
      "entropy": 5.44981107711792,
      "epoch": 3.2101239235454737,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00039836639403975215,
      "loss": 5.0192,
      "mean_token_accuracy": 0.1993451952934265,
      "num_tokens": 70515729.0,
      "step": 38210
    },
    {
      "entropy": 5.440519857406616,
      "epoch": 3.2105440033606385,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039834134909906067,
      "loss": 4.998,
      "mean_token_accuracy": 0.20613971799612046,
      "num_tokens": 70525591.0,
      "step": 38215
    },
    {
      "entropy": 5.555025863647461,
      "epoch": 3.2109640831758033,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003983163019734406,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20049538761377333,
      "num_tokens": 70534808.0,
      "step": 38220
    },
    {
      "entropy": 5.517244911193847,
      "epoch": 3.211384162990968,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039829125266333584,
      "loss": 5.0369,
      "mean_token_accuracy": 0.20173622369766236,
      "num_tokens": 70543804.0,
      "step": 38225
    },
    {
      "entropy": 5.392929983139038,
      "epoch": 3.2118042428061333,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00039826620116918996,
      "loss": 5.0078,
      "mean_token_accuracy": 0.21291868388652802,
      "num_tokens": 70553555.0,
      "step": 38230
    },
    {
      "entropy": 5.461224794387817,
      "epoch": 3.212224322621298,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003982411474914468,
      "loss": 5.0611,
      "mean_token_accuracy": 0.19956330806016923,
      "num_tokens": 70562207.0,
      "step": 38235
    },
    {
      "entropy": 5.476815271377563,
      "epoch": 3.212644402436463,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003982160916305502,
      "loss": 5.0824,
      "mean_token_accuracy": 0.18946980834007263,
      "num_tokens": 70572016.0,
      "step": 38240
    },
    {
      "entropy": 5.564098167419433,
      "epoch": 3.2130644822516277,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039819103358694404,
      "loss": 5.0579,
      "mean_token_accuracy": 0.19282338470220567,
      "num_tokens": 70581842.0,
      "step": 38245
    },
    {
      "entropy": 5.373212623596191,
      "epoch": 3.213484562066793,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003981659733610721,
      "loss": 4.954,
      "mean_token_accuracy": 0.20655377209186554,
      "num_tokens": 70590834.0,
      "step": 38250
    },
    {
      "entropy": 5.433385896682739,
      "epoch": 3.2139046418819577,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003981409109533783,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20078082084655763,
      "num_tokens": 70599739.0,
      "step": 38255
    },
    {
      "entropy": 5.462605619430542,
      "epoch": 3.2143247216971225,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039811584636430676,
      "loss": 5.0351,
      "mean_token_accuracy": 0.19840809106826782,
      "num_tokens": 70608995.0,
      "step": 38260
    },
    {
      "entropy": 5.499395561218262,
      "epoch": 3.2147448015122873,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039809077959430136,
      "loss": 5.0304,
      "mean_token_accuracy": 0.2001940056681633,
      "num_tokens": 70617598.0,
      "step": 38265
    },
    {
      "entropy": 5.495978927612304,
      "epoch": 3.215164881327452,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003980657106438062,
      "loss": 5.0402,
      "mean_token_accuracy": 0.19881621301174163,
      "num_tokens": 70627590.0,
      "step": 38270
    },
    {
      "entropy": 5.492419052124023,
      "epoch": 3.2155849611426173,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003980406395132653,
      "loss": 5.0688,
      "mean_token_accuracy": 0.1988577052950859,
      "num_tokens": 70637216.0,
      "step": 38275
    },
    {
      "entropy": 5.512527656555176,
      "epoch": 3.216005040957782,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039801556620312284,
      "loss": 5.0241,
      "mean_token_accuracy": 0.2025962382555008,
      "num_tokens": 70646330.0,
      "step": 38280
    },
    {
      "entropy": 5.514253997802735,
      "epoch": 3.216425120772947,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039799049071382296,
      "loss": 5.0497,
      "mean_token_accuracy": 0.1984148919582367,
      "num_tokens": 70655568.0,
      "step": 38285
    },
    {
      "entropy": 5.436829376220703,
      "epoch": 3.2168452005881116,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003979654130458099,
      "loss": 4.9758,
      "mean_token_accuracy": 0.20135241895914077,
      "num_tokens": 70664393.0,
      "step": 38290
    },
    {
      "entropy": 5.394458866119384,
      "epoch": 3.2172652804032764,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003979403331995279,
      "loss": 4.9559,
      "mean_token_accuracy": 0.20798639506101607,
      "num_tokens": 70672613.0,
      "step": 38295
    },
    {
      "entropy": 5.47270245552063,
      "epoch": 3.2176853602184416,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039791525117542116,
      "loss": 5.0987,
      "mean_token_accuracy": 0.20502025932073592,
      "num_tokens": 70681695.0,
      "step": 38300
    },
    {
      "entropy": 5.437195491790772,
      "epoch": 3.2181054400336064,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039789016697393406,
      "loss": 4.993,
      "mean_token_accuracy": 0.2027292490005493,
      "num_tokens": 70690871.0,
      "step": 38305
    },
    {
      "entropy": 5.443032550811767,
      "epoch": 3.218525519848771,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003978650805955109,
      "loss": 5.0032,
      "mean_token_accuracy": 0.19855969697237014,
      "num_tokens": 70699908.0,
      "step": 38310
    },
    {
      "entropy": 5.480698347091675,
      "epoch": 3.218945599663936,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039783999204059614,
      "loss": 5.0107,
      "mean_token_accuracy": 0.20936707258224488,
      "num_tokens": 70709247.0,
      "step": 38315
    },
    {
      "entropy": 5.568811321258545,
      "epoch": 3.2193656794791012,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039781490130963416,
      "loss": 5.0812,
      "mean_token_accuracy": 0.19539655596017838,
      "num_tokens": 70718304.0,
      "step": 38320
    },
    {
      "entropy": 5.5112873077392575,
      "epoch": 3.219785759294266,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003977898084030693,
      "loss": 5.0894,
      "mean_token_accuracy": 0.19515542536973954,
      "num_tokens": 70727424.0,
      "step": 38325
    },
    {
      "entropy": 5.54922308921814,
      "epoch": 3.220205839109431,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039776471332134643,
      "loss": 5.0845,
      "mean_token_accuracy": 0.19697028249502183,
      "num_tokens": 70736779.0,
      "step": 38330
    },
    {
      "entropy": 5.445277261734009,
      "epoch": 3.2206259189245956,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039773961606490984,
      "loss": 5.0075,
      "mean_token_accuracy": 0.20105798691511154,
      "num_tokens": 70745907.0,
      "step": 38335
    },
    {
      "entropy": 5.389476919174195,
      "epoch": 3.2210459987397604,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003977145166342041,
      "loss": 4.9303,
      "mean_token_accuracy": 0.205945485830307,
      "num_tokens": 70754775.0,
      "step": 38340
    },
    {
      "entropy": 5.501386976242065,
      "epoch": 3.2214660785549256,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039768941502967383,
      "loss": 5.0406,
      "mean_token_accuracy": 0.19523666948080062,
      "num_tokens": 70763544.0,
      "step": 38345
    },
    {
      "entropy": 5.394075107574463,
      "epoch": 3.2218861583700904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039766431125176383,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2069868564605713,
      "num_tokens": 70772243.0,
      "step": 38350
    },
    {
      "entropy": 5.438135290145874,
      "epoch": 3.222306238185255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039763920530091866,
      "loss": 4.9805,
      "mean_token_accuracy": 0.20090286880731584,
      "num_tokens": 70781239.0,
      "step": 38355
    },
    {
      "entropy": 5.410037279129028,
      "epoch": 3.22272631800042,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039761409717758316,
      "loss": 5.0125,
      "mean_token_accuracy": 0.20604109019041061,
      "num_tokens": 70790496.0,
      "step": 38360
    },
    {
      "entropy": 5.512104606628418,
      "epoch": 3.2231463978155848,
      "grad_norm": 1.0,
      "learning_rate": 0.00039758898688220214,
      "loss": 5.1496,
      "mean_token_accuracy": 0.1875116914510727,
      "num_tokens": 70800705.0,
      "step": 38365
    },
    {
      "entropy": 5.530015182495117,
      "epoch": 3.22356647763075,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003975638744152202,
      "loss": 4.9569,
      "mean_token_accuracy": 0.2097737595438957,
      "num_tokens": 70809347.0,
      "step": 38370
    },
    {
      "entropy": 5.466633272171021,
      "epoch": 3.223986557445915,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003975387597770824,
      "loss": 5.0103,
      "mean_token_accuracy": 0.20760903656482696,
      "num_tokens": 70817720.0,
      "step": 38375
    },
    {
      "entropy": 5.460888862609863,
      "epoch": 3.2244066372610796,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003975136429682336,
      "loss": 5.0427,
      "mean_token_accuracy": 0.20416326075792313,
      "num_tokens": 70826871.0,
      "step": 38380
    },
    {
      "entropy": 5.4933699607849125,
      "epoch": 3.2248267170762444,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003974885239891187,
      "loss": 5.0612,
      "mean_token_accuracy": 0.20361454039812088,
      "num_tokens": 70835884.0,
      "step": 38385
    },
    {
      "entropy": 5.4496935367584225,
      "epoch": 3.225246796891409,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039746340284018257,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19769803136587144,
      "num_tokens": 70844637.0,
      "step": 38390
    },
    {
      "entropy": 5.562241888046264,
      "epoch": 3.2256668767065744,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003974382795218704,
      "loss": 5.0705,
      "mean_token_accuracy": 0.19711460620164872,
      "num_tokens": 70853864.0,
      "step": 38395
    },
    {
      "entropy": 5.499105072021484,
      "epoch": 3.226086956521739,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003974131540346271,
      "loss": 5.0825,
      "mean_token_accuracy": 0.19898367673158646,
      "num_tokens": 70862380.0,
      "step": 38400
    },
    {
      "entropy": 5.520234298706055,
      "epoch": 3.226507036336904,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003973880263788979,
      "loss": 5.1119,
      "mean_token_accuracy": 0.1973409742116928,
      "num_tokens": 70871553.0,
      "step": 38405
    },
    {
      "entropy": 5.571721792221069,
      "epoch": 3.2269271161520687,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003973628965551278,
      "loss": 5.1544,
      "mean_token_accuracy": 0.192457914352417,
      "num_tokens": 70881195.0,
      "step": 38410
    },
    {
      "entropy": 5.5349424362182615,
      "epoch": 3.227347195967234,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039733776456376196,
      "loss": 5.0169,
      "mean_token_accuracy": 0.20262586027383805,
      "num_tokens": 70890968.0,
      "step": 38415
    },
    {
      "entropy": 5.428368854522705,
      "epoch": 3.2277672757823987,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003973126304052456,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20798390805721284,
      "num_tokens": 70900335.0,
      "step": 38420
    },
    {
      "entropy": 5.484357500076294,
      "epoch": 3.2281873555975635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003972874940800241,
      "loss": 5.0409,
      "mean_token_accuracy": 0.20182862877845764,
      "num_tokens": 70909718.0,
      "step": 38425
    },
    {
      "entropy": 5.413263511657715,
      "epoch": 3.2286074354127283,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003972623555885424,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20779234170913696,
      "num_tokens": 70919188.0,
      "step": 38430
    },
    {
      "entropy": 5.498701381683349,
      "epoch": 3.229027515227893,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003972372149312462,
      "loss": 5.0933,
      "mean_token_accuracy": 0.19119768291711808,
      "num_tokens": 70928246.0,
      "step": 38435
    },
    {
      "entropy": 5.544117069244384,
      "epoch": 3.2294475950430583,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039721207210858064,
      "loss": 5.0107,
      "mean_token_accuracy": 0.20352457761764525,
      "num_tokens": 70936086.0,
      "step": 38440
    },
    {
      "entropy": 5.518244504928589,
      "epoch": 3.229867674858223,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039718692712099115,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19960079491138458,
      "num_tokens": 70945408.0,
      "step": 38445
    },
    {
      "entropy": 5.480761528015137,
      "epoch": 3.230287754673388,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039716177996892315,
      "loss": 5.0433,
      "mean_token_accuracy": 0.19446030259132385,
      "num_tokens": 70954678.0,
      "step": 38450
    },
    {
      "entropy": 5.528666830062866,
      "epoch": 3.2307078344885527,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00039713663065282215,
      "loss": 5.0299,
      "mean_token_accuracy": 0.19740652292966843,
      "num_tokens": 70964645.0,
      "step": 38455
    },
    {
      "entropy": 5.424607658386231,
      "epoch": 3.231127914303718,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003971114791731336,
      "loss": 4.9405,
      "mean_token_accuracy": 0.20507633239030837,
      "num_tokens": 70973889.0,
      "step": 38460
    },
    {
      "entropy": 5.488205099105835,
      "epoch": 3.2315479941188827,
      "grad_norm": 1.21875,
      "learning_rate": 0.000397086325530303,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20037824660539627,
      "num_tokens": 70982414.0,
      "step": 38465
    },
    {
      "entropy": 5.437688398361206,
      "epoch": 3.2319680739340475,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003970611697247761,
      "loss": 5.0291,
      "mean_token_accuracy": 0.20293630212545394,
      "num_tokens": 70992332.0,
      "step": 38470
    },
    {
      "entropy": 5.350277376174927,
      "epoch": 3.2323881537492123,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003970360117569984,
      "loss": 4.9191,
      "mean_token_accuracy": 0.21233004480600357,
      "num_tokens": 71000690.0,
      "step": 38475
    },
    {
      "entropy": 5.489262342453003,
      "epoch": 3.232808233564377,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003970108516274156,
      "loss": 5.0571,
      "mean_token_accuracy": 0.2006470039486885,
      "num_tokens": 71009872.0,
      "step": 38480
    },
    {
      "entropy": 5.538922595977783,
      "epoch": 3.2332283133795423,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003969856893364734,
      "loss": 5.0452,
      "mean_token_accuracy": 0.20118416845798492,
      "num_tokens": 71018607.0,
      "step": 38485
    },
    {
      "entropy": 5.4405842304229735,
      "epoch": 3.233648393194707,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039696052488461756,
      "loss": 5.0643,
      "mean_token_accuracy": 0.20420926809310913,
      "num_tokens": 71028092.0,
      "step": 38490
    },
    {
      "entropy": 5.5170207023620605,
      "epoch": 3.234068473009872,
      "grad_norm": 0.984375,
      "learning_rate": 0.00039693535827229384,
      "loss": 5.152,
      "mean_token_accuracy": 0.19070643484592437,
      "num_tokens": 71038576.0,
      "step": 38495
    },
    {
      "entropy": 5.392761659622193,
      "epoch": 3.2344885528250367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039691018949994795,
      "loss": 4.9162,
      "mean_token_accuracy": 0.20588478744029998,
      "num_tokens": 71048358.0,
      "step": 38500
    },
    {
      "entropy": 5.4605082988739015,
      "epoch": 3.2349086326402015,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039688501856802584,
      "loss": 4.9163,
      "mean_token_accuracy": 0.20466964989900588,
      "num_tokens": 71056869.0,
      "step": 38505
    },
    {
      "entropy": 5.542162752151489,
      "epoch": 3.2353287124553667,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039685984547697343,
      "loss": 5.1624,
      "mean_token_accuracy": 0.18217868059873582,
      "num_tokens": 71067036.0,
      "step": 38510
    },
    {
      "entropy": 5.420140743255615,
      "epoch": 3.2357487922705315,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003968346702272365,
      "loss": 5.0158,
      "mean_token_accuracy": 0.19792579412460326,
      "num_tokens": 71076160.0,
      "step": 38515
    },
    {
      "entropy": 5.51313533782959,
      "epoch": 3.2361688720856963,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003968094928192613,
      "loss": 5.0249,
      "mean_token_accuracy": 0.2020052656531334,
      "num_tokens": 71085074.0,
      "step": 38520
    },
    {
      "entropy": 5.535574817657471,
      "epoch": 3.236588951900861,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003967843132534935,
      "loss": 5.0124,
      "mean_token_accuracy": 0.19837789982557297,
      "num_tokens": 71095472.0,
      "step": 38525
    },
    {
      "entropy": 5.5600255012512205,
      "epoch": 3.237009031716026,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003967591315303793,
      "loss": 5.1504,
      "mean_token_accuracy": 0.19021259844303132,
      "num_tokens": 71105483.0,
      "step": 38530
    },
    {
      "entropy": 5.454877853393555,
      "epoch": 3.237429111531191,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039673394765036483,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20599291175603868,
      "num_tokens": 71114821.0,
      "step": 38535
    },
    {
      "entropy": 5.409696245193482,
      "epoch": 3.237849191346356,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039670876161389626,
      "loss": 4.9186,
      "mean_token_accuracy": 0.20736783891916274,
      "num_tokens": 71124283.0,
      "step": 38540
    },
    {
      "entropy": 5.47367525100708,
      "epoch": 3.2382692711615206,
      "grad_norm": 1.125,
      "learning_rate": 0.00039668357342141956,
      "loss": 5.0302,
      "mean_token_accuracy": 0.1999298095703125,
      "num_tokens": 71133499.0,
      "step": 38545
    },
    {
      "entropy": 5.56892204284668,
      "epoch": 3.2386893509766854,
      "grad_norm": 1.0625,
      "learning_rate": 0.000396658383073381,
      "loss": 5.1121,
      "mean_token_accuracy": 0.1898047484457493,
      "num_tokens": 71142912.0,
      "step": 38550
    },
    {
      "entropy": 5.485276126861573,
      "epoch": 3.2391094307918507,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039663319057022683,
      "loss": 5.0671,
      "mean_token_accuracy": 0.20217240750789642,
      "num_tokens": 71152215.0,
      "step": 38555
    },
    {
      "entropy": 5.390960025787353,
      "epoch": 3.2395295106070154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003966079959124034,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20889199078083037,
      "num_tokens": 71161411.0,
      "step": 38560
    },
    {
      "entropy": 5.44251594543457,
      "epoch": 3.2399495904221802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039658279910035683,
      "loss": 5.036,
      "mean_token_accuracy": 0.19915347993373872,
      "num_tokens": 71169869.0,
      "step": 38565
    },
    {
      "entropy": 5.511502265930176,
      "epoch": 3.240369670237345,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003965576001345336,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2019523188471794,
      "num_tokens": 71179282.0,
      "step": 38570
    },
    {
      "entropy": 5.481347846984863,
      "epoch": 3.24078975005251,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039653239901538015,
      "loss": 5.0316,
      "mean_token_accuracy": 0.20530188381671904,
      "num_tokens": 71188089.0,
      "step": 38575
    },
    {
      "entropy": 5.487636709213257,
      "epoch": 3.241209829867675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003965071957433428,
      "loss": 4.9915,
      "mean_token_accuracy": 0.20845285803079605,
      "num_tokens": 71196846.0,
      "step": 38580
    },
    {
      "entropy": 5.421129417419434,
      "epoch": 3.24162990968284,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000396481990318868,
      "loss": 4.9517,
      "mean_token_accuracy": 0.20493856072425842,
      "num_tokens": 71205546.0,
      "step": 38585
    },
    {
      "entropy": 5.502717685699463,
      "epoch": 3.2420499894980046,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003964567827424023,
      "loss": 5.0445,
      "mean_token_accuracy": 0.1922112688422203,
      "num_tokens": 71214112.0,
      "step": 38590
    },
    {
      "entropy": 5.523945999145508,
      "epoch": 3.2424700693131694,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003964315730143922,
      "loss": 5.0411,
      "mean_token_accuracy": 0.19938259869813918,
      "num_tokens": 71223111.0,
      "step": 38595
    },
    {
      "entropy": 5.473343753814698,
      "epoch": 3.242890149128334,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003964063611352844,
      "loss": 5.0976,
      "mean_token_accuracy": 0.19369007647037506,
      "num_tokens": 71232696.0,
      "step": 38600
    },
    {
      "entropy": 5.514780330657959,
      "epoch": 3.2433102289434994,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003963811471055254,
      "loss": 5.0716,
      "mean_token_accuracy": 0.2014769732952118,
      "num_tokens": 71242850.0,
      "step": 38605
    },
    {
      "entropy": 5.479055547714234,
      "epoch": 3.243730308758664,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003963559309255619,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20444455742835999,
      "num_tokens": 71251475.0,
      "step": 38610
    },
    {
      "entropy": 5.361508464813232,
      "epoch": 3.244150388573829,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039633071259584054,
      "loss": 4.9319,
      "mean_token_accuracy": 0.19948916584253312,
      "num_tokens": 71260206.0,
      "step": 38615
    },
    {
      "entropy": 5.384865427017212,
      "epoch": 3.2445704683889938,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003963054921168081,
      "loss": 5.0429,
      "mean_token_accuracy": 0.19477119147777558,
      "num_tokens": 71268755.0,
      "step": 38620
    },
    {
      "entropy": 5.399696636199951,
      "epoch": 3.244990548204159,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003962802694889114,
      "loss": 4.9404,
      "mean_token_accuracy": 0.19947251081466674,
      "num_tokens": 71279530.0,
      "step": 38625
    },
    {
      "entropy": 5.506216526031494,
      "epoch": 3.245410628019324,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039625504471259705,
      "loss": 4.9999,
      "mean_token_accuracy": 0.19765326082706453,
      "num_tokens": 71288101.0,
      "step": 38630
    },
    {
      "entropy": 5.440213775634765,
      "epoch": 3.2458307078344886,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039622981778831206,
      "loss": 5.0488,
      "mean_token_accuracy": 0.1934562161564827,
      "num_tokens": 71297213.0,
      "step": 38635
    },
    {
      "entropy": 5.409008312225342,
      "epoch": 3.2462507876496534,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039620458871650327,
      "loss": 4.9817,
      "mean_token_accuracy": 0.20510858297348022,
      "num_tokens": 71306502.0,
      "step": 38640
    },
    {
      "entropy": 5.463652753829956,
      "epoch": 3.246670867464818,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00039617935749761764,
      "loss": 5.0189,
      "mean_token_accuracy": 0.19861214458942414,
      "num_tokens": 71315190.0,
      "step": 38645
    },
    {
      "entropy": 5.444293022155762,
      "epoch": 3.2470909472799834,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003961541241321021,
      "loss": 4.9298,
      "mean_token_accuracy": 0.20211730748414994,
      "num_tokens": 71325132.0,
      "step": 38650
    },
    {
      "entropy": 5.453870248794556,
      "epoch": 3.247511027095148,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039612888862040354,
      "loss": 5.0069,
      "mean_token_accuracy": 0.19406550973653794,
      "num_tokens": 71334346.0,
      "step": 38655
    },
    {
      "entropy": 5.494700479507446,
      "epoch": 3.247931106910313,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039610365096296916,
      "loss": 5.0423,
      "mean_token_accuracy": 0.20051784217357635,
      "num_tokens": 71343456.0,
      "step": 38660
    },
    {
      "entropy": 5.574425745010376,
      "epoch": 3.2483511867254777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003960784111602459,
      "loss": 5.1804,
      "mean_token_accuracy": 0.19229833036661148,
      "num_tokens": 71353164.0,
      "step": 38665
    },
    {
      "entropy": 5.410580635070801,
      "epoch": 3.2487712665406425,
      "grad_norm": 1.015625,
      "learning_rate": 0.000396053169212681,
      "loss": 4.9508,
      "mean_token_accuracy": 0.2104957014322281,
      "num_tokens": 71362601.0,
      "step": 38670
    },
    {
      "entropy": 5.449848651885986,
      "epoch": 3.2491913463558078,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039602792512072153,
      "loss": 5.0366,
      "mean_token_accuracy": 0.19195229411125184,
      "num_tokens": 71372782.0,
      "step": 38675
    },
    {
      "entropy": 5.498971891403198,
      "epoch": 3.2496114261709725,
      "grad_norm": 1.125,
      "learning_rate": 0.00039600267888481475,
      "loss": 5.0424,
      "mean_token_accuracy": 0.20197752565145494,
      "num_tokens": 71382139.0,
      "step": 38680
    },
    {
      "entropy": 5.481308221817017,
      "epoch": 3.2500315059861373,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003959774305054077,
      "loss": 5.0298,
      "mean_token_accuracy": 0.20136745274066925,
      "num_tokens": 71391503.0,
      "step": 38685
    },
    {
      "entropy": 5.423967552185059,
      "epoch": 3.250451585801302,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003959521799829478,
      "loss": 4.8591,
      "mean_token_accuracy": 0.21278308480978012,
      "num_tokens": 71399730.0,
      "step": 38690
    },
    {
      "entropy": 5.397884368896484,
      "epoch": 3.250871665616467,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003959269273178823,
      "loss": 5.0224,
      "mean_token_accuracy": 0.19874411225318908,
      "num_tokens": 71409540.0,
      "step": 38695
    },
    {
      "entropy": 5.482583522796631,
      "epoch": 3.251291745431632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003959016725106587,
      "loss": 5.1158,
      "mean_token_accuracy": 0.19810305535793304,
      "num_tokens": 71418844.0,
      "step": 38700
    },
    {
      "entropy": 5.439950084686279,
      "epoch": 3.251711825246797,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039587641556172406,
      "loss": 5.085,
      "mean_token_accuracy": 0.2047850951552391,
      "num_tokens": 71429662.0,
      "step": 38705
    },
    {
      "entropy": 5.548661804199218,
      "epoch": 3.2521319050619617,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039585115647152596,
      "loss": 5.1394,
      "mean_token_accuracy": 0.19349878877401352,
      "num_tokens": 71439828.0,
      "step": 38710
    },
    {
      "entropy": 5.584739875793457,
      "epoch": 3.2525519848771265,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003958258952405119,
      "loss": 5.1587,
      "mean_token_accuracy": 0.19693240374326706,
      "num_tokens": 71449303.0,
      "step": 38715
    },
    {
      "entropy": 5.528120422363282,
      "epoch": 3.2529720646922917,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039580063186912936,
      "loss": 5.0638,
      "mean_token_accuracy": 0.19575938880443572,
      "num_tokens": 71459401.0,
      "step": 38720
    },
    {
      "entropy": 5.470040655136108,
      "epoch": 3.2533921445074565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003957753663578258,
      "loss": 5.1065,
      "mean_token_accuracy": 0.1915103927254677,
      "num_tokens": 71468138.0,
      "step": 38725
    },
    {
      "entropy": 5.507784032821656,
      "epoch": 3.2538122243226213,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039575009870704877,
      "loss": 5.0226,
      "mean_token_accuracy": 0.19990885257720947,
      "num_tokens": 71478830.0,
      "step": 38730
    },
    {
      "entropy": 5.482938814163208,
      "epoch": 3.254232304137786,
      "grad_norm": 1.015625,
      "learning_rate": 0.000395724828917246,
      "loss": 4.9861,
      "mean_token_accuracy": 0.2020757883787155,
      "num_tokens": 71487580.0,
      "step": 38735
    },
    {
      "entropy": 5.531164503097534,
      "epoch": 3.254652383952951,
      "grad_norm": 1.125,
      "learning_rate": 0.000395699556988865,
      "loss": 5.2041,
      "mean_token_accuracy": 0.19265652298927308,
      "num_tokens": 71497156.0,
      "step": 38740
    },
    {
      "entropy": 5.450304555892944,
      "epoch": 3.255072463768116,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039567428292235345,
      "loss": 5.029,
      "mean_token_accuracy": 0.20038577020168305,
      "num_tokens": 71506621.0,
      "step": 38745
    },
    {
      "entropy": 5.532428550720215,
      "epoch": 3.255492543583281,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039564900671815916,
      "loss": 5.1462,
      "mean_token_accuracy": 0.18794961273670197,
      "num_tokens": 71516055.0,
      "step": 38750
    },
    {
      "entropy": 5.530526494979858,
      "epoch": 3.2559126233984457,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003956237283767298,
      "loss": 4.9299,
      "mean_token_accuracy": 0.2158134862780571,
      "num_tokens": 71525899.0,
      "step": 38755
    },
    {
      "entropy": 5.445235347747802,
      "epoch": 3.2563327032136105,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003955984478985133,
      "loss": 5.0128,
      "mean_token_accuracy": 0.1981353670358658,
      "num_tokens": 71534391.0,
      "step": 38760
    },
    {
      "entropy": 5.348951530456543,
      "epoch": 3.2567527830287757,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003955731652839574,
      "loss": 4.9534,
      "mean_token_accuracy": 0.20879674702882767,
      "num_tokens": 71543973.0,
      "step": 38765
    },
    {
      "entropy": 5.4233163356781,
      "epoch": 3.2571728628439405,
      "grad_norm": 1.125,
      "learning_rate": 0.0003955478805335099,
      "loss": 4.8696,
      "mean_token_accuracy": 0.20991221815347672,
      "num_tokens": 71552046.0,
      "step": 38770
    },
    {
      "entropy": 5.51062331199646,
      "epoch": 3.2575929426591053,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003955225936476188,
      "loss": 5.1446,
      "mean_token_accuracy": 0.19526879191398622,
      "num_tokens": 71561998.0,
      "step": 38775
    },
    {
      "entropy": 5.512613010406494,
      "epoch": 3.25801302247427,
      "grad_norm": 1.125,
      "learning_rate": 0.000395497304626732,
      "loss": 5.036,
      "mean_token_accuracy": 0.19348481446504592,
      "num_tokens": 71572608.0,
      "step": 38780
    },
    {
      "entropy": 5.510789966583252,
      "epoch": 3.258433102289435,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039547201347129744,
      "loss": 4.9945,
      "mean_token_accuracy": 0.19556383341550826,
      "num_tokens": 71581746.0,
      "step": 38785
    },
    {
      "entropy": 5.469992399215698,
      "epoch": 3.2588531821046,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003954467201817633,
      "loss": 5.0325,
      "mean_token_accuracy": 0.20349490642547607,
      "num_tokens": 71590460.0,
      "step": 38790
    },
    {
      "entropy": 5.481487798690796,
      "epoch": 3.259273261919765,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039542142475857745,
      "loss": 5.0665,
      "mean_token_accuracy": 0.19209987670183182,
      "num_tokens": 71600248.0,
      "step": 38795
    },
    {
      "entropy": 5.551206350326538,
      "epoch": 3.2596933417349296,
      "grad_norm": 1.140625,
      "learning_rate": 0.000395396127202188,
      "loss": 5.0638,
      "mean_token_accuracy": 0.19844427555799485,
      "num_tokens": 71610578.0,
      "step": 38800
    },
    {
      "entropy": 5.472790241241455,
      "epoch": 3.2601134215500944,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003953708275130433,
      "loss": 5.0431,
      "mean_token_accuracy": 0.19692826867103577,
      "num_tokens": 71619076.0,
      "step": 38805
    },
    {
      "entropy": 5.435970258712769,
      "epoch": 3.260533501365259,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039534552569159135,
      "loss": 4.968,
      "mean_token_accuracy": 0.20122818052768707,
      "num_tokens": 71628612.0,
      "step": 38810
    },
    {
      "entropy": 5.421100378036499,
      "epoch": 3.2609535811804244,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003953202217382803,
      "loss": 4.9284,
      "mean_token_accuracy": 0.20804428607225417,
      "num_tokens": 71638349.0,
      "step": 38815
    },
    {
      "entropy": 5.4286283493042,
      "epoch": 3.2613736609955892,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039529491565355854,
      "loss": 4.9254,
      "mean_token_accuracy": 0.20785807371139525,
      "num_tokens": 71647343.0,
      "step": 38820
    },
    {
      "entropy": 5.506438589096069,
      "epoch": 3.261793740810754,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039526960743787426,
      "loss": 5.0542,
      "mean_token_accuracy": 0.2011151507496834,
      "num_tokens": 71657453.0,
      "step": 38825
    },
    {
      "entropy": 5.421693181991577,
      "epoch": 3.262213820625919,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039524429709167586,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20560717731714248,
      "num_tokens": 71666714.0,
      "step": 38830
    },
    {
      "entropy": 5.5303326606750485,
      "epoch": 3.2626339004410836,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039521898461541163,
      "loss": 5.081,
      "mean_token_accuracy": 0.19360461384057998,
      "num_tokens": 71675634.0,
      "step": 38835
    },
    {
      "entropy": 5.451573514938355,
      "epoch": 3.263053980256249,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039519367000953,
      "loss": 5.0021,
      "mean_token_accuracy": 0.20601682513952255,
      "num_tokens": 71684252.0,
      "step": 38840
    },
    {
      "entropy": 5.428953218460083,
      "epoch": 3.2634740600714136,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003951683532744794,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21411378681659698,
      "num_tokens": 71693347.0,
      "step": 38845
    },
    {
      "entropy": 5.394435977935791,
      "epoch": 3.2638941398865784,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039514303441070826,
      "loss": 4.9656,
      "mean_token_accuracy": 0.20570858269929887,
      "num_tokens": 71701915.0,
      "step": 38850
    },
    {
      "entropy": 5.398312520980835,
      "epoch": 3.264314219701743,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039511771341866513,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20534579455852509,
      "num_tokens": 71710933.0,
      "step": 38855
    },
    {
      "entropy": 5.534540033340454,
      "epoch": 3.264734299516908,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003950923902987986,
      "loss": 5.0756,
      "mean_token_accuracy": 0.19492604434490204,
      "num_tokens": 71720111.0,
      "step": 38860
    },
    {
      "entropy": 5.55876088142395,
      "epoch": 3.265154379332073,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039506706505155716,
      "loss": 5.1286,
      "mean_token_accuracy": 0.19555114060640336,
      "num_tokens": 71730008.0,
      "step": 38865
    },
    {
      "entropy": 5.541469764709473,
      "epoch": 3.265574459147238,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039504173767738955,
      "loss": 5.0879,
      "mean_token_accuracy": 0.19867321848869324,
      "num_tokens": 71739312.0,
      "step": 38870
    },
    {
      "entropy": 5.317882776260376,
      "epoch": 3.2659945389624028,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003950164081767443,
      "loss": 4.8765,
      "mean_token_accuracy": 0.21651771962642669,
      "num_tokens": 71748866.0,
      "step": 38875
    },
    {
      "entropy": 5.445690011978149,
      "epoch": 3.2664146187775676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039499107655007024,
      "loss": 4.9224,
      "mean_token_accuracy": 0.20698915421962738,
      "num_tokens": 71757885.0,
      "step": 38880
    },
    {
      "entropy": 5.362045574188232,
      "epoch": 3.266834698592733,
      "grad_norm": 1.03125,
      "learning_rate": 0.000394965742797816,
      "loss": 4.9151,
      "mean_token_accuracy": 0.216729037463665,
      "num_tokens": 71766418.0,
      "step": 38885
    },
    {
      "entropy": 5.429295587539673,
      "epoch": 3.2672547784078976,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039494040692043044,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21482233256101607,
      "num_tokens": 71776335.0,
      "step": 38890
    },
    {
      "entropy": 5.382491683959961,
      "epoch": 3.2676748582230624,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003949150689183623,
      "loss": 4.9284,
      "mean_token_accuracy": 0.20759962350130082,
      "num_tokens": 71784731.0,
      "step": 38895
    },
    {
      "entropy": 5.493082046508789,
      "epoch": 3.268094938038227,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003948897287920605,
      "loss": 5.1097,
      "mean_token_accuracy": 0.19320838153362274,
      "num_tokens": 71795106.0,
      "step": 38900
    },
    {
      "entropy": 5.42684850692749,
      "epoch": 3.2685150178533924,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00039486438654197386,
      "loss": 4.891,
      "mean_token_accuracy": 0.2098298951983452,
      "num_tokens": 71804267.0,
      "step": 38905
    },
    {
      "entropy": 5.426208972930908,
      "epoch": 3.268935097668557,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003948390421685514,
      "loss": 4.9961,
      "mean_token_accuracy": 0.2078319624066353,
      "num_tokens": 71813754.0,
      "step": 38910
    },
    {
      "entropy": 5.6086231708526615,
      "epoch": 3.269355177483722,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003948136956722419,
      "loss": 5.2269,
      "mean_token_accuracy": 0.187793593108654,
      "num_tokens": 71823810.0,
      "step": 38915
    },
    {
      "entropy": 5.510122776031494,
      "epoch": 3.2697752572988867,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039478834705349453,
      "loss": 5.0676,
      "mean_token_accuracy": 0.19333931803703308,
      "num_tokens": 71832263.0,
      "step": 38920
    },
    {
      "entropy": 5.482314586639404,
      "epoch": 3.2701953371140515,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003947629963127583,
      "loss": 4.9867,
      "mean_token_accuracy": 0.19886838644742966,
      "num_tokens": 71841218.0,
      "step": 38925
    },
    {
      "entropy": 5.529309988021851,
      "epoch": 3.2706154169292168,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003947376434504822,
      "loss": 5.0617,
      "mean_token_accuracy": 0.19936977028846742,
      "num_tokens": 71851054.0,
      "step": 38930
    },
    {
      "entropy": 5.445240545272827,
      "epoch": 3.2710354967443815,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00039471228846711545,
      "loss": 4.9851,
      "mean_token_accuracy": 0.20163045674562455,
      "num_tokens": 71860274.0,
      "step": 38935
    },
    {
      "entropy": 5.410710573196411,
      "epoch": 3.2714555765595463,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003946869313631072,
      "loss": 5.0155,
      "mean_token_accuracy": 0.20375421345233918,
      "num_tokens": 71868806.0,
      "step": 38940
    },
    {
      "entropy": 5.431527614593506,
      "epoch": 3.271875656374711,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003946615721389065,
      "loss": 4.9651,
      "mean_token_accuracy": 0.19937784522771834,
      "num_tokens": 71877395.0,
      "step": 38945
    },
    {
      "entropy": 5.4325127601623535,
      "epoch": 3.272295736189876,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003946362107949627,
      "loss": 4.8835,
      "mean_token_accuracy": 0.20907773077487946,
      "num_tokens": 71885373.0,
      "step": 38950
    },
    {
      "entropy": 5.490934276580811,
      "epoch": 3.272715816005041,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000394610847331725,
      "loss": 5.0301,
      "mean_token_accuracy": 0.20028193593025206,
      "num_tokens": 71894392.0,
      "step": 38955
    },
    {
      "entropy": 5.453857517242431,
      "epoch": 3.273135895820206,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039458548174964276,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21017167568206788,
      "num_tokens": 71902319.0,
      "step": 38960
    },
    {
      "entropy": 5.551933288574219,
      "epoch": 3.2735559756353707,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039456011404916526,
      "loss": 5.1261,
      "mean_token_accuracy": 0.1925982803106308,
      "num_tokens": 71912196.0,
      "step": 38965
    },
    {
      "entropy": 5.547677707672119,
      "epoch": 3.2739760554505355,
      "grad_norm": 1.125,
      "learning_rate": 0.00039453474423074196,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19994420260190965,
      "num_tokens": 71920401.0,
      "step": 38970
    },
    {
      "entropy": 5.489377546310425,
      "epoch": 3.2743961352657003,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039450937229482217,
      "loss": 4.9479,
      "mean_token_accuracy": 0.2069629043340683,
      "num_tokens": 71930168.0,
      "step": 38975
    },
    {
      "entropy": 5.436282825469971,
      "epoch": 3.2748162150808655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003944839982418554,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20745370984077455,
      "num_tokens": 71939533.0,
      "step": 38980
    },
    {
      "entropy": 5.4505609512329105,
      "epoch": 3.2752362948960303,
      "grad_norm": 1.0,
      "learning_rate": 0.0003944586220722911,
      "loss": 5.09,
      "mean_token_accuracy": 0.197832290828228,
      "num_tokens": 71949843.0,
      "step": 38985
    },
    {
      "entropy": 5.599993753433227,
      "epoch": 3.275656374711195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003944332437865789,
      "loss": 5.1605,
      "mean_token_accuracy": 0.19549199789762498,
      "num_tokens": 71959581.0,
      "step": 38990
    },
    {
      "entropy": 5.464225721359253,
      "epoch": 3.27607645452636,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00039440786338516826,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20119570344686508,
      "num_tokens": 71968548.0,
      "step": 38995
    },
    {
      "entropy": 5.455074596405029,
      "epoch": 3.2764965343415247,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003943824808685087,
      "loss": 4.9912,
      "mean_token_accuracy": 0.20094634145498275,
      "num_tokens": 71977575.0,
      "step": 39000
    },
    {
      "epoch": 3.2764965343415247,
      "eval_entropy": 5.224005916597909,
      "eval_loss": 5.133132457733154,
      "eval_mean_token_accuracy": 0.20408303188201932,
      "eval_num_tokens": 71977575.0,
      "eval_runtime": 27.3707,
      "eval_samples_per_second": 1365.182,
      "eval_steps_per_second": 170.657,
      "step": 39000
    },
    {
      "entropy": 5.465867233276367,
      "epoch": 3.27691661415669,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039435709623705003,
      "loss": 5.0658,
      "mean_token_accuracy": 0.1986299932003021,
      "num_tokens": 71987312.0,
      "step": 39005
    },
    {
      "entropy": 5.510805940628051,
      "epoch": 3.2773366939718547,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003943317094912419,
      "loss": 5.0824,
      "mean_token_accuracy": 0.1990688309073448,
      "num_tokens": 71996026.0,
      "step": 39010
    },
    {
      "entropy": 5.509410095214844,
      "epoch": 3.2777567737870195,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039430632063153394,
      "loss": 4.9913,
      "mean_token_accuracy": 0.21192355155944825,
      "num_tokens": 72005027.0,
      "step": 39015
    },
    {
      "entropy": 5.446227359771728,
      "epoch": 3.2781768536021842,
      "grad_norm": 1.09375,
      "learning_rate": 0.000394280929658376,
      "loss": 4.9479,
      "mean_token_accuracy": 0.20611267834901809,
      "num_tokens": 72014428.0,
      "step": 39020
    },
    {
      "entropy": 5.492766666412353,
      "epoch": 3.2785969334173495,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039425553657221774,
      "loss": 5.0529,
      "mean_token_accuracy": 0.1997463271021843,
      "num_tokens": 72024289.0,
      "step": 39025
    },
    {
      "entropy": 5.436016464233399,
      "epoch": 3.2790170132325143,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039423014137350917,
      "loss": 5.0183,
      "mean_token_accuracy": 0.2013874188065529,
      "num_tokens": 72034015.0,
      "step": 39030
    },
    {
      "entropy": 5.509592485427857,
      "epoch": 3.279437093047679,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003942047440626999,
      "loss": 5.0511,
      "mean_token_accuracy": 0.19203657060861587,
      "num_tokens": 72042453.0,
      "step": 39035
    },
    {
      "entropy": 5.470217847824097,
      "epoch": 3.279857172862844,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039417934464024005,
      "loss": 4.9635,
      "mean_token_accuracy": 0.20411795228719712,
      "num_tokens": 72052146.0,
      "step": 39040
    },
    {
      "entropy": 5.61448073387146,
      "epoch": 3.280277252678009,
      "grad_norm": 1.125,
      "learning_rate": 0.0003941539431065795,
      "loss": 5.1103,
      "mean_token_accuracy": 0.19303927570581436,
      "num_tokens": 72060314.0,
      "step": 39045
    },
    {
      "entropy": 5.442869043350219,
      "epoch": 3.280697332493174,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039412853946216807,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21227120608091354,
      "num_tokens": 72069654.0,
      "step": 39050
    },
    {
      "entropy": 5.4539031982421875,
      "epoch": 3.2811174123083386,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039410313370745605,
      "loss": 5.0852,
      "mean_token_accuracy": 0.20714098662137986,
      "num_tokens": 72079527.0,
      "step": 39055
    },
    {
      "entropy": 5.470202541351318,
      "epoch": 3.2815374921235034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039407772584289333,
      "loss": 5.0892,
      "mean_token_accuracy": 0.19963895082473754,
      "num_tokens": 72088561.0,
      "step": 39060
    },
    {
      "entropy": 5.4886329650878904,
      "epoch": 3.281957571938668,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00039405231586892996,
      "loss": 4.9638,
      "mean_token_accuracy": 0.20045991241931915,
      "num_tokens": 72097834.0,
      "step": 39065
    },
    {
      "entropy": 5.469976997375488,
      "epoch": 3.2823776517538334,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003940269037860161,
      "loss": 5.0678,
      "mean_token_accuracy": 0.1890168532729149,
      "num_tokens": 72107508.0,
      "step": 39070
    },
    {
      "entropy": 5.444178438186645,
      "epoch": 3.2827977315689982,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003940014895946019,
      "loss": 5.0832,
      "mean_token_accuracy": 0.20726941525936127,
      "num_tokens": 72118112.0,
      "step": 39075
    },
    {
      "entropy": 5.444027662277222,
      "epoch": 3.283217811384163,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039397607329513774,
      "loss": 4.9192,
      "mean_token_accuracy": 0.20178260207176207,
      "num_tokens": 72128141.0,
      "step": 39080
    },
    {
      "entropy": 5.475773286819458,
      "epoch": 3.283637891199328,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003939506548880736,
      "loss": 4.98,
      "mean_token_accuracy": 0.20398510694503785,
      "num_tokens": 72137847.0,
      "step": 39085
    },
    {
      "entropy": 5.421097755432129,
      "epoch": 3.2840579710144926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039392523437385983,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21468516141176225,
      "num_tokens": 72147026.0,
      "step": 39090
    },
    {
      "entropy": 5.405813837051392,
      "epoch": 3.284478050829658,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039389981175294683,
      "loss": 4.9712,
      "mean_token_accuracy": 0.20319415479898453,
      "num_tokens": 72157091.0,
      "step": 39095
    },
    {
      "entropy": 5.433291959762573,
      "epoch": 3.2848981306448226,
      "grad_norm": 1.125,
      "learning_rate": 0.0003938743870257849,
      "loss": 5.0024,
      "mean_token_accuracy": 0.20442201048135758,
      "num_tokens": 72165426.0,
      "step": 39100
    },
    {
      "entropy": 5.4348835945129395,
      "epoch": 3.2853182104599874,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003938489601928244,
      "loss": 5.0549,
      "mean_token_accuracy": 0.20341161489486695,
      "num_tokens": 72174357.0,
      "step": 39105
    },
    {
      "entropy": 5.465460205078125,
      "epoch": 3.285738290275152,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039382353125451576,
      "loss": 5.0093,
      "mean_token_accuracy": 0.20122134685516357,
      "num_tokens": 72183645.0,
      "step": 39110
    },
    {
      "entropy": 5.489626121520996,
      "epoch": 3.286158370090317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039379810021130947,
      "loss": 5.069,
      "mean_token_accuracy": 0.19541551023721696,
      "num_tokens": 72192804.0,
      "step": 39115
    },
    {
      "entropy": 5.515480899810791,
      "epoch": 3.286578449905482,
      "grad_norm": 0.984375,
      "learning_rate": 0.00039377266706365596,
      "loss": 5.0254,
      "mean_token_accuracy": 0.1969981849193573,
      "num_tokens": 72202592.0,
      "step": 39120
    },
    {
      "entropy": 5.4666389465332035,
      "epoch": 3.286998529720647,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039374723181200585,
      "loss": 4.9624,
      "mean_token_accuracy": 0.20995834916830064,
      "num_tokens": 72211561.0,
      "step": 39125
    },
    {
      "entropy": 5.472526264190674,
      "epoch": 3.2874186095358118,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003937217944568096,
      "loss": 5.041,
      "mean_token_accuracy": 0.19691554605960845,
      "num_tokens": 72220960.0,
      "step": 39130
    },
    {
      "entropy": 5.568407440185547,
      "epoch": 3.2878386893509766,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003936963549985179,
      "loss": 5.1993,
      "mean_token_accuracy": 0.19464983195066451,
      "num_tokens": 72229925.0,
      "step": 39135
    },
    {
      "entropy": 5.458086633682251,
      "epoch": 3.2882587691661413,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003936709134375815,
      "loss": 4.967,
      "mean_token_accuracy": 0.20444717407226562,
      "num_tokens": 72238525.0,
      "step": 39140
    },
    {
      "entropy": 5.4491167068481445,
      "epoch": 3.2886788489813066,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003936454697744509,
      "loss": 5.0429,
      "mean_token_accuracy": 0.20057573765516282,
      "num_tokens": 72247882.0,
      "step": 39145
    },
    {
      "entropy": 5.455700922012329,
      "epoch": 3.2890989287964714,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039362002400957695,
      "loss": 4.9879,
      "mean_token_accuracy": 0.2047765925526619,
      "num_tokens": 72256672.0,
      "step": 39150
    },
    {
      "entropy": 5.49384126663208,
      "epoch": 3.289519008611636,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039359457614341026,
      "loss": 5.0264,
      "mean_token_accuracy": 0.2045904889702797,
      "num_tokens": 72265516.0,
      "step": 39155
    },
    {
      "entropy": 5.372196912765503,
      "epoch": 3.289939088426801,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003935691261764018,
      "loss": 4.9883,
      "mean_token_accuracy": 0.20066849291324615,
      "num_tokens": 72274724.0,
      "step": 39160
    },
    {
      "entropy": 5.488720178604126,
      "epoch": 3.2903591682419657,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039354367410900234,
      "loss": 5.044,
      "mean_token_accuracy": 0.20088207572698594,
      "num_tokens": 72283768.0,
      "step": 39165
    },
    {
      "entropy": 5.431426239013672,
      "epoch": 3.290779248057131,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003935182199416627,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2034145787358284,
      "num_tokens": 72293575.0,
      "step": 39170
    },
    {
      "entropy": 5.496453046798706,
      "epoch": 3.2911993278722957,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003934927636748338,
      "loss": 5.0166,
      "mean_token_accuracy": 0.1969471126794815,
      "num_tokens": 72302538.0,
      "step": 39175
    },
    {
      "entropy": 5.500840187072754,
      "epoch": 3.2916194076874605,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039346730530896663,
      "loss": 5.0435,
      "mean_token_accuracy": 0.19919968545436859,
      "num_tokens": 72310919.0,
      "step": 39180
    },
    {
      "entropy": 5.404334449768067,
      "epoch": 3.2920394875026253,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003934418448445122,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21092341840267181,
      "num_tokens": 72320172.0,
      "step": 39185
    },
    {
      "entropy": 5.487982654571534,
      "epoch": 3.2924595673177905,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003934163822819214,
      "loss": 5.0147,
      "mean_token_accuracy": 0.20543356090784073,
      "num_tokens": 72328903.0,
      "step": 39190
    },
    {
      "entropy": 5.488181018829346,
      "epoch": 3.2928796471329553,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003933909176216454,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20868505984544755,
      "num_tokens": 72337723.0,
      "step": 39195
    },
    {
      "entropy": 5.448330879211426,
      "epoch": 3.29329972694812,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039336545086413516,
      "loss": 4.9405,
      "mean_token_accuracy": 0.21833205670118333,
      "num_tokens": 72345923.0,
      "step": 39200
    },
    {
      "entropy": 5.396405982971191,
      "epoch": 3.293719806763285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003933399820098421,
      "loss": 4.9932,
      "mean_token_accuracy": 0.2004032239317894,
      "num_tokens": 72354258.0,
      "step": 39205
    },
    {
      "entropy": 5.485851764678955,
      "epoch": 3.29413988657845,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039331451105921694,
      "loss": 5.0276,
      "mean_token_accuracy": 0.20939842462539673,
      "num_tokens": 72363102.0,
      "step": 39210
    },
    {
      "entropy": 5.54050612449646,
      "epoch": 3.294559966393615,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003932890380127113,
      "loss": 5.007,
      "mean_token_accuracy": 0.2017019122838974,
      "num_tokens": 72371952.0,
      "step": 39215
    },
    {
      "entropy": 5.47012939453125,
      "epoch": 3.2949800462087797,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003932635628707763,
      "loss": 4.9948,
      "mean_token_accuracy": 0.20287920534610748,
      "num_tokens": 72380358.0,
      "step": 39220
    },
    {
      "entropy": 5.440503883361816,
      "epoch": 3.2954001260239445,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039323808563386314,
      "loss": 5.0183,
      "mean_token_accuracy": 0.19754156470298767,
      "num_tokens": 72391226.0,
      "step": 39225
    },
    {
      "entropy": 5.5034104347229,
      "epoch": 3.2958202058391093,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039321260630242313,
      "loss": 5.0851,
      "mean_token_accuracy": 0.19541579633951187,
      "num_tokens": 72401241.0,
      "step": 39230
    },
    {
      "entropy": 5.576357221603393,
      "epoch": 3.2962402856542745,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003931871248769077,
      "loss": 5.1117,
      "mean_token_accuracy": 0.19794097691774368,
      "num_tokens": 72410831.0,
      "step": 39235
    },
    {
      "entropy": 5.533256912231446,
      "epoch": 3.2966603654694393,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003931616413577682,
      "loss": 5.0804,
      "mean_token_accuracy": 0.19481770992279052,
      "num_tokens": 72420435.0,
      "step": 39240
    },
    {
      "entropy": 5.480136203765869,
      "epoch": 3.297080445284604,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039313615574545607,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20749178826808928,
      "num_tokens": 72430358.0,
      "step": 39245
    },
    {
      "entropy": 5.445224094390869,
      "epoch": 3.297500525099769,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003931106680404228,
      "loss": 4.9396,
      "mean_token_accuracy": 0.20232960134744643,
      "num_tokens": 72439298.0,
      "step": 39250
    },
    {
      "entropy": 5.452116203308106,
      "epoch": 3.2979206049149337,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039308517824311986,
      "loss": 4.9556,
      "mean_token_accuracy": 0.2068701907992363,
      "num_tokens": 72448195.0,
      "step": 39255
    },
    {
      "entropy": 5.399233388900757,
      "epoch": 3.298340684730099,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039305968635399883,
      "loss": 5.0242,
      "mean_token_accuracy": 0.1960566222667694,
      "num_tokens": 72457463.0,
      "step": 39260
    },
    {
      "entropy": 5.55393648147583,
      "epoch": 3.2987607645452637,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003930341923735111,
      "loss": 5.1126,
      "mean_token_accuracy": 0.19348105490207673,
      "num_tokens": 72467447.0,
      "step": 39265
    },
    {
      "entropy": 5.543250846862793,
      "epoch": 3.2991808443604285,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003930086963021085,
      "loss": 5.1243,
      "mean_token_accuracy": 0.19608206897974015,
      "num_tokens": 72477926.0,
      "step": 39270
    },
    {
      "entropy": 5.499228429794312,
      "epoch": 3.2996009241755933,
      "grad_norm": 1.125,
      "learning_rate": 0.00039298319814024267,
      "loss": 4.9713,
      "mean_token_accuracy": 0.21072871536016463,
      "num_tokens": 72486045.0,
      "step": 39275
    },
    {
      "entropy": 5.432867527008057,
      "epoch": 3.300021003990758,
      "grad_norm": 1.125,
      "learning_rate": 0.00039295769788836514,
      "loss": 4.9803,
      "mean_token_accuracy": 0.2061684548854828,
      "num_tokens": 72495451.0,
      "step": 39280
    },
    {
      "entropy": 5.494841527938843,
      "epoch": 3.3004410838059233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003929321955469279,
      "loss": 5.0845,
      "mean_token_accuracy": 0.19405404776334761,
      "num_tokens": 72505360.0,
      "step": 39285
    },
    {
      "entropy": 5.367766380310059,
      "epoch": 3.300861163621088,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003929066911163824,
      "loss": 4.9,
      "mean_token_accuracy": 0.21148720383644104,
      "num_tokens": 72514389.0,
      "step": 39290
    },
    {
      "entropy": 5.519782972335816,
      "epoch": 3.301281243436253,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003928811845971806,
      "loss": 5.1385,
      "mean_token_accuracy": 0.18809567838907243,
      "num_tokens": 72523832.0,
      "step": 39295
    },
    {
      "entropy": 5.560894203186035,
      "epoch": 3.3017013232514176,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039285567598977436,
      "loss": 5.0584,
      "mean_token_accuracy": 0.21150924414396285,
      "num_tokens": 72533096.0,
      "step": 39300
    },
    {
      "entropy": 5.516732692718506,
      "epoch": 3.3021214030665824,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039283016529461544,
      "loss": 5.0108,
      "mean_token_accuracy": 0.2007792264223099,
      "num_tokens": 72542493.0,
      "step": 39305
    },
    {
      "entropy": 5.442561960220337,
      "epoch": 3.3025414828817476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003928046525121559,
      "loss": 5.0271,
      "mean_token_accuracy": 0.20532561093568802,
      "num_tokens": 72551388.0,
      "step": 39310
    },
    {
      "entropy": 5.488917684555053,
      "epoch": 3.3029615626969124,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003927791376428476,
      "loss": 4.9974,
      "mean_token_accuracy": 0.19788373559713363,
      "num_tokens": 72560535.0,
      "step": 39315
    },
    {
      "entropy": 5.5197502136230465,
      "epoch": 3.303381642512077,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039275362068714244,
      "loss": 5.1107,
      "mean_token_accuracy": 0.19530973136425017,
      "num_tokens": 72571040.0,
      "step": 39320
    },
    {
      "entropy": 5.545066833496094,
      "epoch": 3.303801722327242,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003927281016454925,
      "loss": 5.1118,
      "mean_token_accuracy": 0.19902200251817703,
      "num_tokens": 72580296.0,
      "step": 39325
    },
    {
      "entropy": 5.5736278057098385,
      "epoch": 3.3042218021424072,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039270258051834995,
      "loss": 5.0835,
      "mean_token_accuracy": 0.1910419911146164,
      "num_tokens": 72589719.0,
      "step": 39330
    },
    {
      "entropy": 5.498506546020508,
      "epoch": 3.304641881957572,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003926770573061668,
      "loss": 5.0118,
      "mean_token_accuracy": 0.20172821283340453,
      "num_tokens": 72597855.0,
      "step": 39335
    },
    {
      "entropy": 5.506922101974487,
      "epoch": 3.305061961772737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003926515320093951,
      "loss": 5.0887,
      "mean_token_accuracy": 0.20082094371318818,
      "num_tokens": 72606433.0,
      "step": 39340
    },
    {
      "entropy": 5.531206130981445,
      "epoch": 3.3054820415879016,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039262600462848713,
      "loss": 5.0782,
      "mean_token_accuracy": 0.20335006564855576,
      "num_tokens": 72615988.0,
      "step": 39345
    },
    {
      "entropy": 5.451373672485351,
      "epoch": 3.305902121403067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000392600475163895,
      "loss": 4.888,
      "mean_token_accuracy": 0.21564795672893525,
      "num_tokens": 72625612.0,
      "step": 39350
    },
    {
      "entropy": 5.399855279922486,
      "epoch": 3.3063222012182316,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00039257494361607105,
      "loss": 4.9747,
      "mean_token_accuracy": 0.1973452553153038,
      "num_tokens": 72635822.0,
      "step": 39355
    },
    {
      "entropy": 5.492930126190186,
      "epoch": 3.3067422810333964,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039254940998546757,
      "loss": 5.0713,
      "mean_token_accuracy": 0.19453591406345366,
      "num_tokens": 72645554.0,
      "step": 39360
    },
    {
      "entropy": 5.570063352584839,
      "epoch": 3.307162360848561,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039252387427253684,
      "loss": 5.1432,
      "mean_token_accuracy": 0.18652114421129226,
      "num_tokens": 72655308.0,
      "step": 39365
    },
    {
      "entropy": 5.4510925769805905,
      "epoch": 3.307582440663726,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000392498336477731,
      "loss": 5.0501,
      "mean_token_accuracy": 0.20214925706386566,
      "num_tokens": 72664155.0,
      "step": 39370
    },
    {
      "entropy": 5.437160301208496,
      "epoch": 3.308002520478891,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039247279660150285,
      "loss": 5.0038,
      "mean_token_accuracy": 0.19994151294231416,
      "num_tokens": 72673607.0,
      "step": 39375
    },
    {
      "entropy": 5.460023021697998,
      "epoch": 3.308422600294056,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003924472546443045,
      "loss": 5.0897,
      "mean_token_accuracy": 0.19616836160421372,
      "num_tokens": 72682427.0,
      "step": 39380
    },
    {
      "entropy": 5.444723176956177,
      "epoch": 3.308842680109221,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039242171060658855,
      "loss": 4.9867,
      "mean_token_accuracy": 0.20638588070869446,
      "num_tokens": 72691178.0,
      "step": 39385
    },
    {
      "entropy": 5.584323406219482,
      "epoch": 3.3092627599243856,
      "grad_norm": 1.125,
      "learning_rate": 0.00039239616448880744,
      "loss": 5.0647,
      "mean_token_accuracy": 0.2044753834605217,
      "num_tokens": 72699729.0,
      "step": 39390
    },
    {
      "entropy": 5.589071130752563,
      "epoch": 3.3096828397395504,
      "grad_norm": 1.125,
      "learning_rate": 0.00039237061629141375,
      "loss": 5.216,
      "mean_token_accuracy": 0.1929125517606735,
      "num_tokens": 72709480.0,
      "step": 39395
    },
    {
      "entropy": 5.459353113174439,
      "epoch": 3.3101029195547156,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039234506601485997,
      "loss": 4.9686,
      "mean_token_accuracy": 0.2084002122282982,
      "num_tokens": 72718672.0,
      "step": 39400
    },
    {
      "entropy": 5.506574296951294,
      "epoch": 3.3105229993698804,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039231951365959876,
      "loss": 5.01,
      "mean_token_accuracy": 0.2016291305422783,
      "num_tokens": 72727362.0,
      "step": 39405
    },
    {
      "entropy": 5.473837423324585,
      "epoch": 3.310943079185045,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003922939592260828,
      "loss": 4.9356,
      "mean_token_accuracy": 0.20765947699546813,
      "num_tokens": 72735981.0,
      "step": 39410
    },
    {
      "entropy": 5.454399442672729,
      "epoch": 3.31136315900021,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00039226840271476483,
      "loss": 5.0148,
      "mean_token_accuracy": 0.20438583791255951,
      "num_tokens": 72745812.0,
      "step": 39415
    },
    {
      "entropy": 5.465248155593872,
      "epoch": 3.3117832388153747,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039224284412609737,
      "loss": 5.0712,
      "mean_token_accuracy": 0.19898517280817032,
      "num_tokens": 72754896.0,
      "step": 39420
    },
    {
      "entropy": 5.42842378616333,
      "epoch": 3.31220331863054,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039221728346053335,
      "loss": 5.0005,
      "mean_token_accuracy": 0.19613821655511857,
      "num_tokens": 72763941.0,
      "step": 39425
    },
    {
      "entropy": 5.506424379348755,
      "epoch": 3.3126233984457047,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039219172071852556,
      "loss": 5.0723,
      "mean_token_accuracy": 0.20268739014863968,
      "num_tokens": 72773694.0,
      "step": 39430
    },
    {
      "entropy": 5.492679834365845,
      "epoch": 3.3130434782608695,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003921661559005267,
      "loss": 5.0184,
      "mean_token_accuracy": 0.21079789102077484,
      "num_tokens": 72783573.0,
      "step": 39435
    },
    {
      "entropy": 5.55807671546936,
      "epoch": 3.3134635580760343,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039214058900698974,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20024627596139907,
      "num_tokens": 72792996.0,
      "step": 39440
    },
    {
      "entropy": 5.667859220504761,
      "epoch": 3.313883637891199,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003921150200383676,
      "loss": 5.2489,
      "mean_token_accuracy": 0.18123804181814193,
      "num_tokens": 72804100.0,
      "step": 39445
    },
    {
      "entropy": 5.464576959609985,
      "epoch": 3.3143037177063643,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039208944899511315,
      "loss": 4.953,
      "mean_token_accuracy": 0.20350924134254456,
      "num_tokens": 72813300.0,
      "step": 39450
    },
    {
      "entropy": 5.589162302017212,
      "epoch": 3.314723797521529,
      "grad_norm": 0.953125,
      "learning_rate": 0.00039206387587767947,
      "loss": 5.1355,
      "mean_token_accuracy": 0.20169551223516463,
      "num_tokens": 72823714.0,
      "step": 39455
    },
    {
      "entropy": 5.4210395336151125,
      "epoch": 3.315143877336694,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00039203830068651946,
      "loss": 4.9161,
      "mean_token_accuracy": 0.2110142543911934,
      "num_tokens": 72833002.0,
      "step": 39460
    },
    {
      "entropy": 5.545590877532959,
      "epoch": 3.3155639571518587,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003920127234220862,
      "loss": 5.1083,
      "mean_token_accuracy": 0.19474445879459382,
      "num_tokens": 72842036.0,
      "step": 39465
    },
    {
      "entropy": 5.458925056457519,
      "epoch": 3.3159840369670235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039198714408483277,
      "loss": 4.9811,
      "mean_token_accuracy": 0.2063574656844139,
      "num_tokens": 72850673.0,
      "step": 39470
    },
    {
      "entropy": 5.494983911514282,
      "epoch": 3.3164041167821887,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003919615626752124,
      "loss": 5.0811,
      "mean_token_accuracy": 0.19670623391866685,
      "num_tokens": 72859530.0,
      "step": 39475
    },
    {
      "entropy": 5.456060075759888,
      "epoch": 3.3168241965973535,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003919359791936781,
      "loss": 5.0196,
      "mean_token_accuracy": 0.20299796462059022,
      "num_tokens": 72868373.0,
      "step": 39480
    },
    {
      "entropy": 5.509439706802368,
      "epoch": 3.3172442764125183,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003919103936406831,
      "loss": 5.0336,
      "mean_token_accuracy": 0.19916834980249404,
      "num_tokens": 72877861.0,
      "step": 39485
    },
    {
      "entropy": 5.4317961692810055,
      "epoch": 3.317664356227683,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039188480601668087,
      "loss": 5.0132,
      "mean_token_accuracy": 0.20705768316984177,
      "num_tokens": 72887001.0,
      "step": 39490
    },
    {
      "entropy": 5.459163808822632,
      "epoch": 3.3180844360428483,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003918592163221243,
      "loss": 5.0067,
      "mean_token_accuracy": 0.20404226779937745,
      "num_tokens": 72895776.0,
      "step": 39495
    },
    {
      "entropy": 5.412143325805664,
      "epoch": 3.318504515858013,
      "grad_norm": 1.046875,
      "learning_rate": 0.000391833624557467,
      "loss": 4.9422,
      "mean_token_accuracy": 0.20344126969575882,
      "num_tokens": 72905129.0,
      "step": 39500
    },
    {
      "entropy": 5.436256313323975,
      "epoch": 3.318924595673178,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003918080307231622,
      "loss": 4.99,
      "mean_token_accuracy": 0.1970217138528824,
      "num_tokens": 72914223.0,
      "step": 39505
    },
    {
      "entropy": 5.438193225860596,
      "epoch": 3.3193446754883427,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039178243481966326,
      "loss": 4.9397,
      "mean_token_accuracy": 0.20183008760213852,
      "num_tokens": 72922653.0,
      "step": 39510
    },
    {
      "entropy": 5.411322116851807,
      "epoch": 3.319764755303508,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003917568368474236,
      "loss": 4.9531,
      "mean_token_accuracy": 0.20327755361795424,
      "num_tokens": 72931663.0,
      "step": 39515
    },
    {
      "entropy": 5.533938646316528,
      "epoch": 3.3201848351186727,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00039173123680689666,
      "loss": 5.121,
      "mean_token_accuracy": 0.1923891469836235,
      "num_tokens": 72941610.0,
      "step": 39520
    },
    {
      "entropy": 5.563170528411865,
      "epoch": 3.3206049149338375,
      "grad_norm": 1.078125,
      "learning_rate": 0.000391705634698536,
      "loss": 5.1362,
      "mean_token_accuracy": 0.195901620388031,
      "num_tokens": 72950082.0,
      "step": 39525
    },
    {
      "entropy": 5.5445574760437015,
      "epoch": 3.3210249947490023,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039168003052279517,
      "loss": 5.0932,
      "mean_token_accuracy": 0.1973117023706436,
      "num_tokens": 72959145.0,
      "step": 39530
    },
    {
      "entropy": 5.439287090301514,
      "epoch": 3.321445074564167,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003916544242801276,
      "loss": 4.9721,
      "mean_token_accuracy": 0.20665373355150224,
      "num_tokens": 72968033.0,
      "step": 39535
    },
    {
      "entropy": 5.41987943649292,
      "epoch": 3.3218651543793323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039162881597098705,
      "loss": 4.9929,
      "mean_token_accuracy": 0.19790396243333816,
      "num_tokens": 72977201.0,
      "step": 39540
    },
    {
      "entropy": 5.510673713684082,
      "epoch": 3.322285234194497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039160320559582695,
      "loss": 4.9891,
      "mean_token_accuracy": 0.2038276359438896,
      "num_tokens": 72986132.0,
      "step": 39545
    },
    {
      "entropy": 5.479355955123902,
      "epoch": 3.322705314009662,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003915775931551012,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20068728774785996,
      "num_tokens": 72995406.0,
      "step": 39550
    },
    {
      "entropy": 5.434365224838257,
      "epoch": 3.3231253938248266,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003915519786492634,
      "loss": 4.955,
      "mean_token_accuracy": 0.21250090897083282,
      "num_tokens": 73004423.0,
      "step": 39555
    },
    {
      "entropy": 5.422234582901001,
      "epoch": 3.3235454736399914,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039152636207876736,
      "loss": 5.0494,
      "mean_token_accuracy": 0.1966592103242874,
      "num_tokens": 73013766.0,
      "step": 39560
    },
    {
      "entropy": 5.498662900924683,
      "epoch": 3.3239655534551567,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039150074344406675,
      "loss": 5.0354,
      "mean_token_accuracy": 0.19781696200370788,
      "num_tokens": 73022918.0,
      "step": 39565
    },
    {
      "entropy": 5.509663915634155,
      "epoch": 3.3243856332703214,
      "grad_norm": 1.0,
      "learning_rate": 0.00039147512274561555,
      "loss": 5.0617,
      "mean_token_accuracy": 0.20077135115861894,
      "num_tokens": 73032792.0,
      "step": 39570
    },
    {
      "entropy": 5.529039669036865,
      "epoch": 3.3248057130854862,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00039144949998386744,
      "loss": 5.0198,
      "mean_token_accuracy": 0.2000294879078865,
      "num_tokens": 73042458.0,
      "step": 39575
    },
    {
      "entropy": 5.420677852630615,
      "epoch": 3.325225792900651,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003914238751592764,
      "loss": 5.0116,
      "mean_token_accuracy": 0.19686253368854523,
      "num_tokens": 73052440.0,
      "step": 39580
    },
    {
      "entropy": 5.416725063323975,
      "epoch": 3.325645872715816,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003913982482722964,
      "loss": 5.0025,
      "mean_token_accuracy": 0.2074135422706604,
      "num_tokens": 73060953.0,
      "step": 39585
    },
    {
      "entropy": 5.419241809844971,
      "epoch": 3.326065952530981,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039137261932338134,
      "loss": 5.0361,
      "mean_token_accuracy": 0.19812024384737015,
      "num_tokens": 73069477.0,
      "step": 39590
    },
    {
      "entropy": 5.511202716827393,
      "epoch": 3.326486032346146,
      "grad_norm": 0.984375,
      "learning_rate": 0.00039134698831298526,
      "loss": 5.026,
      "mean_token_accuracy": 0.2031167969107628,
      "num_tokens": 73078470.0,
      "step": 39595
    },
    {
      "entropy": 5.370931720733642,
      "epoch": 3.3269061121613106,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003913213552415622,
      "loss": 4.901,
      "mean_token_accuracy": 0.21262763142585756,
      "num_tokens": 73086755.0,
      "step": 39600
    },
    {
      "entropy": 5.451963710784912,
      "epoch": 3.3273261919764754,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039129572010956627,
      "loss": 5.1256,
      "mean_token_accuracy": 0.19932074397802352,
      "num_tokens": 73096022.0,
      "step": 39605
    },
    {
      "entropy": 5.461871004104614,
      "epoch": 3.32774627179164,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039127008291745143,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20329087525606154,
      "num_tokens": 73104352.0,
      "step": 39610
    },
    {
      "entropy": 5.5529945373535154,
      "epoch": 3.3281663516068054,
      "grad_norm": 1.125,
      "learning_rate": 0.0003912444436656721,
      "loss": 5.0723,
      "mean_token_accuracy": 0.1985785722732544,
      "num_tokens": 73113277.0,
      "step": 39615
    },
    {
      "entropy": 5.443260145187378,
      "epoch": 3.32858643142197,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039121880235468213,
      "loss": 4.9849,
      "mean_token_accuracy": 0.20552826821804046,
      "num_tokens": 73123450.0,
      "step": 39620
    },
    {
      "entropy": 5.424990463256836,
      "epoch": 3.329006511237135,
      "grad_norm": 1.140625,
      "learning_rate": 0.000391193158984936,
      "loss": 4.9262,
      "mean_token_accuracy": 0.2046126201748848,
      "num_tokens": 73133274.0,
      "step": 39625
    },
    {
      "entropy": 5.445368003845215,
      "epoch": 3.3294265910522998,
      "grad_norm": 1.0625,
      "learning_rate": 0.000391167513556888,
      "loss": 5.0548,
      "mean_token_accuracy": 0.1971559301018715,
      "num_tokens": 73142297.0,
      "step": 39630
    },
    {
      "entropy": 5.484339618682862,
      "epoch": 3.329846670867465,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039114186607099217,
      "loss": 4.9656,
      "mean_token_accuracy": 0.20992633253335952,
      "num_tokens": 73151232.0,
      "step": 39635
    },
    {
      "entropy": 5.48458776473999,
      "epoch": 3.33026675068263,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039111621652770294,
      "loss": 5.0701,
      "mean_token_accuracy": 0.2035965919494629,
      "num_tokens": 73160802.0,
      "step": 39640
    },
    {
      "entropy": 5.384565925598144,
      "epoch": 3.3306868304977946,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003910905649274749,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2099968671798706,
      "num_tokens": 73169105.0,
      "step": 39645
    },
    {
      "entropy": 5.445176649093628,
      "epoch": 3.3311069103129594,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039106491127076213,
      "loss": 5.0768,
      "mean_token_accuracy": 0.20586573779582978,
      "num_tokens": 73177253.0,
      "step": 39650
    },
    {
      "entropy": 5.49315299987793,
      "epoch": 3.3315269901281246,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00039103925555801933,
      "loss": 5.0162,
      "mean_token_accuracy": 0.1966431438922882,
      "num_tokens": 73186276.0,
      "step": 39655
    },
    {
      "entropy": 5.537379455566406,
      "epoch": 3.3319470699432894,
      "grad_norm": 1.125,
      "learning_rate": 0.00039101359778970083,
      "loss": 5.1112,
      "mean_token_accuracy": 0.19408504217863082,
      "num_tokens": 73196248.0,
      "step": 39660
    },
    {
      "entropy": 5.395159339904785,
      "epoch": 3.332367149758454,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003909879379662612,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20683045089244842,
      "num_tokens": 73205262.0,
      "step": 39665
    },
    {
      "entropy": 5.411399841308594,
      "epoch": 3.332787229573619,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039096227608815496,
      "loss": 4.9186,
      "mean_token_accuracy": 0.2111225828528404,
      "num_tokens": 73214713.0,
      "step": 39670
    },
    {
      "entropy": 5.418216133117676,
      "epoch": 3.3332073093887837,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039093661215583664,
      "loss": 5.001,
      "mean_token_accuracy": 0.20157690346240997,
      "num_tokens": 73223043.0,
      "step": 39675
    },
    {
      "entropy": 5.482395982742309,
      "epoch": 3.333627389203949,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039091094616976104,
      "loss": 5.1129,
      "mean_token_accuracy": 0.1947149306535721,
      "num_tokens": 73232868.0,
      "step": 39680
    },
    {
      "entropy": 5.495424890518189,
      "epoch": 3.3340474690191138,
      "grad_norm": 1.125,
      "learning_rate": 0.0003908852781303826,
      "loss": 5.0645,
      "mean_token_accuracy": 0.2084543749690056,
      "num_tokens": 73242094.0,
      "step": 39685
    },
    {
      "entropy": 5.478791856765747,
      "epoch": 3.3344675488342785,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039085960803815625,
      "loss": 5.0071,
      "mean_token_accuracy": 0.19968540370464324,
      "num_tokens": 73251499.0,
      "step": 39690
    },
    {
      "entropy": 5.515465974807739,
      "epoch": 3.3348876286494433,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039083393589353654,
      "loss": 5.0683,
      "mean_token_accuracy": 0.19945953637361527,
      "num_tokens": 73260920.0,
      "step": 39695
    },
    {
      "entropy": 5.4874547004699705,
      "epoch": 3.335307708464608,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003908082616969783,
      "loss": 5.1092,
      "mean_token_accuracy": 0.20116014778614044,
      "num_tokens": 73269775.0,
      "step": 39700
    },
    {
      "entropy": 5.571578073501587,
      "epoch": 3.3357277882797733,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039078258544893635,
      "loss": 5.0812,
      "mean_token_accuracy": 0.19794175326824187,
      "num_tokens": 73278678.0,
      "step": 39705
    },
    {
      "entropy": 5.40465784072876,
      "epoch": 3.336147868094938,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003907569071498655,
      "loss": 4.9613,
      "mean_token_accuracy": 0.20367953926324844,
      "num_tokens": 73287579.0,
      "step": 39710
    },
    {
      "entropy": 5.461197280883789,
      "epoch": 3.336567947910103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003907312268002207,
      "loss": 5.0312,
      "mean_token_accuracy": 0.20665702670812608,
      "num_tokens": 73297006.0,
      "step": 39715
    },
    {
      "entropy": 5.432220363616944,
      "epoch": 3.3369880277252677,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039070554440045677,
      "loss": 5.0462,
      "mean_token_accuracy": 0.20642297863960266,
      "num_tokens": 73306147.0,
      "step": 39720
    },
    {
      "entropy": 5.4759620189666744,
      "epoch": 3.3374081075404325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003906798599510287,
      "loss": 5.0589,
      "mean_token_accuracy": 0.19657204449176788,
      "num_tokens": 73315584.0,
      "step": 39725
    },
    {
      "entropy": 5.465712070465088,
      "epoch": 3.3378281873555977,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039065417345239144,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20486307740211487,
      "num_tokens": 73324756.0,
      "step": 39730
    },
    {
      "entropy": 5.492051267623902,
      "epoch": 3.3382482671707625,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003906284849050001,
      "loss": 5.0266,
      "mean_token_accuracy": 0.1983754813671112,
      "num_tokens": 73334177.0,
      "step": 39735
    },
    {
      "entropy": 5.43417272567749,
      "epoch": 3.3386683469859273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039060279430930966,
      "loss": 5.0045,
      "mean_token_accuracy": 0.20428377091884614,
      "num_tokens": 73343965.0,
      "step": 39740
    },
    {
      "entropy": 5.4432323455810545,
      "epoch": 3.339088426801092,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039057710166577527,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20602914541959763,
      "num_tokens": 73352853.0,
      "step": 39745
    },
    {
      "entropy": 5.395395183563233,
      "epoch": 3.339508506616257,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039055140697485203,
      "loss": 4.8416,
      "mean_token_accuracy": 0.2165542021393776,
      "num_tokens": 73361062.0,
      "step": 39750
    },
    {
      "entropy": 5.41996202468872,
      "epoch": 3.339928586431422,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039052571023699516,
      "loss": 5.0124,
      "mean_token_accuracy": 0.20046370029449462,
      "num_tokens": 73370207.0,
      "step": 39755
    },
    {
      "entropy": 5.490372133255005,
      "epoch": 3.340348666246587,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039050001145265973,
      "loss": 5.1051,
      "mean_token_accuracy": 0.1945744499564171,
      "num_tokens": 73379823.0,
      "step": 39760
    },
    {
      "entropy": 5.399872589111328,
      "epoch": 3.3407687460617517,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039047431062230113,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21180406659841539,
      "num_tokens": 73388765.0,
      "step": 39765
    },
    {
      "entropy": 5.505929613113404,
      "epoch": 3.3411888258769165,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003904486077463746,
      "loss": 5.0166,
      "mean_token_accuracy": 0.20269328355789185,
      "num_tokens": 73397701.0,
      "step": 39770
    },
    {
      "entropy": 5.431462049484253,
      "epoch": 3.3416089056920817,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003904229028253354,
      "loss": 5.0626,
      "mean_token_accuracy": 0.1997684895992279,
      "num_tokens": 73406761.0,
      "step": 39775
    },
    {
      "entropy": 5.455052709579467,
      "epoch": 3.3420289855072465,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003903971958596389,
      "loss": 5.0223,
      "mean_token_accuracy": 0.20125970393419265,
      "num_tokens": 73417195.0,
      "step": 39780
    },
    {
      "entropy": 5.466881322860718,
      "epoch": 3.3424490653224113,
      "grad_norm": 1.125,
      "learning_rate": 0.0003903714868497405,
      "loss": 5.0211,
      "mean_token_accuracy": 0.20088918954133989,
      "num_tokens": 73426048.0,
      "step": 39785
    },
    {
      "entropy": 5.375912094116211,
      "epoch": 3.342869145137576,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039034577579609566,
      "loss": 4.8575,
      "mean_token_accuracy": 0.21280352920293807,
      "num_tokens": 73434055.0,
      "step": 39790
    },
    {
      "entropy": 5.442580938339233,
      "epoch": 3.343289224952741,
      "grad_norm": 1.125,
      "learning_rate": 0.00039032006269915976,
      "loss": 5.0021,
      "mean_token_accuracy": 0.2049369513988495,
      "num_tokens": 73442778.0,
      "step": 39795
    },
    {
      "entropy": 5.367901086807251,
      "epoch": 3.343709304767906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003902943475593883,
      "loss": 4.9469,
      "mean_token_accuracy": 0.2059815615415573,
      "num_tokens": 73451502.0,
      "step": 39800
    },
    {
      "entropy": 5.51497974395752,
      "epoch": 3.344129384583071,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003902686303772369,
      "loss": 5.0851,
      "mean_token_accuracy": 0.1919392466545105,
      "num_tokens": 73460097.0,
      "step": 39805
    },
    {
      "entropy": 5.522514963150025,
      "epoch": 3.3445494643982356,
      "grad_norm": 1.125,
      "learning_rate": 0.000390242911153161,
      "loss": 4.9926,
      "mean_token_accuracy": 0.20035905092954637,
      "num_tokens": 73469015.0,
      "step": 39810
    },
    {
      "entropy": 5.380199289321899,
      "epoch": 3.3449695442134004,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003902171898876162,
      "loss": 4.9003,
      "mean_token_accuracy": 0.21126618534326552,
      "num_tokens": 73477924.0,
      "step": 39815
    },
    {
      "entropy": 5.422903537750244,
      "epoch": 3.3453896240285657,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039019146658105837,
      "loss": 4.9699,
      "mean_token_accuracy": 0.20094941705465316,
      "num_tokens": 73486981.0,
      "step": 39820
    },
    {
      "entropy": 5.508761405944824,
      "epoch": 3.3458097038437304,
      "grad_norm": 1.078125,
      "learning_rate": 0.000390165741233943,
      "loss": 5.0323,
      "mean_token_accuracy": 0.19550971388816835,
      "num_tokens": 73495747.0,
      "step": 39825
    },
    {
      "entropy": 5.438371992111206,
      "epoch": 3.3462297836588952,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039014001384672577,
      "loss": 4.9788,
      "mean_token_accuracy": 0.20578560680150987,
      "num_tokens": 73504959.0,
      "step": 39830
    },
    {
      "entropy": 5.429667329788208,
      "epoch": 3.34664986347406,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003901142844198625,
      "loss": 4.9947,
      "mean_token_accuracy": 0.20224272161722184,
      "num_tokens": 73513554.0,
      "step": 39835
    },
    {
      "entropy": 5.470904588699341,
      "epoch": 3.347069943289225,
      "grad_norm": 0.9375,
      "learning_rate": 0.000390088552953809,
      "loss": 5.0137,
      "mean_token_accuracy": 0.2072012707591057,
      "num_tokens": 73523831.0,
      "step": 39840
    },
    {
      "entropy": 5.454239988327027,
      "epoch": 3.34749002310439,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039006281944902106,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20190001875162125,
      "num_tokens": 73532392.0,
      "step": 39845
    },
    {
      "entropy": 5.452204847335816,
      "epoch": 3.347910102919555,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039003708390595443,
      "loss": 5.0055,
      "mean_token_accuracy": 0.20394090712070465,
      "num_tokens": 73542240.0,
      "step": 39850
    },
    {
      "entropy": 5.557689476013183,
      "epoch": 3.3483301827347196,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039001134632506515,
      "loss": 5.0756,
      "mean_token_accuracy": 0.19815595597028732,
      "num_tokens": 73552372.0,
      "step": 39855
    },
    {
      "entropy": 5.387678956985473,
      "epoch": 3.3487502625498844,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003899856067068091,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20743410438299179,
      "num_tokens": 73561265.0,
      "step": 39860
    },
    {
      "entropy": 5.497483730316162,
      "epoch": 3.349170342365049,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038995986505164224,
      "loss": 5.0574,
      "mean_token_accuracy": 0.19908715635538102,
      "num_tokens": 73570654.0,
      "step": 39865
    },
    {
      "entropy": 5.522952222824097,
      "epoch": 3.3495904221802144,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003899341213600205,
      "loss": 5.0173,
      "mean_token_accuracy": 0.19915869385004042,
      "num_tokens": 73579621.0,
      "step": 39870
    },
    {
      "entropy": 5.51110610961914,
      "epoch": 3.350010501995379,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038990837563240003,
      "loss": 5.0595,
      "mean_token_accuracy": 0.20198958665132521,
      "num_tokens": 73589139.0,
      "step": 39875
    },
    {
      "entropy": 5.411385774612427,
      "epoch": 3.350430581810544,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038988262786923683,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20738916844129562,
      "num_tokens": 73597781.0,
      "step": 39880
    },
    {
      "entropy": 5.411518144607544,
      "epoch": 3.3508506616257088,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000389856878070987,
      "loss": 4.9509,
      "mean_token_accuracy": 0.2080303654074669,
      "num_tokens": 73605886.0,
      "step": 39885
    },
    {
      "entropy": 5.384153366088867,
      "epoch": 3.3512707414408736,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00038983112623810676,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21406611800193787,
      "num_tokens": 73615313.0,
      "step": 39890
    },
    {
      "entropy": 5.429943609237671,
      "epoch": 3.351690821256039,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003898053723710523,
      "loss": 5.0273,
      "mean_token_accuracy": 0.198246069252491,
      "num_tokens": 73626561.0,
      "step": 39895
    },
    {
      "entropy": 5.502477073669434,
      "epoch": 3.3521109010712036,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003897796164702797,
      "loss": 4.9936,
      "mean_token_accuracy": 0.20250643491744996,
      "num_tokens": 73635968.0,
      "step": 39900
    },
    {
      "entropy": 5.398700141906739,
      "epoch": 3.3525309808863684,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003897538585362454,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21909637600183487,
      "num_tokens": 73645276.0,
      "step": 39905
    },
    {
      "entropy": 5.39681224822998,
      "epoch": 3.352951060701533,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003897280985694055,
      "loss": 4.9954,
      "mean_token_accuracy": 0.19953548461198806,
      "num_tokens": 73653315.0,
      "step": 39910
    },
    {
      "entropy": 5.514429378509521,
      "epoch": 3.353371140516698,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038970233657021643,
      "loss": 5.0697,
      "mean_token_accuracy": 0.19768503457307815,
      "num_tokens": 73662994.0,
      "step": 39915
    },
    {
      "entropy": 5.501956224441528,
      "epoch": 3.353791220331863,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003896765725391346,
      "loss": 4.968,
      "mean_token_accuracy": 0.20939837247133256,
      "num_tokens": 73672109.0,
      "step": 39920
    },
    {
      "entropy": 5.427763795852661,
      "epoch": 3.354211300147028,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038965080647661624,
      "loss": 4.9838,
      "mean_token_accuracy": 0.2060981124639511,
      "num_tokens": 73682321.0,
      "step": 39925
    },
    {
      "entropy": 5.42556586265564,
      "epoch": 3.3546313799621927,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003896250383831179,
      "loss": 4.9872,
      "mean_token_accuracy": 0.20797317177057267,
      "num_tokens": 73692494.0,
      "step": 39930
    },
    {
      "entropy": 5.413104104995727,
      "epoch": 3.3550514597773575,
      "grad_norm": 1.09375,
      "learning_rate": 0.000389599268259096,
      "loss": 4.9227,
      "mean_token_accuracy": 0.21091172099113464,
      "num_tokens": 73701113.0,
      "step": 39935
    },
    {
      "entropy": 5.533779573440552,
      "epoch": 3.3554715395925228,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003895734961050071,
      "loss": 5.1317,
      "mean_token_accuracy": 0.1914818301796913,
      "num_tokens": 73710618.0,
      "step": 39940
    },
    {
      "entropy": 5.467933654785156,
      "epoch": 3.3558916194076875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003895477219213078,
      "loss": 4.9741,
      "mean_token_accuracy": 0.20694924741983414,
      "num_tokens": 73719514.0,
      "step": 39945
    },
    {
      "entropy": 5.386525583267212,
      "epoch": 3.3563116992228523,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038952194570845446,
      "loss": 4.9779,
      "mean_token_accuracy": 0.19896466434001922,
      "num_tokens": 73728144.0,
      "step": 39950
    },
    {
      "entropy": 5.419619131088257,
      "epoch": 3.356731779038017,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038949616746690387,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20337124168872833,
      "num_tokens": 73736810.0,
      "step": 39955
    },
    {
      "entropy": 5.428277826309204,
      "epoch": 3.3571518588531823,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003894703871971126,
      "loss": 4.9773,
      "mean_token_accuracy": 0.20561828166246415,
      "num_tokens": 73746108.0,
      "step": 39960
    },
    {
      "entropy": 5.544042348861694,
      "epoch": 3.357571938668347,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038944460489953744,
      "loss": 5.1118,
      "mean_token_accuracy": 0.19121332168579103,
      "num_tokens": 73755553.0,
      "step": 39965
    },
    {
      "entropy": 5.482161951065064,
      "epoch": 3.357992018483512,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038941882057463503,
      "loss": 5.0304,
      "mean_token_accuracy": 0.2053460121154785,
      "num_tokens": 73764801.0,
      "step": 39970
    },
    {
      "entropy": 5.38171648979187,
      "epoch": 3.3584120982986767,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038939303422286204,
      "loss": 4.915,
      "mean_token_accuracy": 0.20882476419210433,
      "num_tokens": 73773244.0,
      "step": 39975
    },
    {
      "entropy": 5.465098667144775,
      "epoch": 3.3588321781138415,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003893672458446754,
      "loss": 5.0125,
      "mean_token_accuracy": 0.2004648983478546,
      "num_tokens": 73783141.0,
      "step": 39980
    },
    {
      "entropy": 5.446253633499145,
      "epoch": 3.3592522579290067,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003893414554405319,
      "loss": 4.9958,
      "mean_token_accuracy": 0.2025010034441948,
      "num_tokens": 73791854.0,
      "step": 39985
    },
    {
      "entropy": 5.503292846679687,
      "epoch": 3.3596723377441715,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003893156630108884,
      "loss": 4.954,
      "mean_token_accuracy": 0.2053287923336029,
      "num_tokens": 73800558.0,
      "step": 39990
    },
    {
      "entropy": 5.488830089569092,
      "epoch": 3.3600924175593363,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003892898685562018,
      "loss": 5.0188,
      "mean_token_accuracy": 0.20546677708625793,
      "num_tokens": 73808938.0,
      "step": 39995
    },
    {
      "entropy": 5.343171215057373,
      "epoch": 3.360512497374501,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000389264072076929,
      "loss": 4.9469,
      "mean_token_accuracy": 0.20086036771535873,
      "num_tokens": 73817810.0,
      "step": 40000
    },
    {
      "entropy": 5.444939804077149,
      "epoch": 3.360932577189666,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038923827357352703,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19802677035331726,
      "num_tokens": 73827198.0,
      "step": 40005
    },
    {
      "entropy": 5.4722977638244625,
      "epoch": 3.361352657004831,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038921247304645287,
      "loss": 5.036,
      "mean_token_accuracy": 0.2017348065972328,
      "num_tokens": 73836406.0,
      "step": 40010
    },
    {
      "entropy": 5.4291260719299315,
      "epoch": 3.361772736819996,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003891866704961635,
      "loss": 5.017,
      "mean_token_accuracy": 0.19951038509607316,
      "num_tokens": 73846092.0,
      "step": 40015
    },
    {
      "entropy": 5.365188503265381,
      "epoch": 3.3621928166351607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038916086592311614,
      "loss": 4.8665,
      "mean_token_accuracy": 0.21454322338104248,
      "num_tokens": 73854447.0,
      "step": 40020
    },
    {
      "entropy": 5.48971471786499,
      "epoch": 3.3626128964503255,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003891350593277677,
      "loss": 5.0697,
      "mean_token_accuracy": 0.19642435908317565,
      "num_tokens": 73864060.0,
      "step": 40025
    },
    {
      "entropy": 5.44871940612793,
      "epoch": 3.3630329762654902,
      "grad_norm": 1.125,
      "learning_rate": 0.0003891092507105756,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2145434409379959,
      "num_tokens": 73872719.0,
      "step": 40030
    },
    {
      "entropy": 5.50097165107727,
      "epoch": 3.3634530560806555,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038908344007199683,
      "loss": 5.0455,
      "mean_token_accuracy": 0.2045575574040413,
      "num_tokens": 73881927.0,
      "step": 40035
    },
    {
      "entropy": 5.4099798679351805,
      "epoch": 3.3638731358958203,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038905762741248866,
      "loss": 4.9218,
      "mean_token_accuracy": 0.20595065653324127,
      "num_tokens": 73890814.0,
      "step": 40040
    },
    {
      "entropy": 5.460403776168823,
      "epoch": 3.364293215710985,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003890318127325083,
      "loss": 4.9501,
      "mean_token_accuracy": 0.20722517967224122,
      "num_tokens": 73899615.0,
      "step": 40045
    },
    {
      "entropy": 5.4704807758331295,
      "epoch": 3.36471329552615,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003890059960325132,
      "loss": 5.0417,
      "mean_token_accuracy": 0.19995758086442947,
      "num_tokens": 73908955.0,
      "step": 40050
    },
    {
      "entropy": 5.494711971282959,
      "epoch": 3.3651333753413146,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038898017731296046,
      "loss": 5.0409,
      "mean_token_accuracy": 0.20951238125562668,
      "num_tokens": 73917429.0,
      "step": 40055
    },
    {
      "entropy": 5.490182781219483,
      "epoch": 3.36555345515648,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038895435657430763,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20517728179693223,
      "num_tokens": 73926114.0,
      "step": 40060
    },
    {
      "entropy": 5.454232406616211,
      "epoch": 3.3659735349716446,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038892853381701205,
      "loss": 4.9948,
      "mean_token_accuracy": 0.20288797467947006,
      "num_tokens": 73935560.0,
      "step": 40065
    },
    {
      "entropy": 5.4154078483581545,
      "epoch": 3.3663936147868094,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003889027090415312,
      "loss": 4.9009,
      "mean_token_accuracy": 0.20612566471099852,
      "num_tokens": 73944563.0,
      "step": 40070
    },
    {
      "entropy": 5.393708276748657,
      "epoch": 3.366813694601974,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038887688224832243,
      "loss": 4.9588,
      "mean_token_accuracy": 0.2040835589170456,
      "num_tokens": 73953936.0,
      "step": 40075
    },
    {
      "entropy": 5.414011001586914,
      "epoch": 3.3672337744171394,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038885105343784346,
      "loss": 5.03,
      "mean_token_accuracy": 0.2012826293706894,
      "num_tokens": 73963953.0,
      "step": 40080
    },
    {
      "entropy": 5.560592365264893,
      "epoch": 3.3676538542323042,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003888252226105516,
      "loss": 5.09,
      "mean_token_accuracy": 0.2048245370388031,
      "num_tokens": 73973702.0,
      "step": 40085
    },
    {
      "entropy": 5.513804769515991,
      "epoch": 3.368073934047469,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038879938976690457,
      "loss": 5.1185,
      "mean_token_accuracy": 0.19360488206148146,
      "num_tokens": 73984051.0,
      "step": 40090
    },
    {
      "entropy": 5.422868537902832,
      "epoch": 3.368494013862634,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038877355490735994,
      "loss": 4.9838,
      "mean_token_accuracy": 0.20492325276136397,
      "num_tokens": 73993632.0,
      "step": 40095
    },
    {
      "entropy": 5.453632640838623,
      "epoch": 3.3689140936777986,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00038874771803237536,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20490719079971315,
      "num_tokens": 74003073.0,
      "step": 40100
    },
    {
      "entropy": 5.435187196731567,
      "epoch": 3.369334173492964,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038872187914240847,
      "loss": 4.9224,
      "mean_token_accuracy": 0.1964834988117218,
      "num_tokens": 74011879.0,
      "step": 40105
    },
    {
      "entropy": 5.457554578781128,
      "epoch": 3.3697542533081286,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003886960382379171,
      "loss": 4.9515,
      "mean_token_accuracy": 0.212699493765831,
      "num_tokens": 74020850.0,
      "step": 40110
    },
    {
      "entropy": 5.5195478916168215,
      "epoch": 3.3701743331232934,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038867019531935897,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19690089225769042,
      "num_tokens": 74030317.0,
      "step": 40115
    },
    {
      "entropy": 5.528742408752441,
      "epoch": 3.370594412938458,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038864435038719186,
      "loss": 5.0089,
      "mean_token_accuracy": 0.2054593026638031,
      "num_tokens": 74039675.0,
      "step": 40120
    },
    {
      "entropy": 5.488582897186279,
      "epoch": 3.3710144927536234,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038861850344187356,
      "loss": 5.0019,
      "mean_token_accuracy": 0.19993501007556916,
      "num_tokens": 74049248.0,
      "step": 40125
    },
    {
      "entropy": 5.380663967132568,
      "epoch": 3.371434572568788,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000388592654483862,
      "loss": 4.9233,
      "mean_token_accuracy": 0.21270245164632798,
      "num_tokens": 74057568.0,
      "step": 40130
    },
    {
      "entropy": 5.578536701202393,
      "epoch": 3.371854652383953,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038856680351361504,
      "loss": 5.2155,
      "mean_token_accuracy": 0.18489163517951965,
      "num_tokens": 74067855.0,
      "step": 40135
    },
    {
      "entropy": 5.4491363048553465,
      "epoch": 3.3722747321991178,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003885409505315907,
      "loss": 4.9591,
      "mean_token_accuracy": 0.20314182937145234,
      "num_tokens": 74076871.0,
      "step": 40140
    },
    {
      "entropy": 5.44821400642395,
      "epoch": 3.3726948120142826,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003885150955382469,
      "loss": 5.0465,
      "mean_token_accuracy": 0.20226654559373855,
      "num_tokens": 74085710.0,
      "step": 40145
    },
    {
      "entropy": 5.615245628356933,
      "epoch": 3.373114891829448,
      "grad_norm": 1.125,
      "learning_rate": 0.00038848923853404153,
      "loss": 5.1263,
      "mean_token_accuracy": 0.19918455481529235,
      "num_tokens": 74094768.0,
      "step": 40150
    },
    {
      "entropy": 5.438761758804321,
      "epoch": 3.3735349716446126,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003884633795194328,
      "loss": 4.938,
      "mean_token_accuracy": 0.21200217455625534,
      "num_tokens": 74104169.0,
      "step": 40155
    },
    {
      "entropy": 5.377452182769775,
      "epoch": 3.3739550514597774,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038843751849487875,
      "loss": 5.0455,
      "mean_token_accuracy": 0.20387629717588424,
      "num_tokens": 74113493.0,
      "step": 40160
    },
    {
      "entropy": 5.351119661331177,
      "epoch": 3.374375131274942,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038841165546083737,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21583041399717331,
      "num_tokens": 74121774.0,
      "step": 40165
    },
    {
      "entropy": 5.491842794418335,
      "epoch": 3.374795211090107,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038838579041776703,
      "loss": 5.0462,
      "mean_token_accuracy": 0.19633707106113435,
      "num_tokens": 74131644.0,
      "step": 40170
    },
    {
      "entropy": 5.61126971244812,
      "epoch": 3.375215290905272,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003883599233661257,
      "loss": 5.1994,
      "mean_token_accuracy": 0.19192510545253755,
      "num_tokens": 74142178.0,
      "step": 40175
    },
    {
      "entropy": 5.544053173065185,
      "epoch": 3.375635370720437,
      "grad_norm": 1.125,
      "learning_rate": 0.00038833405430637174,
      "loss": 5.0132,
      "mean_token_accuracy": 0.20145414024591446,
      "num_tokens": 74151823.0,
      "step": 40180
    },
    {
      "entropy": 5.455138301849365,
      "epoch": 3.3760554505356017,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038830818323896336,
      "loss": 4.946,
      "mean_token_accuracy": 0.20921961963176727,
      "num_tokens": 74160161.0,
      "step": 40185
    },
    {
      "entropy": 5.368802452087403,
      "epoch": 3.3764755303507665,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003882823101643589,
      "loss": 5.0174,
      "mean_token_accuracy": 0.2004290133714676,
      "num_tokens": 74168947.0,
      "step": 40190
    },
    {
      "entropy": 5.535355424880981,
      "epoch": 3.3768956101659313,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003882564350830166,
      "loss": 5.0202,
      "mean_token_accuracy": 0.197880819439888,
      "num_tokens": 74178156.0,
      "step": 40195
    },
    {
      "entropy": 5.535935258865356,
      "epoch": 3.3773156899810965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038823055799539494,
      "loss": 4.9954,
      "mean_token_accuracy": 0.20945094525814056,
      "num_tokens": 74187314.0,
      "step": 40200
    },
    {
      "entropy": 5.4654860496521,
      "epoch": 3.3777357697962613,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038820467890195214,
      "loss": 4.9762,
      "mean_token_accuracy": 0.2032724514603615,
      "num_tokens": 74196569.0,
      "step": 40205
    },
    {
      "entropy": 5.452509117126465,
      "epoch": 3.378155849611426,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003881787978031469,
      "loss": 4.953,
      "mean_token_accuracy": 0.20414326041936875,
      "num_tokens": 74205320.0,
      "step": 40210
    },
    {
      "entropy": 5.44056921005249,
      "epoch": 3.378575929426591,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038815291469943735,
      "loss": 4.9785,
      "mean_token_accuracy": 0.197767037153244,
      "num_tokens": 74214561.0,
      "step": 40215
    },
    {
      "entropy": 5.43310866355896,
      "epoch": 3.3789960092417557,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003881270295912823,
      "loss": 4.9776,
      "mean_token_accuracy": 0.204453007876873,
      "num_tokens": 74224015.0,
      "step": 40220
    },
    {
      "entropy": 5.517815113067627,
      "epoch": 3.379416089056921,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00038810114247914016,
      "loss": 5.1121,
      "mean_token_accuracy": 0.191264545917511,
      "num_tokens": 74233921.0,
      "step": 40225
    },
    {
      "entropy": 5.5204681873321535,
      "epoch": 3.3798361688720857,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003880752533634695,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2137249916791916,
      "num_tokens": 74242227.0,
      "step": 40230
    },
    {
      "entropy": 5.314469671249389,
      "epoch": 3.3802562486872505,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003880493622447289,
      "loss": 4.8925,
      "mean_token_accuracy": 0.2100583627820015,
      "num_tokens": 74251103.0,
      "step": 40235
    },
    {
      "entropy": 5.479018020629883,
      "epoch": 3.3806763285024153,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003880234691233771,
      "loss": 5.1214,
      "mean_token_accuracy": 0.19542838633060455,
      "num_tokens": 74261237.0,
      "step": 40240
    },
    {
      "entropy": 5.498105812072754,
      "epoch": 3.3810964083175805,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003879975739998728,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20549378097057341,
      "num_tokens": 74269776.0,
      "step": 40245
    },
    {
      "entropy": 5.465292263031006,
      "epoch": 3.3815164881327453,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003879716768746746,
      "loss": 4.9041,
      "mean_token_accuracy": 0.20449284017086028,
      "num_tokens": 74277989.0,
      "step": 40250
    },
    {
      "entropy": 5.446615219116211,
      "epoch": 3.38193656794791,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038794577774824135,
      "loss": 4.9669,
      "mean_token_accuracy": 0.20100870579481125,
      "num_tokens": 74287849.0,
      "step": 40255
    },
    {
      "entropy": 5.43276948928833,
      "epoch": 3.382356647763075,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038791987662103187,
      "loss": 4.9421,
      "mean_token_accuracy": 0.20415633618831636,
      "num_tokens": 74296989.0,
      "step": 40260
    },
    {
      "entropy": 5.496999835968017,
      "epoch": 3.38277672757824,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003878939734935048,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20012001544237137,
      "num_tokens": 74305592.0,
      "step": 40265
    },
    {
      "entropy": 5.454124593734742,
      "epoch": 3.383196807393405,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038786806836611917,
      "loss": 5.026,
      "mean_token_accuracy": 0.20510165840387345,
      "num_tokens": 74314738.0,
      "step": 40270
    },
    {
      "entropy": 5.5624199390411375,
      "epoch": 3.3836168872085697,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003878421612393338,
      "loss": 5.0591,
      "mean_token_accuracy": 0.1928695932030678,
      "num_tokens": 74323947.0,
      "step": 40275
    },
    {
      "entropy": 5.4305336475372314,
      "epoch": 3.3840369670237345,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038781625211360763,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20105755776166917,
      "num_tokens": 74333700.0,
      "step": 40280
    },
    {
      "entropy": 5.409491539001465,
      "epoch": 3.3844570468388993,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038779034098939975,
      "loss": 4.9771,
      "mean_token_accuracy": 0.2014092594385147,
      "num_tokens": 74342669.0,
      "step": 40285
    },
    {
      "entropy": 5.485744667053223,
      "epoch": 3.3848771266540645,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003877644278671689,
      "loss": 4.9965,
      "mean_token_accuracy": 0.20406735837459564,
      "num_tokens": 74351823.0,
      "step": 40290
    },
    {
      "entropy": 5.40206093788147,
      "epoch": 3.3852972064692293,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038773851274737445,
      "loss": 4.8883,
      "mean_token_accuracy": 0.21664421111345292,
      "num_tokens": 74360127.0,
      "step": 40295
    },
    {
      "entropy": 5.507190895080567,
      "epoch": 3.385717286284394,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038771259563047514,
      "loss": 5.0636,
      "mean_token_accuracy": 0.2057281255722046,
      "num_tokens": 74369059.0,
      "step": 40300
    },
    {
      "entropy": 5.435095930099488,
      "epoch": 3.386137366099559,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003876866765169303,
      "loss": 5.0378,
      "mean_token_accuracy": 0.19915671944618224,
      "num_tokens": 74378649.0,
      "step": 40305
    },
    {
      "entropy": 5.519116544723511,
      "epoch": 3.3865574459147236,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038766075540719904,
      "loss": 5.05,
      "mean_token_accuracy": 0.19067341834306717,
      "num_tokens": 74388067.0,
      "step": 40310
    },
    {
      "entropy": 5.425424480438233,
      "epoch": 3.386977525729889,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038763483230174053,
      "loss": 4.9573,
      "mean_token_accuracy": 0.20737352073192597,
      "num_tokens": 74396482.0,
      "step": 40315
    },
    {
      "entropy": 5.443493843078613,
      "epoch": 3.3873976055450536,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003876089072010139,
      "loss": 5.0153,
      "mean_token_accuracy": 0.19658247381448746,
      "num_tokens": 74405547.0,
      "step": 40320
    },
    {
      "entropy": 5.452578020095825,
      "epoch": 3.3878176853602184,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003875829801054785,
      "loss": 4.9515,
      "mean_token_accuracy": 0.2086341366171837,
      "num_tokens": 74414836.0,
      "step": 40325
    },
    {
      "entropy": 5.509425687789917,
      "epoch": 3.388237765175383,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038755705101559357,
      "loss": 5.0381,
      "mean_token_accuracy": 0.2051796108484268,
      "num_tokens": 74423545.0,
      "step": 40330
    },
    {
      "entropy": 5.421851396560669,
      "epoch": 3.388657844990548,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003875311199318185,
      "loss": 4.9361,
      "mean_token_accuracy": 0.20867653489112853,
      "num_tokens": 74432287.0,
      "step": 40335
    },
    {
      "entropy": 5.440361261367798,
      "epoch": 3.3890779248057132,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038750518685461247,
      "loss": 5.1149,
      "mean_token_accuracy": 0.19672640413045883,
      "num_tokens": 74442175.0,
      "step": 40340
    },
    {
      "entropy": 5.433359432220459,
      "epoch": 3.389498004620878,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000387479251784435,
      "loss": 4.9334,
      "mean_token_accuracy": 0.20822295546531677,
      "num_tokens": 74450564.0,
      "step": 40345
    },
    {
      "entropy": 5.556117105484009,
      "epoch": 3.389918084436043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038745331472174563,
      "loss": 5.2253,
      "mean_token_accuracy": 0.19139765352010726,
      "num_tokens": 74460743.0,
      "step": 40350
    },
    {
      "entropy": 5.508415174484253,
      "epoch": 3.3903381642512076,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038742737566700356,
      "loss": 5.0702,
      "mean_token_accuracy": 0.19567021876573562,
      "num_tokens": 74469915.0,
      "step": 40355
    },
    {
      "entropy": 5.529552555084228,
      "epoch": 3.3907582440663724,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038740143462066845,
      "loss": 5.0543,
      "mean_token_accuracy": 0.19802587926387788,
      "num_tokens": 74480307.0,
      "step": 40360
    },
    {
      "entropy": 5.55032377243042,
      "epoch": 3.3911783238815376,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038737549158319984,
      "loss": 5.04,
      "mean_token_accuracy": 0.1975310280919075,
      "num_tokens": 74490058.0,
      "step": 40365
    },
    {
      "entropy": 5.4853044033050535,
      "epoch": 3.3915984036967024,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038734954655505723,
      "loss": 4.9704,
      "mean_token_accuracy": 0.204733906686306,
      "num_tokens": 74499183.0,
      "step": 40370
    },
    {
      "entropy": 5.539429092407227,
      "epoch": 3.392018483511867,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038732359953670025,
      "loss": 5.0246,
      "mean_token_accuracy": 0.2065304547548294,
      "num_tokens": 74508213.0,
      "step": 40375
    },
    {
      "entropy": 5.361800241470337,
      "epoch": 3.392438563327032,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038729765052858856,
      "loss": 4.9448,
      "mean_token_accuracy": 0.20518131405115128,
      "num_tokens": 74517520.0,
      "step": 40380
    },
    {
      "entropy": 5.377232217788697,
      "epoch": 3.392858643142197,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038727169953118185,
      "loss": 4.9141,
      "mean_token_accuracy": 0.20951476991176604,
      "num_tokens": 74527236.0,
      "step": 40385
    },
    {
      "entropy": 5.467092370986938,
      "epoch": 3.393278722957362,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003872457465449398,
      "loss": 5.0511,
      "mean_token_accuracy": 0.1995369091629982,
      "num_tokens": 74535886.0,
      "step": 40390
    },
    {
      "entropy": 5.484516334533692,
      "epoch": 3.393698802772527,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038721979157032205,
      "loss": 5.0491,
      "mean_token_accuracy": 0.19565394073724746,
      "num_tokens": 74545751.0,
      "step": 40395
    },
    {
      "entropy": 5.521516466140747,
      "epoch": 3.3941188825876916,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038719383460778856,
      "loss": 5.02,
      "mean_token_accuracy": 0.20443420708179474,
      "num_tokens": 74555271.0,
      "step": 40400
    },
    {
      "entropy": 5.540292406082154,
      "epoch": 3.3945389624028564,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038716787565779904,
      "loss": 5.0495,
      "mean_token_accuracy": 0.19560931622982025,
      "num_tokens": 74564215.0,
      "step": 40405
    },
    {
      "entropy": 5.446998548507691,
      "epoch": 3.3949590422180216,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038714191472081333,
      "loss": 5.0185,
      "mean_token_accuracy": 0.2113236129283905,
      "num_tokens": 74573251.0,
      "step": 40410
    },
    {
      "entropy": 5.4307427406311035,
      "epoch": 3.3953791220331864,
      "grad_norm": 1.0,
      "learning_rate": 0.0003871159517972914,
      "loss": 5.0212,
      "mean_token_accuracy": 0.19926595240831374,
      "num_tokens": 74582476.0,
      "step": 40415
    },
    {
      "entropy": 5.492230844497681,
      "epoch": 3.395799201848351,
      "grad_norm": 1.125,
      "learning_rate": 0.0003870899868876931,
      "loss": 5.027,
      "mean_token_accuracy": 0.19902862906455993,
      "num_tokens": 74591488.0,
      "step": 40420
    },
    {
      "entropy": 5.467468690872193,
      "epoch": 3.396219281663516,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038706401999247833,
      "loss": 4.9627,
      "mean_token_accuracy": 0.20162304639816284,
      "num_tokens": 74601871.0,
      "step": 40425
    },
    {
      "entropy": 5.436094951629639,
      "epoch": 3.396639361478681,
      "grad_norm": 1.0,
      "learning_rate": 0.0003870380511121072,
      "loss": 5.0181,
      "mean_token_accuracy": 0.1927729383111,
      "num_tokens": 74611618.0,
      "step": 40430
    },
    {
      "entropy": 5.306555223464966,
      "epoch": 3.397059441293846,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003870120802470397,
      "loss": 4.8555,
      "mean_token_accuracy": 0.21009015589952468,
      "num_tokens": 74621000.0,
      "step": 40435
    },
    {
      "entropy": 5.579749584197998,
      "epoch": 3.3974795211090107,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038698610739773586,
      "loss": 5.0953,
      "mean_token_accuracy": 0.19614443182945251,
      "num_tokens": 74629767.0,
      "step": 40440
    },
    {
      "entropy": 5.394829416275025,
      "epoch": 3.3978996009241755,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038696013256465586,
      "loss": 4.8863,
      "mean_token_accuracy": 0.215895876288414,
      "num_tokens": 74638415.0,
      "step": 40445
    },
    {
      "entropy": 5.478527212142945,
      "epoch": 3.3983196807393403,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003869341557482597,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1934507668018341,
      "num_tokens": 74647262.0,
      "step": 40450
    },
    {
      "entropy": 5.479902458190918,
      "epoch": 3.3987397605545056,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003869081769490076,
      "loss": 5.0185,
      "mean_token_accuracy": 0.20063655972480773,
      "num_tokens": 74656317.0,
      "step": 40455
    },
    {
      "entropy": 5.478737258911133,
      "epoch": 3.3991598403696703,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003868821961673598,
      "loss": 4.9611,
      "mean_token_accuracy": 0.20383478999137877,
      "num_tokens": 74665209.0,
      "step": 40460
    },
    {
      "entropy": 5.435899114608764,
      "epoch": 3.399579920184835,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038685621340377646,
      "loss": 5.0047,
      "mean_token_accuracy": 0.2037412241101265,
      "num_tokens": 74675175.0,
      "step": 40465
    },
    {
      "entropy": 5.411191987991333,
      "epoch": 3.4,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038683022865871794,
      "loss": 5.0218,
      "mean_token_accuracy": 0.2040274515748024,
      "num_tokens": 74684207.0,
      "step": 40470
    },
    {
      "entropy": 5.522229146957398,
      "epoch": 3.4004200798151647,
      "grad_norm": 1.125,
      "learning_rate": 0.00038680424193264447,
      "loss": 5.095,
      "mean_token_accuracy": 0.1993575006723404,
      "num_tokens": 74692721.0,
      "step": 40475
    },
    {
      "entropy": 5.397816371917725,
      "epoch": 3.40084015963033,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038677825322601644,
      "loss": 4.918,
      "mean_token_accuracy": 0.21406650245189668,
      "num_tokens": 74701515.0,
      "step": 40480
    },
    {
      "entropy": 5.4765058040618895,
      "epoch": 3.4012602394454947,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038675226253929426,
      "loss": 5.0083,
      "mean_token_accuracy": 0.20230465978384018,
      "num_tokens": 74710585.0,
      "step": 40485
    },
    {
      "entropy": 5.491699695587158,
      "epoch": 3.4016803192606595,
      "grad_norm": 1.125,
      "learning_rate": 0.0003867262698729382,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20742523968219756,
      "num_tokens": 74719151.0,
      "step": 40490
    },
    {
      "entropy": 5.472250843048096,
      "epoch": 3.4021003990758243,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003867002752274088,
      "loss": 4.9918,
      "mean_token_accuracy": 0.19909026026725768,
      "num_tokens": 74729182.0,
      "step": 40495
    },
    {
      "entropy": 5.542170000076294,
      "epoch": 3.402520478890989,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003866742786031666,
      "loss": 5.038,
      "mean_token_accuracy": 0.2033754050731659,
      "num_tokens": 74738434.0,
      "step": 40500
    },
    {
      "entropy": 5.4281126976013185,
      "epoch": 3.4029405587061543,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003866482800006721,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20513907223939895,
      "num_tokens": 74748400.0,
      "step": 40505
    },
    {
      "entropy": 5.416245222091675,
      "epoch": 3.403360638521319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038662227942038574,
      "loss": 5.0315,
      "mean_token_accuracy": 0.20559964329004288,
      "num_tokens": 74757901.0,
      "step": 40510
    },
    {
      "entropy": 5.472030448913574,
      "epoch": 3.403780718336484,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038659627686276824,
      "loss": 5.0126,
      "mean_token_accuracy": 0.20252881348133087,
      "num_tokens": 74766042.0,
      "step": 40515
    },
    {
      "entropy": 5.5173734664917,
      "epoch": 3.4042007981516487,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038657027232828,
      "loss": 5.0636,
      "mean_token_accuracy": 0.19326265007257462,
      "num_tokens": 74775056.0,
      "step": 40520
    },
    {
      "entropy": 5.494423580169678,
      "epoch": 3.4046208779668135,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038654426581738205,
      "loss": 5.0543,
      "mean_token_accuracy": 0.19466761201620103,
      "num_tokens": 74784829.0,
      "step": 40525
    },
    {
      "entropy": 5.417185544967651,
      "epoch": 3.4050409577819787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038651825733053473,
      "loss": 4.9227,
      "mean_token_accuracy": 0.20635501444339752,
      "num_tokens": 74794497.0,
      "step": 40530
    },
    {
      "entropy": 5.504184150695801,
      "epoch": 3.4054610375971435,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038649224686819895,
      "loss": 5.065,
      "mean_token_accuracy": 0.20072720795869828,
      "num_tokens": 74803630.0,
      "step": 40535
    },
    {
      "entropy": 5.495270442962647,
      "epoch": 3.4058811174123083,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003864662344308354,
      "loss": 5.0154,
      "mean_token_accuracy": 0.20455118715763093,
      "num_tokens": 74812616.0,
      "step": 40540
    },
    {
      "entropy": 5.421155261993408,
      "epoch": 3.406301197227473,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038644022001890494,
      "loss": 4.8745,
      "mean_token_accuracy": 0.21108186095952988,
      "num_tokens": 74820455.0,
      "step": 40545
    },
    {
      "entropy": 5.505782127380371,
      "epoch": 3.4067212770426383,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038641420363286834,
      "loss": 5.0866,
      "mean_token_accuracy": 0.19642014801502228,
      "num_tokens": 74830264.0,
      "step": 40550
    },
    {
      "entropy": 5.485417604446411,
      "epoch": 3.407141356857803,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038638818527318654,
      "loss": 5.0428,
      "mean_token_accuracy": 0.19686658084392547,
      "num_tokens": 74840859.0,
      "step": 40555
    },
    {
      "entropy": 5.3772368907928465,
      "epoch": 3.407561436672968,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038636216494032026,
      "loss": 4.8822,
      "mean_token_accuracy": 0.21115367859601974,
      "num_tokens": 74848979.0,
      "step": 40560
    },
    {
      "entropy": 5.437415266036988,
      "epoch": 3.4079815164881326,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003863361426347307,
      "loss": 5.0337,
      "mean_token_accuracy": 0.1947709485888481,
      "num_tokens": 74857874.0,
      "step": 40565
    },
    {
      "entropy": 5.514195966720581,
      "epoch": 3.408401596303298,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003863101183568786,
      "loss": 4.9666,
      "mean_token_accuracy": 0.20696549862623215,
      "num_tokens": 74866692.0,
      "step": 40570
    },
    {
      "entropy": 5.547784662246704,
      "epoch": 3.4088216761184627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003862840921072252,
      "loss": 5.0919,
      "mean_token_accuracy": 0.19766554683446885,
      "num_tokens": 74876562.0,
      "step": 40575
    },
    {
      "entropy": 5.474359130859375,
      "epoch": 3.4092417559336274,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003862580638862314,
      "loss": 5.0189,
      "mean_token_accuracy": 0.20415627658367158,
      "num_tokens": 74886253.0,
      "step": 40580
    },
    {
      "entropy": 5.478776645660401,
      "epoch": 3.4096618357487922,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003862320336943582,
      "loss": 4.9835,
      "mean_token_accuracy": 0.20162849724292756,
      "num_tokens": 74895078.0,
      "step": 40585
    },
    {
      "entropy": 5.431770372390747,
      "epoch": 3.410081915563957,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003862060015320668,
      "loss": 4.9943,
      "mean_token_accuracy": 0.20644888430833816,
      "num_tokens": 74904492.0,
      "step": 40590
    },
    {
      "entropy": 5.434218263626098,
      "epoch": 3.4105019953791222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003861799673998184,
      "loss": 4.9727,
      "mean_token_accuracy": 0.20646823793649674,
      "num_tokens": 74912976.0,
      "step": 40595
    },
    {
      "entropy": 5.496877574920655,
      "epoch": 3.410922075194287,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003861539312980741,
      "loss": 5.094,
      "mean_token_accuracy": 0.20286500602960586,
      "num_tokens": 74921932.0,
      "step": 40600
    },
    {
      "entropy": 5.47724609375,
      "epoch": 3.411342155009452,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038612789322729513,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20088566541671754,
      "num_tokens": 74930955.0,
      "step": 40605
    },
    {
      "entropy": 5.4441032886505125,
      "epoch": 3.4117622348246166,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038610185318794284,
      "loss": 4.9527,
      "mean_token_accuracy": 0.20379171818494796,
      "num_tokens": 74940590.0,
      "step": 40610
    },
    {
      "entropy": 5.332818508148193,
      "epoch": 3.4121823146397814,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038607581118047836,
      "loss": 4.8463,
      "mean_token_accuracy": 0.2176797091960907,
      "num_tokens": 74948891.0,
      "step": 40615
    },
    {
      "entropy": 5.424838018417359,
      "epoch": 3.4126023944549466,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038604976720536306,
      "loss": 5.0076,
      "mean_token_accuracy": 0.20408626645803452,
      "num_tokens": 74957596.0,
      "step": 40620
    },
    {
      "entropy": 5.455378818511963,
      "epoch": 3.4130224742701114,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038602372126305827,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20081445425748826,
      "num_tokens": 74966038.0,
      "step": 40625
    },
    {
      "entropy": 5.513239908218384,
      "epoch": 3.413442554085276,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038599767335402554,
      "loss": 5.0286,
      "mean_token_accuracy": 0.201531258225441,
      "num_tokens": 74975712.0,
      "step": 40630
    },
    {
      "entropy": 5.475709676742554,
      "epoch": 3.413862633900441,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003859716234787261,
      "loss": 5.01,
      "mean_token_accuracy": 0.20347533673048018,
      "num_tokens": 74985067.0,
      "step": 40635
    },
    {
      "entropy": 5.4363292217254635,
      "epoch": 3.4142827137156058,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003859455716376216,
      "loss": 5.0166,
      "mean_token_accuracy": 0.204985311627388,
      "num_tokens": 74994576.0,
      "step": 40640
    },
    {
      "entropy": 5.509305667877197,
      "epoch": 3.414702793530771,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038591951783117337,
      "loss": 5.0817,
      "mean_token_accuracy": 0.20557031780481339,
      "num_tokens": 75003553.0,
      "step": 40645
    },
    {
      "entropy": 5.4629151821136475,
      "epoch": 3.415122873345936,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003858934620598429,
      "loss": 5.0053,
      "mean_token_accuracy": 0.20083437412977218,
      "num_tokens": 75014293.0,
      "step": 40650
    },
    {
      "entropy": 5.526914978027344,
      "epoch": 3.4155429531611006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038586740432409197,
      "loss": 5.0387,
      "mean_token_accuracy": 0.20552701652050018,
      "num_tokens": 75024229.0,
      "step": 40655
    },
    {
      "entropy": 5.511394596099853,
      "epoch": 3.4159630329762654,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038584134462438203,
      "loss": 5.0046,
      "mean_token_accuracy": 0.1998521015048027,
      "num_tokens": 75033278.0,
      "step": 40660
    },
    {
      "entropy": 5.471908473968506,
      "epoch": 3.41638311279143,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038581528296117475,
      "loss": 5.0509,
      "mean_token_accuracy": 0.19808753579854965,
      "num_tokens": 75041656.0,
      "step": 40665
    },
    {
      "entropy": 5.456634187698365,
      "epoch": 3.4168031926065954,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038578921933493177,
      "loss": 5.0235,
      "mean_token_accuracy": 0.1995171919465065,
      "num_tokens": 75050512.0,
      "step": 40670
    },
    {
      "entropy": 5.503889703750611,
      "epoch": 3.41722327242176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038576315374611486,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20569607615470886,
      "num_tokens": 75059617.0,
      "step": 40675
    },
    {
      "entropy": 5.443442535400391,
      "epoch": 3.417643352236925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003857370861951857,
      "loss": 5.0271,
      "mean_token_accuracy": 0.20374262779951097,
      "num_tokens": 75068484.0,
      "step": 40680
    },
    {
      "entropy": 5.430636739730835,
      "epoch": 3.4180634320520897,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000385711016682606,
      "loss": 5.0308,
      "mean_token_accuracy": 0.1981089547276497,
      "num_tokens": 75078348.0,
      "step": 40685
    },
    {
      "entropy": 5.442523193359375,
      "epoch": 3.418483511867255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038568494520883776,
      "loss": 4.9373,
      "mean_token_accuracy": 0.20581999868154527,
      "num_tokens": 75087545.0,
      "step": 40690
    },
    {
      "entropy": 5.477462100982666,
      "epoch": 3.4189035916824198,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003856588717743426,
      "loss": 5.0005,
      "mean_token_accuracy": 0.20889517813920974,
      "num_tokens": 75096879.0,
      "step": 40695
    },
    {
      "entropy": 5.458183622360229,
      "epoch": 3.4193236714975845,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038563279637958257,
      "loss": 5.0108,
      "mean_token_accuracy": 0.19879422038793565,
      "num_tokens": 75105432.0,
      "step": 40700
    },
    {
      "entropy": 5.508874702453613,
      "epoch": 3.4197437513127493,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003856067190250195,
      "loss": 4.9652,
      "mean_token_accuracy": 0.204096482694149,
      "num_tokens": 75114344.0,
      "step": 40705
    },
    {
      "entropy": 5.519516229629517,
      "epoch": 3.420163831127914,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003855806397111154,
      "loss": 5.0059,
      "mean_token_accuracy": 0.20041621774435042,
      "num_tokens": 75123133.0,
      "step": 40710
    },
    {
      "entropy": 5.430754852294922,
      "epoch": 3.4205839109430793,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003855545584383322,
      "loss": 5.0022,
      "mean_token_accuracy": 0.20122570544481277,
      "num_tokens": 75131891.0,
      "step": 40715
    },
    {
      "entropy": 5.484272193908692,
      "epoch": 3.421003990758244,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003855284752071319,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20136144757270813,
      "num_tokens": 75140984.0,
      "step": 40720
    },
    {
      "entropy": 5.4948381900787355,
      "epoch": 3.421424070573409,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003855023900179766,
      "loss": 5.0551,
      "mean_token_accuracy": 0.20161013603210448,
      "num_tokens": 75150234.0,
      "step": 40725
    },
    {
      "entropy": 5.532878017425537,
      "epoch": 3.4218441503885737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003854763028713283,
      "loss": 5.0352,
      "mean_token_accuracy": 0.20172789990901946,
      "num_tokens": 75159821.0,
      "step": 40730
    },
    {
      "entropy": 5.520913457870483,
      "epoch": 3.422264230203739,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038545021376764927,
      "loss": 4.9884,
      "mean_token_accuracy": 0.20758794099092484,
      "num_tokens": 75168998.0,
      "step": 40735
    },
    {
      "entropy": 5.551850461959839,
      "epoch": 3.4226843100189037,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003854241227074015,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19725649058818817,
      "num_tokens": 75178661.0,
      "step": 40740
    },
    {
      "entropy": 5.487232398986817,
      "epoch": 3.4231043898340685,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003853980296910474,
      "loss": 5.0306,
      "mean_token_accuracy": 0.20963283479213715,
      "num_tokens": 75188226.0,
      "step": 40745
    },
    {
      "entropy": 5.522430658340454,
      "epoch": 3.4235244696492333,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003853719347190489,
      "loss": 5.0389,
      "mean_token_accuracy": 0.19534048736095427,
      "num_tokens": 75198500.0,
      "step": 40750
    },
    {
      "entropy": 5.455517339706421,
      "epoch": 3.423944549464398,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038534583779186837,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20306572169065476,
      "num_tokens": 75208190.0,
      "step": 40755
    },
    {
      "entropy": 5.419887113571167,
      "epoch": 3.4243646292795633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003853197389099682,
      "loss": 4.9749,
      "mean_token_accuracy": 0.19491011798381805,
      "num_tokens": 75218341.0,
      "step": 40760
    },
    {
      "entropy": 5.438328266143799,
      "epoch": 3.424784709094728,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038529363807381073,
      "loss": 5.0443,
      "mean_token_accuracy": 0.20393554866313934,
      "num_tokens": 75227327.0,
      "step": 40765
    },
    {
      "entropy": 5.554452848434448,
      "epoch": 3.425204788909893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003852675352838582,
      "loss": 5.1008,
      "mean_token_accuracy": 0.18999476432800294,
      "num_tokens": 75236217.0,
      "step": 40770
    },
    {
      "entropy": 5.482580518722534,
      "epoch": 3.4256248687250577,
      "grad_norm": 1.125,
      "learning_rate": 0.0003852414305405731,
      "loss": 4.9782,
      "mean_token_accuracy": 0.204641754925251,
      "num_tokens": 75244899.0,
      "step": 40775
    },
    {
      "entropy": 5.429069662094117,
      "epoch": 3.4260449485402225,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038521532384441776,
      "loss": 4.9696,
      "mean_token_accuracy": 0.20816980302333832,
      "num_tokens": 75254020.0,
      "step": 40780
    },
    {
      "entropy": 5.519163608551025,
      "epoch": 3.4264650283553877,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003851892151958548,
      "loss": 5.0548,
      "mean_token_accuracy": 0.19956021904945373,
      "num_tokens": 75263135.0,
      "step": 40785
    },
    {
      "entropy": 5.420156955718994,
      "epoch": 3.4268851081705525,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003851631045953466,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20346290469169617,
      "num_tokens": 75271488.0,
      "step": 40790
    },
    {
      "entropy": 5.481930828094482,
      "epoch": 3.4273051879857173,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003851369920433557,
      "loss": 5.0561,
      "mean_token_accuracy": 0.19735123962163925,
      "num_tokens": 75279702.0,
      "step": 40795
    },
    {
      "entropy": 5.471161413192749,
      "epoch": 3.427725267800882,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003851108775403446,
      "loss": 5.058,
      "mean_token_accuracy": 0.20395553410053252,
      "num_tokens": 75289891.0,
      "step": 40800
    },
    {
      "entropy": 5.444450855255127,
      "epoch": 3.428145347616047,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003850847610867761,
      "loss": 4.9879,
      "mean_token_accuracy": 0.20115752071142196,
      "num_tokens": 75298620.0,
      "step": 40805
    },
    {
      "entropy": 5.455213022232056,
      "epoch": 3.428565427431212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003850586426831128,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21131363958120347,
      "num_tokens": 75307554.0,
      "step": 40810
    },
    {
      "entropy": 5.4988593578338625,
      "epoch": 3.428985507246377,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003850325223298173,
      "loss": 5.0,
      "mean_token_accuracy": 0.19837337285280227,
      "num_tokens": 75316507.0,
      "step": 40815
    },
    {
      "entropy": 5.504372596740723,
      "epoch": 3.4294055870615416,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038500640002735225,
      "loss": 5.0057,
      "mean_token_accuracy": 0.20341958701610566,
      "num_tokens": 75326831.0,
      "step": 40820
    },
    {
      "entropy": 5.503168201446533,
      "epoch": 3.4298256668767064,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003849802757761805,
      "loss": 5.0386,
      "mean_token_accuracy": 0.202643021941185,
      "num_tokens": 75335292.0,
      "step": 40825
    },
    {
      "entropy": 5.401393222808838,
      "epoch": 3.430245746691871,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003849541495767648,
      "loss": 4.9762,
      "mean_token_accuracy": 0.1975756123661995,
      "num_tokens": 75343858.0,
      "step": 40830
    },
    {
      "entropy": 5.382503318786621,
      "epoch": 3.4306658265070364,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038492802142956796,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21392674446105958,
      "num_tokens": 75353065.0,
      "step": 40835
    },
    {
      "entropy": 5.491394710540772,
      "epoch": 3.4310859063222012,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003849018913350528,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20546640306711197,
      "num_tokens": 75361224.0,
      "step": 40840
    },
    {
      "entropy": 5.453875350952148,
      "epoch": 3.431505986137366,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003848757592936823,
      "loss": 4.9932,
      "mean_token_accuracy": 0.20678966641426086,
      "num_tokens": 75370703.0,
      "step": 40845
    },
    {
      "entropy": 5.506053924560547,
      "epoch": 3.431926065952531,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038484962530591923,
      "loss": 5.1037,
      "mean_token_accuracy": 0.19531260579824447,
      "num_tokens": 75380893.0,
      "step": 40850
    },
    {
      "entropy": 5.44675760269165,
      "epoch": 3.432346145767696,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003848234893722267,
      "loss": 4.9823,
      "mean_token_accuracy": 0.20177959650754929,
      "num_tokens": 75390258.0,
      "step": 40855
    },
    {
      "entropy": 5.494674015045166,
      "epoch": 3.432766225582861,
      "grad_norm": 1.125,
      "learning_rate": 0.0003847973514930675,
      "loss": 5.0998,
      "mean_token_accuracy": 0.19596589654684066,
      "num_tokens": 75398988.0,
      "step": 40860
    },
    {
      "entropy": 5.4555439949035645,
      "epoch": 3.4331863053980256,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00038477121166890485,
      "loss": 4.9349,
      "mean_token_accuracy": 0.21815367937088012,
      "num_tokens": 75407961.0,
      "step": 40865
    },
    {
      "entropy": 5.529971504211426,
      "epoch": 3.4336063852131904,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003847450699002016,
      "loss": 5.0866,
      "mean_token_accuracy": 0.20491639971733094,
      "num_tokens": 75417626.0,
      "step": 40870
    },
    {
      "entropy": 5.474290132522583,
      "epoch": 3.4340264650283556,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038471892618742107,
      "loss": 4.988,
      "mean_token_accuracy": 0.20280582010746,
      "num_tokens": 75427310.0,
      "step": 40875
    },
    {
      "entropy": 5.4606681823730465,
      "epoch": 3.4344465448435204,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003846927805310261,
      "loss": 5.0335,
      "mean_token_accuracy": 0.19954955875873565,
      "num_tokens": 75436441.0,
      "step": 40880
    },
    {
      "entropy": 5.361448001861572,
      "epoch": 3.434866624658685,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038466663293148016,
      "loss": 4.8687,
      "mean_token_accuracy": 0.21194545477628707,
      "num_tokens": 75445458.0,
      "step": 40885
    },
    {
      "entropy": 5.48385968208313,
      "epoch": 3.43528670447385,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038464048338924617,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20452190935611725,
      "num_tokens": 75453512.0,
      "step": 40890
    },
    {
      "entropy": 5.473313283920288,
      "epoch": 3.4357067842890148,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038461433190478754,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20254811346530915,
      "num_tokens": 75462441.0,
      "step": 40895
    },
    {
      "entropy": 5.498736476898193,
      "epoch": 3.43612686410418,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003845881784785674,
      "loss": 5.1194,
      "mean_token_accuracy": 0.19093137681484224,
      "num_tokens": 75471779.0,
      "step": 40900
    },
    {
      "entropy": 5.539383792877198,
      "epoch": 3.436546943919345,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038456202311104923,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20367665737867355,
      "num_tokens": 75481295.0,
      "step": 40905
    },
    {
      "entropy": 5.457128524780273,
      "epoch": 3.4369670237345096,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038453586580269617,
      "loss": 4.9844,
      "mean_token_accuracy": 0.20950897336006163,
      "num_tokens": 75490481.0,
      "step": 40910
    },
    {
      "entropy": 5.462775373458863,
      "epoch": 3.4373871035496744,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003845097065539716,
      "loss": 5.0475,
      "mean_token_accuracy": 0.2031432643532753,
      "num_tokens": 75499641.0,
      "step": 40915
    },
    {
      "entropy": 5.478801727294922,
      "epoch": 3.437807183364839,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038448354536533907,
      "loss": 5.0368,
      "mean_token_accuracy": 0.1972571685910225,
      "num_tokens": 75509765.0,
      "step": 40920
    },
    {
      "entropy": 5.3933806896209715,
      "epoch": 3.4382272631800044,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038445738223726186,
      "loss": 4.9697,
      "mean_token_accuracy": 0.20693077147006989,
      "num_tokens": 75518557.0,
      "step": 40925
    },
    {
      "entropy": 5.559487295150757,
      "epoch": 3.438647342995169,
      "grad_norm": 1.125,
      "learning_rate": 0.00038443121717020344,
      "loss": 5.1354,
      "mean_token_accuracy": 0.19531406760215758,
      "num_tokens": 75529176.0,
      "step": 40930
    },
    {
      "entropy": 5.5553652286529545,
      "epoch": 3.439067422810334,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003844050501646274,
      "loss": 5.0594,
      "mean_token_accuracy": 0.19539279490709305,
      "num_tokens": 75539783.0,
      "step": 40935
    },
    {
      "entropy": 5.495352172851563,
      "epoch": 3.4394875026254987,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003843788812209972,
      "loss": 4.9412,
      "mean_token_accuracy": 0.20469011813402177,
      "num_tokens": 75549593.0,
      "step": 40940
    },
    {
      "entropy": 5.50433201789856,
      "epoch": 3.4399075824406635,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038435271033977655,
      "loss": 5.0944,
      "mean_token_accuracy": 0.19288884699344636,
      "num_tokens": 75558975.0,
      "step": 40945
    },
    {
      "entropy": 5.51657772064209,
      "epoch": 3.4403276622558288,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003843265375214289,
      "loss": 5.0643,
      "mean_token_accuracy": 0.19603597074747087,
      "num_tokens": 75568554.0,
      "step": 40950
    },
    {
      "entropy": 5.485136795043945,
      "epoch": 3.4407477420709935,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038430036276641783,
      "loss": 4.967,
      "mean_token_accuracy": 0.20668002516031264,
      "num_tokens": 75578048.0,
      "step": 40955
    },
    {
      "entropy": 5.484333515167236,
      "epoch": 3.4411678218861583,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003842741860752072,
      "loss": 5.0758,
      "mean_token_accuracy": 0.2013871878385544,
      "num_tokens": 75586200.0,
      "step": 40960
    },
    {
      "entropy": 5.406482362747193,
      "epoch": 3.441587901701323,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003842480074482606,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20459170192480086,
      "num_tokens": 75596437.0,
      "step": 40965
    },
    {
      "entropy": 5.522357606887818,
      "epoch": 3.442007981516488,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003842218268860418,
      "loss": 5.1149,
      "mean_token_accuracy": 0.19689613133668898,
      "num_tokens": 75605893.0,
      "step": 40970
    },
    {
      "entropy": 5.469048261642456,
      "epoch": 3.442428061331653,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038419564438901463,
      "loss": 4.9375,
      "mean_token_accuracy": 0.203371824324131,
      "num_tokens": 75614837.0,
      "step": 40975
    },
    {
      "entropy": 5.411543703079223,
      "epoch": 3.442848141146818,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038416945995764273,
      "loss": 4.9485,
      "mean_token_accuracy": 0.20960523337125778,
      "num_tokens": 75623854.0,
      "step": 40980
    },
    {
      "entropy": 5.416290760040283,
      "epoch": 3.4432682209619827,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003841432735923902,
      "loss": 4.958,
      "mean_token_accuracy": 0.20379687398672103,
      "num_tokens": 75632988.0,
      "step": 40985
    },
    {
      "entropy": 5.427675437927246,
      "epoch": 3.4436883007771475,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003841170852937207,
      "loss": 4.9656,
      "mean_token_accuracy": 0.2125040903687477,
      "num_tokens": 75641342.0,
      "step": 40990
    },
    {
      "entropy": 5.578057909011841,
      "epoch": 3.4441083805923127,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003840908950620982,
      "loss": 5.0695,
      "mean_token_accuracy": 0.19317152202129365,
      "num_tokens": 75650529.0,
      "step": 40995
    },
    {
      "entropy": 5.423354530334473,
      "epoch": 3.4445284604074775,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003840647028979867,
      "loss": 4.9912,
      "mean_token_accuracy": 0.2039853036403656,
      "num_tokens": 75658669.0,
      "step": 41000
    },
    {
      "entropy": 5.428369617462158,
      "epoch": 3.4449485402226423,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003840385088018502,
      "loss": 4.9945,
      "mean_token_accuracy": 0.1974459782242775,
      "num_tokens": 75667282.0,
      "step": 41005
    },
    {
      "entropy": 5.4238403797149655,
      "epoch": 3.445368620037807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003840123127741526,
      "loss": 4.9829,
      "mean_token_accuracy": 0.20298832058906555,
      "num_tokens": 75677501.0,
      "step": 41010
    },
    {
      "entropy": 5.507348489761353,
      "epoch": 3.4457886998529723,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038398611481535813,
      "loss": 5.0374,
      "mean_token_accuracy": 0.20321206897497177,
      "num_tokens": 75686813.0,
      "step": 41015
    },
    {
      "entropy": 5.454885053634643,
      "epoch": 3.446208779668137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003839599149259306,
      "loss": 4.9285,
      "mean_token_accuracy": 0.2083237037062645,
      "num_tokens": 75696832.0,
      "step": 41020
    },
    {
      "entropy": 5.468069076538086,
      "epoch": 3.446628859483302,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038393371310633433,
      "loss": 4.9775,
      "mean_token_accuracy": 0.21064782738685608,
      "num_tokens": 75705187.0,
      "step": 41025
    },
    {
      "entropy": 5.39578948020935,
      "epoch": 3.4470489392984667,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003839075093570334,
      "loss": 4.9008,
      "mean_token_accuracy": 0.21310412734746934,
      "num_tokens": 75714647.0,
      "step": 41030
    },
    {
      "entropy": 5.468989324569702,
      "epoch": 3.4474690191136315,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038388130367849205,
      "loss": 4.9994,
      "mean_token_accuracy": 0.20659517049789428,
      "num_tokens": 75723312.0,
      "step": 41035
    },
    {
      "entropy": 5.359749412536621,
      "epoch": 3.4478890989287967,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003838550960711745,
      "loss": 4.9214,
      "mean_token_accuracy": 0.20414272099733352,
      "num_tokens": 75732562.0,
      "step": 41040
    },
    {
      "entropy": 5.401244211196899,
      "epoch": 3.4483091787439615,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003838288865355449,
      "loss": 5.0317,
      "mean_token_accuracy": 0.20013727992773056,
      "num_tokens": 75741381.0,
      "step": 41045
    },
    {
      "entropy": 5.534995889663696,
      "epoch": 3.4487292585591263,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003838026750720677,
      "loss": 5.0511,
      "mean_token_accuracy": 0.20616113692522048,
      "num_tokens": 75750088.0,
      "step": 41050
    },
    {
      "entropy": 5.4376506328582765,
      "epoch": 3.449149338374291,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038377646168120716,
      "loss": 4.9523,
      "mean_token_accuracy": 0.20904172509908675,
      "num_tokens": 75758854.0,
      "step": 41055
    },
    {
      "entropy": 5.416173028945923,
      "epoch": 3.449569418189456,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003837502463634275,
      "loss": 4.9332,
      "mean_token_accuracy": 0.20767395794391633,
      "num_tokens": 75767780.0,
      "step": 41060
    },
    {
      "entropy": 5.488742351531982,
      "epoch": 3.449989498004621,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003837240291191933,
      "loss": 5.0282,
      "mean_token_accuracy": 0.20316307991743088,
      "num_tokens": 75776466.0,
      "step": 41065
    },
    {
      "entropy": 5.4711535453796385,
      "epoch": 3.450409577819786,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038369780994896885,
      "loss": 5.0525,
      "mean_token_accuracy": 0.19806579649448394,
      "num_tokens": 75785320.0,
      "step": 41070
    },
    {
      "entropy": 5.476105785369873,
      "epoch": 3.4508296576349506,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003836715888532187,
      "loss": 5.0922,
      "mean_token_accuracy": 0.19805455207824707,
      "num_tokens": 75794234.0,
      "step": 41075
    },
    {
      "entropy": 5.543690299987793,
      "epoch": 3.4512497374501154,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003836453658324074,
      "loss": 5.0011,
      "mean_token_accuracy": 0.2021676555275917,
      "num_tokens": 75803360.0,
      "step": 41080
    },
    {
      "entropy": 5.493349981307984,
      "epoch": 3.45166981726528,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003836191408869993,
      "loss": 5.0395,
      "mean_token_accuracy": 0.20089788734912872,
      "num_tokens": 75812803.0,
      "step": 41085
    },
    {
      "entropy": 5.424468374252319,
      "epoch": 3.4520898970804454,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000383592914017459,
      "loss": 4.9795,
      "mean_token_accuracy": 0.19792817831039428,
      "num_tokens": 75821808.0,
      "step": 41090
    },
    {
      "entropy": 5.5005804061889645,
      "epoch": 3.4525099768956102,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038356668522425126,
      "loss": 5.0221,
      "mean_token_accuracy": 0.20542535930871964,
      "num_tokens": 75831241.0,
      "step": 41095
    },
    {
      "entropy": 5.493171501159668,
      "epoch": 3.452930056710775,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038354045450784047,
      "loss": 4.9927,
      "mean_token_accuracy": 0.2027008965611458,
      "num_tokens": 75840641.0,
      "step": 41100
    },
    {
      "entropy": 5.467362737655639,
      "epoch": 3.45335013652594,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003835142218686915,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2138818770647049,
      "num_tokens": 75850598.0,
      "step": 41105
    },
    {
      "entropy": 5.546457099914551,
      "epoch": 3.4537702163411046,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038348798730726894,
      "loss": 5.002,
      "mean_token_accuracy": 0.20313488990068435,
      "num_tokens": 75859090.0,
      "step": 41110
    },
    {
      "entropy": 5.323027753829956,
      "epoch": 3.45419029615627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003834617508240376,
      "loss": 4.8877,
      "mean_token_accuracy": 0.21294118762016295,
      "num_tokens": 75868241.0,
      "step": 41115
    },
    {
      "entropy": 5.515311431884766,
      "epoch": 3.4546103759714346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003834355124194621,
      "loss": 5.0206,
      "mean_token_accuracy": 0.19905922412872315,
      "num_tokens": 75877440.0,
      "step": 41120
    },
    {
      "entropy": 5.501564073562622,
      "epoch": 3.4550304557865994,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003834092720940074,
      "loss": 5.0698,
      "mean_token_accuracy": 0.19058767408132554,
      "num_tokens": 75886847.0,
      "step": 41125
    },
    {
      "entropy": 5.461261463165283,
      "epoch": 3.455450535601764,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003833830298481383,
      "loss": 4.949,
      "mean_token_accuracy": 0.20387173891067506,
      "num_tokens": 75895510.0,
      "step": 41130
    },
    {
      "entropy": 5.471967792510986,
      "epoch": 3.455870615416929,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003833567856823195,
      "loss": 4.9888,
      "mean_token_accuracy": 0.2012672394514084,
      "num_tokens": 75903942.0,
      "step": 41135
    },
    {
      "entropy": 5.559611463546753,
      "epoch": 3.456290695232094,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038333053959701613,
      "loss": 5.1451,
      "mean_token_accuracy": 0.19757984429597855,
      "num_tokens": 75913758.0,
      "step": 41140
    },
    {
      "entropy": 5.437851619720459,
      "epoch": 3.456710775047259,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003833042915926931,
      "loss": 4.8674,
      "mean_token_accuracy": 0.21220056265592574,
      "num_tokens": 75922648.0,
      "step": 41145
    },
    {
      "entropy": 5.364724397659302,
      "epoch": 3.4571308548624238,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038327804166981525,
      "loss": 4.9402,
      "mean_token_accuracy": 0.2019209235906601,
      "num_tokens": 75930532.0,
      "step": 41150
    },
    {
      "entropy": 5.447324514389038,
      "epoch": 3.4575509346775886,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038325178982884754,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2094930350780487,
      "num_tokens": 75939641.0,
      "step": 41155
    },
    {
      "entropy": 5.551626205444336,
      "epoch": 3.457971014492754,
      "grad_norm": 1.125,
      "learning_rate": 0.0003832255360702553,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20238253325223923,
      "num_tokens": 75948197.0,
      "step": 41160
    },
    {
      "entropy": 5.497283601760865,
      "epoch": 3.4583910943079186,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038319928039450327,
      "loss": 5.0058,
      "mean_token_accuracy": 0.198265840113163,
      "num_tokens": 75957723.0,
      "step": 41165
    },
    {
      "entropy": 5.468540239334106,
      "epoch": 3.4588111741230834,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038317302280205674,
      "loss": 4.9827,
      "mean_token_accuracy": 0.2085440680384636,
      "num_tokens": 75966115.0,
      "step": 41170
    },
    {
      "entropy": 5.44486255645752,
      "epoch": 3.459231253938248,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003831467632933808,
      "loss": 4.953,
      "mean_token_accuracy": 0.2131699040532112,
      "num_tokens": 75975533.0,
      "step": 41175
    },
    {
      "entropy": 5.485549783706665,
      "epoch": 3.4596513337534134,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003831205018689407,
      "loss": 5.054,
      "mean_token_accuracy": 0.2000655859708786,
      "num_tokens": 75985019.0,
      "step": 41180
    },
    {
      "entropy": 5.436942434310913,
      "epoch": 3.460071413568578,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038309423852920155,
      "loss": 4.9946,
      "mean_token_accuracy": 0.20653147250413895,
      "num_tokens": 75993643.0,
      "step": 41185
    },
    {
      "entropy": 5.387221908569336,
      "epoch": 3.460491493383743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003830679732746286,
      "loss": 4.9724,
      "mean_token_accuracy": 0.2077716752886772,
      "num_tokens": 76001924.0,
      "step": 41190
    },
    {
      "entropy": 5.461418390274048,
      "epoch": 3.4609115731989077,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003830417061056872,
      "loss": 4.9209,
      "mean_token_accuracy": 0.20479716360569,
      "num_tokens": 76010886.0,
      "step": 41195
    },
    {
      "entropy": 5.377589702606201,
      "epoch": 3.4613316530140725,
      "grad_norm": 1.0,
      "learning_rate": 0.0003830154370228426,
      "loss": 4.9741,
      "mean_token_accuracy": 0.19836679548025132,
      "num_tokens": 76020232.0,
      "step": 41200
    },
    {
      "entropy": 5.447286367416382,
      "epoch": 3.4617517328292378,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038298916602656015,
      "loss": 5.0442,
      "mean_token_accuracy": 0.2030433803796768,
      "num_tokens": 76028998.0,
      "step": 41205
    },
    {
      "entropy": 5.483960771560669,
      "epoch": 3.4621718126444025,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038296289311730514,
      "loss": 5.0253,
      "mean_token_accuracy": 0.20193239003419877,
      "num_tokens": 76038283.0,
      "step": 41210
    },
    {
      "entropy": 5.545046758651734,
      "epoch": 3.4625918924595673,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003829366182955432,
      "loss": 5.0472,
      "mean_token_accuracy": 0.20486552715301515,
      "num_tokens": 76046224.0,
      "step": 41215
    },
    {
      "entropy": 5.513156986236572,
      "epoch": 3.463011972274732,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003829103415617396,
      "loss": 5.0507,
      "mean_token_accuracy": 0.20042028725147248,
      "num_tokens": 76055485.0,
      "step": 41220
    },
    {
      "entropy": 5.5351213932037355,
      "epoch": 3.463432052089897,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038288406291635984,
      "loss": 5.0754,
      "mean_token_accuracy": 0.20757190585136415,
      "num_tokens": 76065540.0,
      "step": 41225
    },
    {
      "entropy": 5.387123775482178,
      "epoch": 3.463852131905062,
      "grad_norm": 1.125,
      "learning_rate": 0.0003828577823598695,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20220641791820526,
      "num_tokens": 76074498.0,
      "step": 41230
    },
    {
      "entropy": 5.485323429107666,
      "epoch": 3.464272211720227,
      "grad_norm": 1.0625,
      "learning_rate": 0.000382831499892734,
      "loss": 5.0121,
      "mean_token_accuracy": 0.20539361238479614,
      "num_tokens": 76083852.0,
      "step": 41235
    },
    {
      "entropy": 5.600309562683106,
      "epoch": 3.4646922915353917,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003828052155154191,
      "loss": 5.0287,
      "mean_token_accuracy": 0.1990771248936653,
      "num_tokens": 76093024.0,
      "step": 41240
    },
    {
      "entropy": 5.457973289489746,
      "epoch": 3.4651123713505565,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003827789292283903,
      "loss": 4.9785,
      "mean_token_accuracy": 0.20452190339565277,
      "num_tokens": 76102121.0,
      "step": 41245
    },
    {
      "entropy": 5.516783618927002,
      "epoch": 3.4655324511657213,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003827526410321132,
      "loss": 5.035,
      "mean_token_accuracy": 0.2010795295238495,
      "num_tokens": 76111142.0,
      "step": 41250
    },
    {
      "entropy": 5.513057613372803,
      "epoch": 3.4659525309808865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003827263509270536,
      "loss": 5.1277,
      "mean_token_accuracy": 0.19275751560926438,
      "num_tokens": 76120731.0,
      "step": 41255
    },
    {
      "entropy": 5.5383704662322994,
      "epoch": 3.4663726107960513,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038270005891367714,
      "loss": 5.0549,
      "mean_token_accuracy": 0.20636885315179826,
      "num_tokens": 76130596.0,
      "step": 41260
    },
    {
      "entropy": 5.566382646560669,
      "epoch": 3.466792690611216,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038267376499244964,
      "loss": 5.0195,
      "mean_token_accuracy": 0.20265950560569762,
      "num_tokens": 76139608.0,
      "step": 41265
    },
    {
      "entropy": 5.380749654769898,
      "epoch": 3.467212770426381,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003826474691638368,
      "loss": 4.918,
      "mean_token_accuracy": 0.20795429944992067,
      "num_tokens": 76148920.0,
      "step": 41270
    },
    {
      "entropy": 5.420613431930542,
      "epoch": 3.4676328502415457,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003826211714283045,
      "loss": 5.0139,
      "mean_token_accuracy": 0.20709948390722274,
      "num_tokens": 76158078.0,
      "step": 41275
    },
    {
      "entropy": 5.552948999404907,
      "epoch": 3.468052930056711,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003825948717863186,
      "loss": 5.165,
      "mean_token_accuracy": 0.19669618606567382,
      "num_tokens": 76167646.0,
      "step": 41280
    },
    {
      "entropy": 5.447742700576782,
      "epoch": 3.4684730098718757,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038256857023834484,
      "loss": 4.9421,
      "mean_token_accuracy": 0.20243945419788362,
      "num_tokens": 76176963.0,
      "step": 41285
    },
    {
      "entropy": 5.51794056892395,
      "epoch": 3.4688930896870405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003825422667848494,
      "loss": 5.0566,
      "mean_token_accuracy": 0.19262427985668182,
      "num_tokens": 76186315.0,
      "step": 41290
    },
    {
      "entropy": 5.57488431930542,
      "epoch": 3.4693131695022053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038251596142629794,
      "loss": 5.1056,
      "mean_token_accuracy": 0.1958943247795105,
      "num_tokens": 76195425.0,
      "step": 41295
    },
    {
      "entropy": 5.480070161819458,
      "epoch": 3.4697332493173705,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003824896541631566,
      "loss": 5.082,
      "mean_token_accuracy": 0.19537518918514252,
      "num_tokens": 76204843.0,
      "step": 41300
    },
    {
      "entropy": 5.42186713218689,
      "epoch": 3.4701533291325353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038246334499589143,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20921023339033126,
      "num_tokens": 76213880.0,
      "step": 41305
    },
    {
      "entropy": 5.537519550323486,
      "epoch": 3.4705734089477,
      "grad_norm": 1.0,
      "learning_rate": 0.0003824370339249684,
      "loss": 4.9897,
      "mean_token_accuracy": 0.20871757268905639,
      "num_tokens": 76222849.0,
      "step": 41310
    },
    {
      "entropy": 5.461475467681884,
      "epoch": 3.470993488762865,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003824107209508537,
      "loss": 4.9523,
      "mean_token_accuracy": 0.20866914540529252,
      "num_tokens": 76231893.0,
      "step": 41315
    },
    {
      "entropy": 5.422045087814331,
      "epoch": 3.47141356857803,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003823844060740133,
      "loss": 4.9617,
      "mean_token_accuracy": 0.2071611002087593,
      "num_tokens": 76241866.0,
      "step": 41320
    },
    {
      "entropy": 5.452189588546753,
      "epoch": 3.471833648393195,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038235808929491355,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20642954260110855,
      "num_tokens": 76251297.0,
      "step": 41325
    },
    {
      "entropy": 5.45873441696167,
      "epoch": 3.4722537282083596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003823317706140205,
      "loss": 4.9774,
      "mean_token_accuracy": 0.2017608866095543,
      "num_tokens": 76260685.0,
      "step": 41330
    },
    {
      "entropy": 5.389981079101562,
      "epoch": 3.4726738080235244,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038230545003180034,
      "loss": 4.9629,
      "mean_token_accuracy": 0.21167934238910674,
      "num_tokens": 76269843.0,
      "step": 41335
    },
    {
      "entropy": 5.433181095123291,
      "epoch": 3.473093887838689,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038227912754871943,
      "loss": 4.9773,
      "mean_token_accuracy": 0.20413582175970077,
      "num_tokens": 76279589.0,
      "step": 41340
    },
    {
      "entropy": 5.5117449283599855,
      "epoch": 3.4735139676538545,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038225280316524403,
      "loss": 5.0647,
      "mean_token_accuracy": 0.19547542482614516,
      "num_tokens": 76288639.0,
      "step": 41345
    },
    {
      "entropy": 5.566097259521484,
      "epoch": 3.4739340474690192,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003822264768818404,
      "loss": 5.0505,
      "mean_token_accuracy": 0.19808817356824876,
      "num_tokens": 76297998.0,
      "step": 41350
    },
    {
      "entropy": 5.464324045181274,
      "epoch": 3.474354127284184,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000382200148698975,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21217647045850754,
      "num_tokens": 76307604.0,
      "step": 41355
    },
    {
      "entropy": 5.547789430618286,
      "epoch": 3.474774207099349,
      "grad_norm": 1.0,
      "learning_rate": 0.00038217381861711407,
      "loss": 5.1268,
      "mean_token_accuracy": 0.20522319078445433,
      "num_tokens": 76317697.0,
      "step": 41360
    },
    {
      "entropy": 5.488579797744751,
      "epoch": 3.4751942869145136,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038214748663672424,
      "loss": 5.0682,
      "mean_token_accuracy": 0.1999670073390007,
      "num_tokens": 76327261.0,
      "step": 41365
    },
    {
      "entropy": 5.484370183944702,
      "epoch": 3.475614366729679,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038212115275827177,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20906213521957398,
      "num_tokens": 76336891.0,
      "step": 41370
    },
    {
      "entropy": 5.425146913528442,
      "epoch": 3.4760344465448436,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003820948169822233,
      "loss": 4.9963,
      "mean_token_accuracy": 0.1989768236875534,
      "num_tokens": 76346283.0,
      "step": 41375
    },
    {
      "entropy": 5.412164688110352,
      "epoch": 3.4764545263600084,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038206847930904526,
      "loss": 5.0054,
      "mean_token_accuracy": 0.1992933541536331,
      "num_tokens": 76355670.0,
      "step": 41380
    },
    {
      "entropy": 5.478922939300537,
      "epoch": 3.476874606175173,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038204213973920423,
      "loss": 4.9594,
      "mean_token_accuracy": 0.2093760073184967,
      "num_tokens": 76364450.0,
      "step": 41385
    },
    {
      "entropy": 5.450120973587036,
      "epoch": 3.477294685990338,
      "grad_norm": 0.984375,
      "learning_rate": 0.00038201579827316684,
      "loss": 5.0502,
      "mean_token_accuracy": 0.19755571186542512,
      "num_tokens": 76374463.0,
      "step": 41390
    },
    {
      "entropy": 5.548734426498413,
      "epoch": 3.477714765805503,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003819894549113997,
      "loss": 5.1052,
      "mean_token_accuracy": 0.20361521989107131,
      "num_tokens": 76383772.0,
      "step": 41395
    },
    {
      "entropy": 5.510997533798218,
      "epoch": 3.478134845620668,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00038196310965436944,
      "loss": 4.9835,
      "mean_token_accuracy": 0.20368375927209853,
      "num_tokens": 76392463.0,
      "step": 41400
    },
    {
      "entropy": 5.437485027313232,
      "epoch": 3.478554925435833,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038193676250254275,
      "loss": 4.943,
      "mean_token_accuracy": 0.2049142301082611,
      "num_tokens": 76401193.0,
      "step": 41405
    },
    {
      "entropy": 5.5142559051513675,
      "epoch": 3.4789750052509976,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038191041345638643,
      "loss": 5.1035,
      "mean_token_accuracy": 0.19211620539426805,
      "num_tokens": 76410069.0,
      "step": 41410
    },
    {
      "entropy": 5.497405099868774,
      "epoch": 3.4793950850661624,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038188406251636717,
      "loss": 4.9701,
      "mean_token_accuracy": 0.20461506247520447,
      "num_tokens": 76419501.0,
      "step": 41415
    },
    {
      "entropy": 5.434232473373413,
      "epoch": 3.4798151648813276,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038185770968295177,
      "loss": 4.9166,
      "mean_token_accuracy": 0.20560792982578277,
      "num_tokens": 76428387.0,
      "step": 41420
    },
    {
      "entropy": 5.489426708221435,
      "epoch": 3.4802352446964924,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038183135495660703,
      "loss": 4.9807,
      "mean_token_accuracy": 0.20898114889860153,
      "num_tokens": 76437358.0,
      "step": 41425
    },
    {
      "entropy": 5.4951738834381105,
      "epoch": 3.480655324511657,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003818049983377999,
      "loss": 5.0611,
      "mean_token_accuracy": 0.20083450824022292,
      "num_tokens": 76446368.0,
      "step": 41430
    },
    {
      "entropy": 5.477327680587768,
      "epoch": 3.481075404326822,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003817786398269972,
      "loss": 4.9643,
      "mean_token_accuracy": 0.20491454899311065,
      "num_tokens": 76455646.0,
      "step": 41435
    },
    {
      "entropy": 5.456271696090698,
      "epoch": 3.4814954841419867,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003817522794246658,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20652543902397155,
      "num_tokens": 76464257.0,
      "step": 41440
    },
    {
      "entropy": 5.536167335510254,
      "epoch": 3.481915563957152,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038172591713127286,
      "loss": 5.0635,
      "mean_token_accuracy": 0.2037614479660988,
      "num_tokens": 76473196.0,
      "step": 41445
    },
    {
      "entropy": 5.561646270751953,
      "epoch": 3.4823356437723167,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003816995529472853,
      "loss": 5.0207,
      "mean_token_accuracy": 0.19664284586906433,
      "num_tokens": 76482448.0,
      "step": 41450
    },
    {
      "entropy": 5.459619426727295,
      "epoch": 3.4827557235874815,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038167318687316995,
      "loss": 4.9973,
      "mean_token_accuracy": 0.20430362820625306,
      "num_tokens": 76491934.0,
      "step": 41455
    },
    {
      "entropy": 5.549050998687744,
      "epoch": 3.4831758034026463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038164681890939404,
      "loss": 5.065,
      "mean_token_accuracy": 0.1905907854437828,
      "num_tokens": 76501156.0,
      "step": 41460
    },
    {
      "entropy": 5.529715490341187,
      "epoch": 3.4835958832178116,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003816204490564247,
      "loss": 5.1184,
      "mean_token_accuracy": 0.2037932261824608,
      "num_tokens": 76510618.0,
      "step": 41465
    },
    {
      "entropy": 5.574638032913208,
      "epoch": 3.4840159630329763,
      "grad_norm": 1.0625,
      "learning_rate": 0.000381594077314729,
      "loss": 5.075,
      "mean_token_accuracy": 0.1907612606883049,
      "num_tokens": 76519996.0,
      "step": 41470
    },
    {
      "entropy": 5.550010824203492,
      "epoch": 3.484436042848141,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038156770368477416,
      "loss": 5.0479,
      "mean_token_accuracy": 0.2071364030241966,
      "num_tokens": 76529689.0,
      "step": 41475
    },
    {
      "entropy": 5.592033433914184,
      "epoch": 3.484856122663306,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038154132816702724,
      "loss": 5.095,
      "mean_token_accuracy": 0.19663961976766586,
      "num_tokens": 76539661.0,
      "step": 41480
    },
    {
      "entropy": 5.441378784179688,
      "epoch": 3.485276202478471,
      "grad_norm": 1.25,
      "learning_rate": 0.0003815149507619556,
      "loss": 4.9755,
      "mean_token_accuracy": 0.2033386066555977,
      "num_tokens": 76549506.0,
      "step": 41485
    },
    {
      "entropy": 5.496755170822143,
      "epoch": 3.485696282293636,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003814885714700264,
      "loss": 5.0136,
      "mean_token_accuracy": 0.20057215988636018,
      "num_tokens": 76558789.0,
      "step": 41490
    },
    {
      "entropy": 5.312194871902466,
      "epoch": 3.4861163621088007,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038146219029170703,
      "loss": 4.7946,
      "mean_token_accuracy": 0.21828253716230392,
      "num_tokens": 76567206.0,
      "step": 41495
    },
    {
      "entropy": 5.408727979660034,
      "epoch": 3.4865364419239655,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003814358072274648,
      "loss": 5.0413,
      "mean_token_accuracy": 0.20279616117477417,
      "num_tokens": 76576853.0,
      "step": 41500
    },
    {
      "entropy": 5.525135946273804,
      "epoch": 3.4869565217391303,
      "grad_norm": 1.03125,
      "learning_rate": 0.000381409422277767,
      "loss": 5.058,
      "mean_token_accuracy": 0.20065719187259673,
      "num_tokens": 76585933.0,
      "step": 41505
    },
    {
      "entropy": 5.492447376251221,
      "epoch": 3.4873766015542955,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038138303544308104,
      "loss": 4.9729,
      "mean_token_accuracy": 0.20314229428768157,
      "num_tokens": 76596144.0,
      "step": 41510
    },
    {
      "entropy": 5.422357845306396,
      "epoch": 3.4877966813694603,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038135664672387455,
      "loss": 4.8904,
      "mean_token_accuracy": 0.20911831110715867,
      "num_tokens": 76605607.0,
      "step": 41515
    },
    {
      "entropy": 5.501010751724243,
      "epoch": 3.488216761184625,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038133025612061466,
      "loss": 4.9868,
      "mean_token_accuracy": 0.2071363851428032,
      "num_tokens": 76614312.0,
      "step": 41520
    },
    {
      "entropy": 5.401148509979248,
      "epoch": 3.48863684099979,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038130386363376915,
      "loss": 5.0002,
      "mean_token_accuracy": 0.19533602446317672,
      "num_tokens": 76623065.0,
      "step": 41525
    },
    {
      "entropy": 5.429398679733277,
      "epoch": 3.4890569208149547,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038127746926380535,
      "loss": 4.9921,
      "mean_token_accuracy": 0.20523834079504014,
      "num_tokens": 76632745.0,
      "step": 41530
    },
    {
      "entropy": 5.447716856002808,
      "epoch": 3.48947700063012,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000381251073011191,
      "loss": 4.97,
      "mean_token_accuracy": 0.21005673408508302,
      "num_tokens": 76641623.0,
      "step": 41535
    },
    {
      "entropy": 5.4381507396697994,
      "epoch": 3.4898970804452847,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038122467487639353,
      "loss": 4.9468,
      "mean_token_accuracy": 0.20720257610082626,
      "num_tokens": 76651807.0,
      "step": 41540
    },
    {
      "entropy": 5.436652040481567,
      "epoch": 3.4903171602604495,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038119827485988064,
      "loss": 4.9368,
      "mean_token_accuracy": 0.2162679299712181,
      "num_tokens": 76660685.0,
      "step": 41545
    },
    {
      "entropy": 5.389433145523071,
      "epoch": 3.4907372400756143,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038117187296212004,
      "loss": 4.9593,
      "mean_token_accuracy": 0.20965682715177536,
      "num_tokens": 76669883.0,
      "step": 41550
    },
    {
      "entropy": 5.505159044265747,
      "epoch": 3.491157319890779,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003811454691835794,
      "loss": 5.0871,
      "mean_token_accuracy": 0.19735568016767502,
      "num_tokens": 76679632.0,
      "step": 41555
    },
    {
      "entropy": 5.471107864379883,
      "epoch": 3.4915773997059443,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038111906352472635,
      "loss": 4.9751,
      "mean_token_accuracy": 0.20092821419239043,
      "num_tokens": 76689812.0,
      "step": 41560
    },
    {
      "entropy": 5.392461252212525,
      "epoch": 3.491997479521109,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038109265598602876,
      "loss": 4.8893,
      "mean_token_accuracy": 0.21603091955184936,
      "num_tokens": 76698812.0,
      "step": 41565
    },
    {
      "entropy": 5.477409219741821,
      "epoch": 3.492417559336274,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003810662465679544,
      "loss": 4.9629,
      "mean_token_accuracy": 0.204941463470459,
      "num_tokens": 76708533.0,
      "step": 41570
    },
    {
      "entropy": 5.367539978027343,
      "epoch": 3.4928376391514386,
      "grad_norm": 1.125,
      "learning_rate": 0.0003810398352709711,
      "loss": 4.9531,
      "mean_token_accuracy": 0.2046682730317116,
      "num_tokens": 76717374.0,
      "step": 41575
    },
    {
      "entropy": 5.430282354354858,
      "epoch": 3.4932577189666034,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003810134220955467,
      "loss": 4.9109,
      "mean_token_accuracy": 0.20071182698011397,
      "num_tokens": 76727220.0,
      "step": 41580
    },
    {
      "entropy": 5.456218099594116,
      "epoch": 3.4936777987817687,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038098700704214915,
      "loss": 4.9284,
      "mean_token_accuracy": 0.2067394196987152,
      "num_tokens": 76735294.0,
      "step": 41585
    },
    {
      "entropy": 5.40539665222168,
      "epoch": 3.4940978785969334,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038096059011124634,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20655650943517684,
      "num_tokens": 76744713.0,
      "step": 41590
    },
    {
      "entropy": 5.417543220520019,
      "epoch": 3.4945179584120982,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038093417130330627,
      "loss": 5.0187,
      "mean_token_accuracy": 0.20302670150995256,
      "num_tokens": 76754310.0,
      "step": 41595
    },
    {
      "entropy": 5.473627138137817,
      "epoch": 3.494938038227263,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003809077506187968,
      "loss": 4.937,
      "mean_token_accuracy": 0.20508293509483339,
      "num_tokens": 76763381.0,
      "step": 41600
    },
    {
      "entropy": 5.567989683151245,
      "epoch": 3.4953581180424282,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038088132805818615,
      "loss": 5.0067,
      "mean_token_accuracy": 0.20718769282102584,
      "num_tokens": 76773529.0,
      "step": 41605
    },
    {
      "entropy": 5.373406267166137,
      "epoch": 3.495778197857593,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003808549036219423,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2076689586043358,
      "num_tokens": 76782942.0,
      "step": 41610
    },
    {
      "entropy": 5.403111743927002,
      "epoch": 3.496198277672758,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003808284773105333,
      "loss": 4.9911,
      "mean_token_accuracy": 0.19913169592618943,
      "num_tokens": 76792492.0,
      "step": 41615
    },
    {
      "entropy": 5.514293432235718,
      "epoch": 3.4966183574879226,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038080204912442736,
      "loss": 5.084,
      "mean_token_accuracy": 0.20874502509832382,
      "num_tokens": 76801123.0,
      "step": 41620
    },
    {
      "entropy": 5.468956136703492,
      "epoch": 3.497038437303088,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003807756190640926,
      "loss": 4.9592,
      "mean_token_accuracy": 0.20443494468927384,
      "num_tokens": 76809967.0,
      "step": 41625
    },
    {
      "entropy": 5.447124385833741,
      "epoch": 3.4974585171182526,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003807491871299971,
      "loss": 4.995,
      "mean_token_accuracy": 0.20210361033678054,
      "num_tokens": 76819493.0,
      "step": 41630
    },
    {
      "entropy": 5.413346815109253,
      "epoch": 3.4978785969334174,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003807227533226094,
      "loss": 4.9438,
      "mean_token_accuracy": 0.21345119327306747,
      "num_tokens": 76828904.0,
      "step": 41635
    },
    {
      "entropy": 5.433348321914673,
      "epoch": 3.498298676748582,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003806963176423974,
      "loss": 4.9695,
      "mean_token_accuracy": 0.2031904935836792,
      "num_tokens": 76837779.0,
      "step": 41640
    },
    {
      "entropy": 5.423989582061767,
      "epoch": 3.498718756563747,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038066988008982956,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20887357294559478,
      "num_tokens": 76847351.0,
      "step": 41645
    },
    {
      "entropy": 5.4215833187103275,
      "epoch": 3.499138836378912,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038064344066537426,
      "loss": 4.9931,
      "mean_token_accuracy": 0.2038634642958641,
      "num_tokens": 76856845.0,
      "step": 41650
    },
    {
      "entropy": 5.419219255447388,
      "epoch": 3.499558916194077,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003806169993694998,
      "loss": 4.9494,
      "mean_token_accuracy": 0.2114509090781212,
      "num_tokens": 76866644.0,
      "step": 41655
    },
    {
      "entropy": 5.54095458984375,
      "epoch": 3.499978996009242,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003805905562026746,
      "loss": 5.0078,
      "mean_token_accuracy": 0.20069498866796492,
      "num_tokens": 76876853.0,
      "step": 41660
    },
    {
      "entropy": 5.491980028152466,
      "epoch": 3.5003990758244066,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003805641111653671,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20863333940505982,
      "num_tokens": 76885474.0,
      "step": 41665
    },
    {
      "entropy": 5.427982378005981,
      "epoch": 3.5008191556395714,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038053766425804567,
      "loss": 5.0518,
      "mean_token_accuracy": 0.20182936638593674,
      "num_tokens": 76894270.0,
      "step": 41670
    },
    {
      "entropy": 5.4582959651947025,
      "epoch": 3.5012392354547366,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003805112154811789,
      "loss": 5.0372,
      "mean_token_accuracy": 0.19543871879577637,
      "num_tokens": 76903142.0,
      "step": 41675
    },
    {
      "entropy": 5.452586317062378,
      "epoch": 3.5016593152699014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038048476483523524,
      "loss": 4.9668,
      "mean_token_accuracy": 0.2071259245276451,
      "num_tokens": 76912221.0,
      "step": 41680
    },
    {
      "entropy": 5.463269186019898,
      "epoch": 3.502079395085066,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003804583123206833,
      "loss": 4.9673,
      "mean_token_accuracy": 0.1995807021856308,
      "num_tokens": 76921107.0,
      "step": 41685
    },
    {
      "entropy": 5.448861455917358,
      "epoch": 3.502499474900231,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003804318579379917,
      "loss": 4.9793,
      "mean_token_accuracy": 0.20725654512643815,
      "num_tokens": 76930101.0,
      "step": 41690
    },
    {
      "entropy": 5.467288541793823,
      "epoch": 3.5029195547153957,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038040540168762896,
      "loss": 4.9795,
      "mean_token_accuracy": 0.20588523000478745,
      "num_tokens": 76938360.0,
      "step": 41695
    },
    {
      "entropy": 5.428915023803711,
      "epoch": 3.503339634530561,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003803789435700638,
      "loss": 4.9604,
      "mean_token_accuracy": 0.21153463274240494,
      "num_tokens": 76947636.0,
      "step": 41700
    },
    {
      "entropy": 5.417567873001099,
      "epoch": 3.5037597143457258,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038035248358576494,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20383463203907012,
      "num_tokens": 76956383.0,
      "step": 41705
    },
    {
      "entropy": 5.4720940589904785,
      "epoch": 3.5041797941608905,
      "grad_norm": 0.9375,
      "learning_rate": 0.00038032602173520105,
      "loss": 4.9816,
      "mean_token_accuracy": 0.20148734897375106,
      "num_tokens": 76966690.0,
      "step": 41710
    },
    {
      "entropy": 5.427651500701904,
      "epoch": 3.5045998739760553,
      "grad_norm": 1.015625,
      "learning_rate": 0.000380299558018841,
      "loss": 4.921,
      "mean_token_accuracy": 0.20785754173994064,
      "num_tokens": 76975910.0,
      "step": 41715
    },
    {
      "entropy": 5.464867353439331,
      "epoch": 3.50501995379122,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038027309243715345,
      "loss": 5.046,
      "mean_token_accuracy": 0.20439719259738923,
      "num_tokens": 76984562.0,
      "step": 41720
    },
    {
      "entropy": 5.437987470626831,
      "epoch": 3.5054400336063853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003802466249906073,
      "loss": 5.0276,
      "mean_token_accuracy": 0.21215792000293732,
      "num_tokens": 76994533.0,
      "step": 41725
    },
    {
      "entropy": 5.38671841621399,
      "epoch": 3.50586011342155,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003802201556796713,
      "loss": 4.9044,
      "mean_token_accuracy": 0.20820268541574477,
      "num_tokens": 77003380.0,
      "step": 41730
    },
    {
      "entropy": 5.442011785507202,
      "epoch": 3.506280193236715,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003801936845048145,
      "loss": 4.9377,
      "mean_token_accuracy": 0.20462663173675538,
      "num_tokens": 77011633.0,
      "step": 41735
    },
    {
      "entropy": 5.4338420867919925,
      "epoch": 3.5067002730518797,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003801672114665058,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2091939464211464,
      "num_tokens": 77019578.0,
      "step": 41740
    },
    {
      "entropy": 5.438662433624268,
      "epoch": 3.5071203528670445,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00038014073656521404,
      "loss": 4.9942,
      "mean_token_accuracy": 0.2033005639910698,
      "num_tokens": 77029989.0,
      "step": 41745
    },
    {
      "entropy": 5.480625963211059,
      "epoch": 3.5075404326822097,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003801142598014083,
      "loss": 4.9123,
      "mean_token_accuracy": 0.209139646589756,
      "num_tokens": 77038598.0,
      "step": 41750
    },
    {
      "entropy": 5.455771255493164,
      "epoch": 3.5079605124973745,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003800877811755575,
      "loss": 4.9688,
      "mean_token_accuracy": 0.2023881882429123,
      "num_tokens": 77047903.0,
      "step": 41755
    },
    {
      "entropy": 5.4197601795196535,
      "epoch": 3.5083805923125393,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003800613006881308,
      "loss": 4.9798,
      "mean_token_accuracy": 0.19827140271663665,
      "num_tokens": 77058100.0,
      "step": 41760
    },
    {
      "entropy": 5.487070465087891,
      "epoch": 3.5088006721277045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003800348183395972,
      "loss": 5.0165,
      "mean_token_accuracy": 0.2006743296980858,
      "num_tokens": 77067263.0,
      "step": 41765
    },
    {
      "entropy": 5.453813695907593,
      "epoch": 3.5092207519428693,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000380008334130426,
      "loss": 4.8875,
      "mean_token_accuracy": 0.2132476046681404,
      "num_tokens": 77076126.0,
      "step": 41770
    },
    {
      "entropy": 5.441852045059204,
      "epoch": 3.509640831758034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037998184806108614,
      "loss": 5.0157,
      "mean_token_accuracy": 0.19839726835489274,
      "num_tokens": 77085125.0,
      "step": 41775
    },
    {
      "entropy": 5.454864358901977,
      "epoch": 3.510060911573199,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003799553601320469,
      "loss": 5.0721,
      "mean_token_accuracy": 0.2022296518087387,
      "num_tokens": 77095129.0,
      "step": 41780
    },
    {
      "entropy": 5.473551082611084,
      "epoch": 3.5104809913883637,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003799288703437776,
      "loss": 4.884,
      "mean_token_accuracy": 0.22007420063018798,
      "num_tokens": 77104291.0,
      "step": 41785
    },
    {
      "entropy": 5.462594652175904,
      "epoch": 3.510901071203529,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003799023786967473,
      "loss": 5.0444,
      "mean_token_accuracy": 0.19900732189416886,
      "num_tokens": 77113909.0,
      "step": 41790
    },
    {
      "entropy": 5.408506536483765,
      "epoch": 3.5113211510186937,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003798758851914254,
      "loss": 4.9128,
      "mean_token_accuracy": 0.2111838549375534,
      "num_tokens": 77122819.0,
      "step": 41795
    },
    {
      "entropy": 5.402147388458252,
      "epoch": 3.5117412308338585,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003798493898282813,
      "loss": 4.9436,
      "mean_token_accuracy": 0.20872508883476257,
      "num_tokens": 77132194.0,
      "step": 41800
    },
    {
      "entropy": 5.3920886516571045,
      "epoch": 3.5121613106490233,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037982289260778406,
      "loss": 4.8725,
      "mean_token_accuracy": 0.21803192496299745,
      "num_tokens": 77140867.0,
      "step": 41805
    },
    {
      "entropy": 5.436008644104004,
      "epoch": 3.512581390464188,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037979639353040337,
      "loss": 4.953,
      "mean_token_accuracy": 0.20701647847890853,
      "num_tokens": 77149518.0,
      "step": 41810
    },
    {
      "entropy": 5.46375207901001,
      "epoch": 3.5130014702793533,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003797698925966085,
      "loss": 4.9714,
      "mean_token_accuracy": 0.20925831347703933,
      "num_tokens": 77158918.0,
      "step": 41815
    },
    {
      "entropy": 5.4783440113067625,
      "epoch": 3.513421550094518,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037974338980686895,
      "loss": 4.9979,
      "mean_token_accuracy": 0.20868863463401793,
      "num_tokens": 77167704.0,
      "step": 41820
    },
    {
      "entropy": 5.409232759475708,
      "epoch": 3.513841629909683,
      "grad_norm": 1.125,
      "learning_rate": 0.00037971688516165415,
      "loss": 4.9558,
      "mean_token_accuracy": 0.2040602907538414,
      "num_tokens": 77176515.0,
      "step": 41825
    },
    {
      "entropy": 5.462631750106811,
      "epoch": 3.5142617097248476,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037969037866143366,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20808477997779845,
      "num_tokens": 77185823.0,
      "step": 41830
    },
    {
      "entropy": 5.565123128890991,
      "epoch": 3.5146817895400124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000379663870306677,
      "loss": 5.1369,
      "mean_token_accuracy": 0.1948266252875328,
      "num_tokens": 77194924.0,
      "step": 41835
    },
    {
      "entropy": 5.485260105133056,
      "epoch": 3.5151018693551777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003796373600978539,
      "loss": 4.9994,
      "mean_token_accuracy": 0.20115850120782852,
      "num_tokens": 77204225.0,
      "step": 41840
    },
    {
      "entropy": 5.425109004974365,
      "epoch": 3.5155219491703424,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037961084803543366,
      "loss": 4.9707,
      "mean_token_accuracy": 0.2040533557534218,
      "num_tokens": 77213497.0,
      "step": 41845
    },
    {
      "entropy": 5.35848445892334,
      "epoch": 3.5159420289855072,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003795843341198862,
      "loss": 4.8813,
      "mean_token_accuracy": 0.20770494192838668,
      "num_tokens": 77223240.0,
      "step": 41850
    },
    {
      "entropy": 5.416385269165039,
      "epoch": 3.516362108800672,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003795578183516811,
      "loss": 4.9335,
      "mean_token_accuracy": 0.20744079649448394,
      "num_tokens": 77231892.0,
      "step": 41855
    },
    {
      "entropy": 5.45360631942749,
      "epoch": 3.516782188615837,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037953130073128813,
      "loss": 4.9705,
      "mean_token_accuracy": 0.1986423835158348,
      "num_tokens": 77241383.0,
      "step": 41860
    },
    {
      "entropy": 5.445881605148315,
      "epoch": 3.517202268431002,
      "grad_norm": 1.0625,
      "learning_rate": 0.000379504781259177,
      "loss": 5.0846,
      "mean_token_accuracy": 0.1969805508852005,
      "num_tokens": 77250772.0,
      "step": 41865
    },
    {
      "entropy": 5.394086790084839,
      "epoch": 3.517622348246167,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037947825993581743,
      "loss": 4.874,
      "mean_token_accuracy": 0.21613519489765168,
      "num_tokens": 77259806.0,
      "step": 41870
    },
    {
      "entropy": 5.538832426071167,
      "epoch": 3.5180424280613316,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037945173676167933,
      "loss": 5.1587,
      "mean_token_accuracy": 0.19643352478742598,
      "num_tokens": 77269459.0,
      "step": 41875
    },
    {
      "entropy": 5.495407629013061,
      "epoch": 3.5184625078764964,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003794252117372324,
      "loss": 5.0196,
      "mean_token_accuracy": 0.2005572110414505,
      "num_tokens": 77279906.0,
      "step": 41880
    },
    {
      "entropy": 5.542854118347168,
      "epoch": 3.518882587691661,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037939868486294676,
      "loss": 5.0403,
      "mean_token_accuracy": 0.2052692860364914,
      "num_tokens": 77288532.0,
      "step": 41885
    },
    {
      "entropy": 5.468048143386841,
      "epoch": 3.5193026675068264,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037937215613929203,
      "loss": 4.967,
      "mean_token_accuracy": 0.19978157132864,
      "num_tokens": 77297761.0,
      "step": 41890
    },
    {
      "entropy": 5.463889074325562,
      "epoch": 3.519722747321991,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003793456255667384,
      "loss": 4.9314,
      "mean_token_accuracy": 0.2092806354165077,
      "num_tokens": 77306156.0,
      "step": 41895
    },
    {
      "entropy": 5.434077262878418,
      "epoch": 3.520142827137156,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037931909314575573,
      "loss": 5.0147,
      "mean_token_accuracy": 0.20358773916959763,
      "num_tokens": 77315757.0,
      "step": 41900
    },
    {
      "entropy": 5.484953880310059,
      "epoch": 3.520562906952321,
      "grad_norm": 1.015625,
      "learning_rate": 0.000379292558876814,
      "loss": 5.0196,
      "mean_token_accuracy": 0.2063642293214798,
      "num_tokens": 77324878.0,
      "step": 41905
    },
    {
      "entropy": 5.517754364013672,
      "epoch": 3.5209829867674856,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003792660227603833,
      "loss": 5.0385,
      "mean_token_accuracy": 0.19144693464040757,
      "num_tokens": 77334048.0,
      "step": 41910
    },
    {
      "entropy": 5.563178968429566,
      "epoch": 3.521403066582651,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003792394847969337,
      "loss": 5.035,
      "mean_token_accuracy": 0.20359151661396027,
      "num_tokens": 77342993.0,
      "step": 41915
    },
    {
      "entropy": 5.386972236633301,
      "epoch": 3.5218231463978156,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003792129449869353,
      "loss": 4.9842,
      "mean_token_accuracy": 0.2002910554409027,
      "num_tokens": 77351418.0,
      "step": 41920
    },
    {
      "entropy": 5.4707932472229,
      "epoch": 3.5222432262129804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037918640333085824,
      "loss": 4.9647,
      "mean_token_accuracy": 0.20439096838235854,
      "num_tokens": 77360236.0,
      "step": 41925
    },
    {
      "entropy": 5.498051261901855,
      "epoch": 3.5226633060281456,
      "grad_norm": 1.0,
      "learning_rate": 0.0003791598598291727,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20461732447147368,
      "num_tokens": 77369731.0,
      "step": 41930
    },
    {
      "entropy": 5.480401659011841,
      "epoch": 3.5230833858433104,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003791333144823489,
      "loss": 5.008,
      "mean_token_accuracy": 0.2058209151029587,
      "num_tokens": 77378058.0,
      "step": 41935
    },
    {
      "entropy": 5.474089622497559,
      "epoch": 3.523503465658475,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003791067672908571,
      "loss": 5.0559,
      "mean_token_accuracy": 0.20773122906684877,
      "num_tokens": 77387201.0,
      "step": 41940
    },
    {
      "entropy": 5.5145509243011475,
      "epoch": 3.52392354547364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003790802182551674,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20176197290420533,
      "num_tokens": 77395896.0,
      "step": 41945
    },
    {
      "entropy": 5.542530632019043,
      "epoch": 3.5243436252888047,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037905366737575033,
      "loss": 5.0333,
      "mean_token_accuracy": 0.20216994136571884,
      "num_tokens": 77405560.0,
      "step": 41950
    },
    {
      "entropy": 5.466624546051025,
      "epoch": 3.52476370510397,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003790271146530761,
      "loss": 5.0013,
      "mean_token_accuracy": 0.20502804666757585,
      "num_tokens": 77414222.0,
      "step": 41955
    },
    {
      "entropy": 5.513001441955566,
      "epoch": 3.5251837849191348,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037900056008761506,
      "loss": 5.032,
      "mean_token_accuracy": 0.2027486488223076,
      "num_tokens": 77423308.0,
      "step": 41960
    },
    {
      "entropy": 5.373373365402221,
      "epoch": 3.5256038647342995,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003789740036798377,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20685062259435655,
      "num_tokens": 77431905.0,
      "step": 41965
    },
    {
      "entropy": 5.523012256622314,
      "epoch": 3.5260239445494643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003789474454302144,
      "loss": 5.0196,
      "mean_token_accuracy": 0.19937238842248917,
      "num_tokens": 77439857.0,
      "step": 41970
    },
    {
      "entropy": 5.505891370773315,
      "epoch": 3.526444024364629,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037892088533921555,
      "loss": 5.0526,
      "mean_token_accuracy": 0.20392104387283325,
      "num_tokens": 77449209.0,
      "step": 41975
    },
    {
      "entropy": 5.460748720169067,
      "epoch": 3.5268641041797943,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003788943234073118,
      "loss": 5.04,
      "mean_token_accuracy": 0.19366613030433655,
      "num_tokens": 77457966.0,
      "step": 41980
    },
    {
      "entropy": 5.407882022857666,
      "epoch": 3.527284183994959,
      "grad_norm": 1.125,
      "learning_rate": 0.00037886775963497354,
      "loss": 5.0071,
      "mean_token_accuracy": 0.2084927037358284,
      "num_tokens": 77468235.0,
      "step": 41985
    },
    {
      "entropy": 5.463525915145874,
      "epoch": 3.527704263810124,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037884119402267145,
      "loss": 4.9015,
      "mean_token_accuracy": 0.2120345950126648,
      "num_tokens": 77476717.0,
      "step": 41990
    },
    {
      "entropy": 5.453153276443482,
      "epoch": 3.5281243436252887,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037881462657087607,
      "loss": 4.9575,
      "mean_token_accuracy": 0.2137133702635765,
      "num_tokens": 77485400.0,
      "step": 41995
    },
    {
      "entropy": 5.534685754776001,
      "epoch": 3.5285444234404535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003787880572800581,
      "loss": 5.0586,
      "mean_token_accuracy": 0.2001255050301552,
      "num_tokens": 77494266.0,
      "step": 42000
    },
    {
      "epoch": 3.5285444234404535,
      "eval_entropy": 5.18405020770361,
      "eval_loss": 5.103304862976074,
      "eval_mean_token_accuracy": 0.20719974023295762,
      "eval_num_tokens": 77494266.0,
      "eval_runtime": 27.4964,
      "eval_samples_per_second": 1358.939,
      "eval_steps_per_second": 169.876,
      "step": 42000
    },
    {
      "entropy": 5.408473205566406,
      "epoch": 3.5289645032556187,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037876148615068804,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20554998368024827,
      "num_tokens": 77503768.0,
      "step": 42005
    },
    {
      "entropy": 5.389193105697632,
      "epoch": 3.5293845830707835,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037873491318323675,
      "loss": 4.8978,
      "mean_token_accuracy": 0.21567351669073104,
      "num_tokens": 77511995.0,
      "step": 42010
    },
    {
      "entropy": 5.492789936065674,
      "epoch": 3.5298046628859483,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003787083383781749,
      "loss": 5.0187,
      "mean_token_accuracy": 0.20078884065151215,
      "num_tokens": 77521906.0,
      "step": 42015
    },
    {
      "entropy": 5.579571151733399,
      "epoch": 3.530224742701113,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003786817617359732,
      "loss": 5.0917,
      "mean_token_accuracy": 0.1989911749958992,
      "num_tokens": 77531584.0,
      "step": 42020
    },
    {
      "entropy": 5.51625690460205,
      "epoch": 3.530644822516278,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037865518325710246,
      "loss": 4.9913,
      "mean_token_accuracy": 0.19864429533481598,
      "num_tokens": 77541574.0,
      "step": 42025
    },
    {
      "entropy": 5.468390369415284,
      "epoch": 3.531064902331443,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037862860294203353,
      "loss": 5.074,
      "mean_token_accuracy": 0.19828397631645203,
      "num_tokens": 77551068.0,
      "step": 42030
    },
    {
      "entropy": 5.418768882751465,
      "epoch": 3.531484982146608,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003786020207912373,
      "loss": 4.9936,
      "mean_token_accuracy": 0.20149386674165726,
      "num_tokens": 77560333.0,
      "step": 42035
    },
    {
      "entropy": 5.449148654937744,
      "epoch": 3.5319050619617727,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037857543680518466,
      "loss": 4.9551,
      "mean_token_accuracy": 0.21492572128772736,
      "num_tokens": 77569109.0,
      "step": 42040
    },
    {
      "entropy": 5.420411157608032,
      "epoch": 3.5323251417769375,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037854885098434634,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20854323357343674,
      "num_tokens": 77579108.0,
      "step": 42045
    },
    {
      "entropy": 5.500924301147461,
      "epoch": 3.5327452215921022,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037852226332919347,
      "loss": 5.0883,
      "mean_token_accuracy": 0.19972997009754181,
      "num_tokens": 77588178.0,
      "step": 42050
    },
    {
      "entropy": 5.5727311134338375,
      "epoch": 3.5331653014072675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037849567384019714,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20739773511886597,
      "num_tokens": 77596074.0,
      "step": 42055
    },
    {
      "entropy": 5.425034856796264,
      "epoch": 3.5335853812224323,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003784690825178281,
      "loss": 5.0001,
      "mean_token_accuracy": 0.2052980214357376,
      "num_tokens": 77605061.0,
      "step": 42060
    },
    {
      "entropy": 5.48408317565918,
      "epoch": 3.534005461037597,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037844248936255763,
      "loss": 5.0962,
      "mean_token_accuracy": 0.1907970502972603,
      "num_tokens": 77613843.0,
      "step": 42065
    },
    {
      "entropy": 5.514048719406128,
      "epoch": 3.5344255408527623,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037841589437485673,
      "loss": 5.0484,
      "mean_token_accuracy": 0.1950298637151718,
      "num_tokens": 77623126.0,
      "step": 42070
    },
    {
      "entropy": 5.51355881690979,
      "epoch": 3.534845620667927,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037838929755519655,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21526299864053727,
      "num_tokens": 77632348.0,
      "step": 42075
    },
    {
      "entropy": 5.443536043167114,
      "epoch": 3.535265700483092,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037836269890404806,
      "loss": 4.956,
      "mean_token_accuracy": 0.2033406004309654,
      "num_tokens": 77641512.0,
      "step": 42080
    },
    {
      "entropy": 5.456309986114502,
      "epoch": 3.5356857802982566,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003783360984218827,
      "loss": 4.9451,
      "mean_token_accuracy": 0.20635688155889512,
      "num_tokens": 77650650.0,
      "step": 42085
    },
    {
      "entropy": 5.478555297851562,
      "epoch": 3.5361058601134214,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037830949610917155,
      "loss": 5.0612,
      "mean_token_accuracy": 0.20170481503009796,
      "num_tokens": 77659885.0,
      "step": 42090
    },
    {
      "entropy": 5.486775732040405,
      "epoch": 3.5365259399285867,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003782828919663858,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20485018491744994,
      "num_tokens": 77669281.0,
      "step": 42095
    },
    {
      "entropy": 5.442463970184326,
      "epoch": 3.5369460197437514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037825628599399687,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21412232518196106,
      "num_tokens": 77677568.0,
      "step": 42100
    },
    {
      "entropy": 5.44026141166687,
      "epoch": 3.5373660995589162,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037822967819247603,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2010442927479744,
      "num_tokens": 77686187.0,
      "step": 42105
    },
    {
      "entropy": 5.470219135284424,
      "epoch": 3.537786179374081,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037820306856229454,
      "loss": 4.9597,
      "mean_token_accuracy": 0.2104240447282791,
      "num_tokens": 77694826.0,
      "step": 42110
    },
    {
      "entropy": 5.464857721328736,
      "epoch": 3.538206259189246,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003781764571039238,
      "loss": 5.0387,
      "mean_token_accuracy": 0.2007108822464943,
      "num_tokens": 77703556.0,
      "step": 42115
    },
    {
      "entropy": 5.431994867324829,
      "epoch": 3.538626339004411,
      "grad_norm": 1.0,
      "learning_rate": 0.0003781498438178353,
      "loss": 4.963,
      "mean_token_accuracy": 0.20492938309907913,
      "num_tokens": 77713617.0,
      "step": 42120
    },
    {
      "entropy": 5.503875398635865,
      "epoch": 3.539046418819576,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003781232287045004,
      "loss": 4.9974,
      "mean_token_accuracy": 0.20485129505395888,
      "num_tokens": 77722254.0,
      "step": 42125
    },
    {
      "entropy": 5.418516159057617,
      "epoch": 3.5394664986347406,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003780966117643907,
      "loss": 4.9229,
      "mean_token_accuracy": 0.21605043560266496,
      "num_tokens": 77730802.0,
      "step": 42130
    },
    {
      "entropy": 5.451363801956177,
      "epoch": 3.5398865784499054,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003780699929979775,
      "loss": 4.9651,
      "mean_token_accuracy": 0.20662005692720414,
      "num_tokens": 77740426.0,
      "step": 42135
    },
    {
      "entropy": 5.388500308990478,
      "epoch": 3.54030665826507,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003780433724057325,
      "loss": 4.9533,
      "mean_token_accuracy": 0.210431706905365,
      "num_tokens": 77749559.0,
      "step": 42140
    },
    {
      "entropy": 5.429781675338745,
      "epoch": 3.5407267380802354,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003780167499881272,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20217789709568024,
      "num_tokens": 77759127.0,
      "step": 42145
    },
    {
      "entropy": 5.42663984298706,
      "epoch": 3.5411468178954,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037799012574563323,
      "loss": 5.0087,
      "mean_token_accuracy": 0.2014210119843483,
      "num_tokens": 77768246.0,
      "step": 42150
    },
    {
      "entropy": 5.4369124412536625,
      "epoch": 3.541566897710565,
      "grad_norm": 1.125,
      "learning_rate": 0.0003779634996787222,
      "loss": 4.9014,
      "mean_token_accuracy": 0.20690985321998595,
      "num_tokens": 77777217.0,
      "step": 42155
    },
    {
      "entropy": 5.476494359970093,
      "epoch": 3.5419869775257298,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003779368717878658,
      "loss": 4.9656,
      "mean_token_accuracy": 0.20179176777601243,
      "num_tokens": 77785610.0,
      "step": 42160
    },
    {
      "entropy": 5.405790519714356,
      "epoch": 3.5424070573408946,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037791024207353577,
      "loss": 4.9312,
      "mean_token_accuracy": 0.204990653693676,
      "num_tokens": 77795449.0,
      "step": 42165
    },
    {
      "entropy": 5.449801397323609,
      "epoch": 3.54282713715606,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00037788361053620376,
      "loss": 5.0447,
      "mean_token_accuracy": 0.20327091068029404,
      "num_tokens": 77805073.0,
      "step": 42170
    },
    {
      "entropy": 5.466721296310425,
      "epoch": 3.5432472169712246,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037785697717634156,
      "loss": 4.9743,
      "mean_token_accuracy": 0.19953031688928605,
      "num_tokens": 77813901.0,
      "step": 42175
    },
    {
      "entropy": 5.40221757888794,
      "epoch": 3.5436672967863894,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003778303419944209,
      "loss": 4.9362,
      "mean_token_accuracy": 0.2041926234960556,
      "num_tokens": 77823712.0,
      "step": 42180
    },
    {
      "entropy": 5.495376634597778,
      "epoch": 3.544087376601554,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003778037049909137,
      "loss": 4.9855,
      "mean_token_accuracy": 0.20774625539779662,
      "num_tokens": 77833425.0,
      "step": 42185
    },
    {
      "entropy": 5.407156467437744,
      "epoch": 3.544507456416719,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037777706616629194,
      "loss": 4.9504,
      "mean_token_accuracy": 0.20334288626909255,
      "num_tokens": 77842324.0,
      "step": 42190
    },
    {
      "entropy": 5.414376306533813,
      "epoch": 3.544927536231884,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037775042552102715,
      "loss": 4.9037,
      "mean_token_accuracy": 0.20537534207105637,
      "num_tokens": 77851973.0,
      "step": 42195
    },
    {
      "entropy": 5.502293825149536,
      "epoch": 3.545347616047049,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037772378305559167,
      "loss": 5.0287,
      "mean_token_accuracy": 0.19865405112504958,
      "num_tokens": 77860964.0,
      "step": 42200
    },
    {
      "entropy": 5.504457235336304,
      "epoch": 3.5457676958622137,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037769713877045723,
      "loss": 4.9852,
      "mean_token_accuracy": 0.2013077214360237,
      "num_tokens": 77869427.0,
      "step": 42205
    },
    {
      "entropy": 5.490885782241821,
      "epoch": 3.546187775677379,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037767049266609583,
      "loss": 4.9548,
      "mean_token_accuracy": 0.20819417983293534,
      "num_tokens": 77879059.0,
      "step": 42210
    },
    {
      "entropy": 5.408472299575806,
      "epoch": 3.5466078554925433,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037764384474297955,
      "loss": 4.9003,
      "mean_token_accuracy": 0.211522676050663,
      "num_tokens": 77888185.0,
      "step": 42215
    },
    {
      "entropy": 5.549646663665771,
      "epoch": 3.5470279353077085,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003776171950015804,
      "loss": 5.0716,
      "mean_token_accuracy": 0.20001590102910996,
      "num_tokens": 77897427.0,
      "step": 42220
    },
    {
      "entropy": 5.460158443450927,
      "epoch": 3.5474480151228733,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037759054344237044,
      "loss": 4.9141,
      "mean_token_accuracy": 0.20787637382745744,
      "num_tokens": 77905935.0,
      "step": 42225
    },
    {
      "entropy": 5.446090650558472,
      "epoch": 3.547868094938038,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003775638900658219,
      "loss": 4.9531,
      "mean_token_accuracy": 0.20266959816217422,
      "num_tokens": 77915527.0,
      "step": 42230
    },
    {
      "entropy": 5.539556884765625,
      "epoch": 3.5482881747532034,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037753723487240684,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20958302468061446,
      "num_tokens": 77925402.0,
      "step": 42235
    },
    {
      "entropy": 5.411311244964599,
      "epoch": 3.548708254568368,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003775105778625974,
      "loss": 4.9364,
      "mean_token_accuracy": 0.20700719058513642,
      "num_tokens": 77934859.0,
      "step": 42240
    },
    {
      "entropy": 5.478155040740967,
      "epoch": 3.549128334383533,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037748391903686594,
      "loss": 5.006,
      "mean_token_accuracy": 0.2076345846056938,
      "num_tokens": 77943353.0,
      "step": 42245
    },
    {
      "entropy": 5.364722013473511,
      "epoch": 3.5495484141986977,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003774572583956845,
      "loss": 5.0008,
      "mean_token_accuracy": 0.20590744763612748,
      "num_tokens": 77953708.0,
      "step": 42250
    },
    {
      "entropy": 5.45063362121582,
      "epoch": 3.5499684940138625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003774305959395256,
      "loss": 4.9691,
      "mean_token_accuracy": 0.21370414644479752,
      "num_tokens": 77961680.0,
      "step": 42255
    },
    {
      "entropy": 5.511509799957276,
      "epoch": 3.5503885738290277,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037740393166886137,
      "loss": 5.0221,
      "mean_token_accuracy": 0.20541918873786927,
      "num_tokens": 77971347.0,
      "step": 42260
    },
    {
      "entropy": 5.506279706954956,
      "epoch": 3.5508086536441925,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003773772655841643,
      "loss": 5.0506,
      "mean_token_accuracy": 0.1975969597697258,
      "num_tokens": 77981346.0,
      "step": 42265
    },
    {
      "entropy": 5.3847894191741945,
      "epoch": 3.5512287334593573,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037735059768590666,
      "loss": 4.9688,
      "mean_token_accuracy": 0.21334087997674941,
      "num_tokens": 77989984.0,
      "step": 42270
    },
    {
      "entropy": 5.413444519042969,
      "epoch": 3.551648813274522,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003773239279745609,
      "loss": 5.0152,
      "mean_token_accuracy": 0.20684606283903123,
      "num_tokens": 77998933.0,
      "step": 42275
    },
    {
      "entropy": 5.529790687561035,
      "epoch": 3.552068893089687,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037729725645059943,
      "loss": 5.0539,
      "mean_token_accuracy": 0.20274995863437653,
      "num_tokens": 78008894.0,
      "step": 42280
    },
    {
      "entropy": 5.532739353179932,
      "epoch": 3.552488972904852,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037727058311449476,
      "loss": 4.9679,
      "mean_token_accuracy": 0.21070341169834136,
      "num_tokens": 78018789.0,
      "step": 42285
    },
    {
      "entropy": 5.415424060821533,
      "epoch": 3.552909052720017,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037724390796671926,
      "loss": 5.0073,
      "mean_token_accuracy": 0.2003138020634651,
      "num_tokens": 78027141.0,
      "step": 42290
    },
    {
      "entropy": 5.400521850585937,
      "epoch": 3.5533291325351817,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037721723100774576,
      "loss": 4.9932,
      "mean_token_accuracy": 0.20449649393558503,
      "num_tokens": 78036361.0,
      "step": 42295
    },
    {
      "entropy": 5.528419303894043,
      "epoch": 3.5537492123503465,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003771905522380465,
      "loss": 5.0605,
      "mean_token_accuracy": 0.20221484899520875,
      "num_tokens": 78045564.0,
      "step": 42300
    },
    {
      "entropy": 5.490653085708618,
      "epoch": 3.5541692921655113,
      "grad_norm": 1.0,
      "learning_rate": 0.00037716387165809425,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20413567274808883,
      "num_tokens": 78054548.0,
      "step": 42305
    },
    {
      "entropy": 5.460630989074707,
      "epoch": 3.5545893719806765,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037713718926836166,
      "loss": 5.0203,
      "mean_token_accuracy": 0.19823253750801087,
      "num_tokens": 78064042.0,
      "step": 42310
    },
    {
      "entropy": 5.463803243637085,
      "epoch": 3.5550094517958413,
      "grad_norm": 1.125,
      "learning_rate": 0.00037711050506932127,
      "loss": 4.9165,
      "mean_token_accuracy": 0.21124443113803865,
      "num_tokens": 78072708.0,
      "step": 42315
    },
    {
      "entropy": 5.43557596206665,
      "epoch": 3.555429531611006,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003770838190614459,
      "loss": 4.9402,
      "mean_token_accuracy": 0.2073458343744278,
      "num_tokens": 78081626.0,
      "step": 42320
    },
    {
      "entropy": 5.4025876998901365,
      "epoch": 3.555849611426171,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003770571312452083,
      "loss": 4.9679,
      "mean_token_accuracy": 0.20897025763988494,
      "num_tokens": 78091194.0,
      "step": 42325
    },
    {
      "entropy": 5.462578582763672,
      "epoch": 3.5562696912413356,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037703044162108116,
      "loss": 5.129,
      "mean_token_accuracy": 0.19842705726623536,
      "num_tokens": 78100570.0,
      "step": 42330
    },
    {
      "entropy": 5.550949573516846,
      "epoch": 3.556689771056501,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003770037501895373,
      "loss": 5.0348,
      "mean_token_accuracy": 0.20377752631902696,
      "num_tokens": 78110255.0,
      "step": 42335
    },
    {
      "entropy": 5.503460311889649,
      "epoch": 3.5571098508716656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003769770569510495,
      "loss": 4.9448,
      "mean_token_accuracy": 0.21272388249635696,
      "num_tokens": 78119947.0,
      "step": 42340
    },
    {
      "entropy": 5.393247175216675,
      "epoch": 3.5575299306868304,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003769503619060906,
      "loss": 5.0179,
      "mean_token_accuracy": 0.197049018740654,
      "num_tokens": 78129717.0,
      "step": 42345
    },
    {
      "entropy": 5.457515716552734,
      "epoch": 3.557950010501995,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003769236650551336,
      "loss": 5.0236,
      "mean_token_accuracy": 0.20466381758451463,
      "num_tokens": 78137953.0,
      "step": 42350
    },
    {
      "entropy": 5.597595930099487,
      "epoch": 3.55837009031716,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003768969663986513,
      "loss": 5.1587,
      "mean_token_accuracy": 0.19487226009368896,
      "num_tokens": 78147951.0,
      "step": 42355
    },
    {
      "entropy": 5.561582231521607,
      "epoch": 3.5587901701323252,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037687026593711674,
      "loss": 5.0429,
      "mean_token_accuracy": 0.20614995956420898,
      "num_tokens": 78157634.0,
      "step": 42360
    },
    {
      "entropy": 5.469174814224243,
      "epoch": 3.55921024994749,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003768435636710029,
      "loss": 4.9515,
      "mean_token_accuracy": 0.2167370542883873,
      "num_tokens": 78166462.0,
      "step": 42365
    },
    {
      "entropy": 5.3871835231781,
      "epoch": 3.559630329762655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037681685960078273,
      "loss": 4.9873,
      "mean_token_accuracy": 0.20835974365472792,
      "num_tokens": 78175878.0,
      "step": 42370
    },
    {
      "entropy": 5.514828491210937,
      "epoch": 3.56005040957782,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037679015372692925,
      "loss": 5.0051,
      "mean_token_accuracy": 0.20439629703760148,
      "num_tokens": 78185885.0,
      "step": 42375
    },
    {
      "entropy": 5.513430833816528,
      "epoch": 3.560470489392985,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003767634460499157,
      "loss": 5.0617,
      "mean_token_accuracy": 0.1963864415884018,
      "num_tokens": 78194315.0,
      "step": 42380
    },
    {
      "entropy": 5.481532573699951,
      "epoch": 3.5608905692081496,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037673673657021516,
      "loss": 5.0336,
      "mean_token_accuracy": 0.20596559792757035,
      "num_tokens": 78203558.0,
      "step": 42385
    },
    {
      "entropy": 5.4681354522705075,
      "epoch": 3.5613106490233144,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037671002528830063,
      "loss": 4.9338,
      "mean_token_accuracy": 0.20868785232305526,
      "num_tokens": 78212435.0,
      "step": 42390
    },
    {
      "entropy": 5.449689626693726,
      "epoch": 3.561730728838479,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037668331220464544,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20109506994485854,
      "num_tokens": 78221546.0,
      "step": 42395
    },
    {
      "entropy": 5.382751417160034,
      "epoch": 3.5621508086536444,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037665659731972266,
      "loss": 4.9637,
      "mean_token_accuracy": 0.20806601792573928,
      "num_tokens": 78230802.0,
      "step": 42400
    },
    {
      "entropy": 5.4859215259552006,
      "epoch": 3.562570888468809,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003766298806340057,
      "loss": 5.0252,
      "mean_token_accuracy": 0.1981206178665161,
      "num_tokens": 78241238.0,
      "step": 42405
    },
    {
      "entropy": 5.500022029876709,
      "epoch": 3.562990968283974,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003766031621479676,
      "loss": 4.9943,
      "mean_token_accuracy": 0.19913587123155593,
      "num_tokens": 78250544.0,
      "step": 42410
    },
    {
      "entropy": 5.513527679443359,
      "epoch": 3.563411048099139,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003765764418620819,
      "loss": 5.0222,
      "mean_token_accuracy": 0.20162014216184615,
      "num_tokens": 78260790.0,
      "step": 42415
    },
    {
      "entropy": 5.447948265075683,
      "epoch": 3.5638311279143036,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037654971977682186,
      "loss": 4.9912,
      "mean_token_accuracy": 0.208669912815094,
      "num_tokens": 78270583.0,
      "step": 42420
    },
    {
      "entropy": 5.4648919105529785,
      "epoch": 3.564251207729469,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003765229958926608,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20112790465354918,
      "num_tokens": 78280180.0,
      "step": 42425
    },
    {
      "entropy": 5.478744029998779,
      "epoch": 3.5646712875446336,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003764962702100721,
      "loss": 5.0119,
      "mean_token_accuracy": 0.2034053847193718,
      "num_tokens": 78289614.0,
      "step": 42430
    },
    {
      "entropy": 5.4775653839111325,
      "epoch": 3.5650913673597984,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003764695427295292,
      "loss": 5.0372,
      "mean_token_accuracy": 0.19789644330739975,
      "num_tokens": 78298739.0,
      "step": 42435
    },
    {
      "entropy": 5.4754410743713375,
      "epoch": 3.565511447174963,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037644281345150564,
      "loss": 5.0301,
      "mean_token_accuracy": 0.2057444617152214,
      "num_tokens": 78308412.0,
      "step": 42440
    },
    {
      "entropy": 5.520783853530884,
      "epoch": 3.565931526990128,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003764160823764749,
      "loss": 4.9058,
      "mean_token_accuracy": 0.21384751200675964,
      "num_tokens": 78316616.0,
      "step": 42445
    },
    {
      "entropy": 5.417092657089233,
      "epoch": 3.566351606805293,
      "grad_norm": 1.125,
      "learning_rate": 0.0003763893495049105,
      "loss": 4.9691,
      "mean_token_accuracy": 0.21430428475141525,
      "num_tokens": 78326211.0,
      "step": 42450
    },
    {
      "entropy": 5.478300523757935,
      "epoch": 3.566771686620458,
      "grad_norm": 1.125,
      "learning_rate": 0.00037636261483728594,
      "loss": 5.0943,
      "mean_token_accuracy": 0.19717392921447754,
      "num_tokens": 78335125.0,
      "step": 42455
    },
    {
      "entropy": 5.416744232177734,
      "epoch": 3.5671917664356227,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003763358783740749,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2101605460047722,
      "num_tokens": 78343709.0,
      "step": 42460
    },
    {
      "entropy": 5.472873830795288,
      "epoch": 3.5676118462507875,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037630914011575094,
      "loss": 4.9774,
      "mean_token_accuracy": 0.2087823361158371,
      "num_tokens": 78352841.0,
      "step": 42465
    },
    {
      "entropy": 5.416181468963623,
      "epoch": 3.5680319260659523,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037628240006278767,
      "loss": 4.9709,
      "mean_token_accuracy": 0.2107744410634041,
      "num_tokens": 78362335.0,
      "step": 42470
    },
    {
      "entropy": 5.391063404083252,
      "epoch": 3.5684520058811176,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003762556582156589,
      "loss": 5.0081,
      "mean_token_accuracy": 0.20286656618118287,
      "num_tokens": 78371300.0,
      "step": 42475
    },
    {
      "entropy": 5.4218430519104,
      "epoch": 3.5688720856962823,
      "grad_norm": 1.125,
      "learning_rate": 0.00037622891457483827,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2087465614080429,
      "num_tokens": 78379995.0,
      "step": 42480
    },
    {
      "entropy": 5.398010921478272,
      "epoch": 3.569292165511447,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003762021691407995,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20335749983787538,
      "num_tokens": 78389586.0,
      "step": 42485
    },
    {
      "entropy": 5.444451189041137,
      "epoch": 3.569712245326612,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003761754219140165,
      "loss": 5.0105,
      "mean_token_accuracy": 0.20368486791849136,
      "num_tokens": 78399039.0,
      "step": 42490
    },
    {
      "entropy": 5.428288698196411,
      "epoch": 3.5701323251417767,
      "grad_norm": 1.0,
      "learning_rate": 0.00037614867289496287,
      "loss": 4.9318,
      "mean_token_accuracy": 0.2050960034132004,
      "num_tokens": 78408023.0,
      "step": 42495
    },
    {
      "entropy": 5.4477746963500975,
      "epoch": 3.570552404956942,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037612192208411266,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20476814210414887,
      "num_tokens": 78416542.0,
      "step": 42500
    },
    {
      "entropy": 5.516491889953613,
      "epoch": 3.5709724847721067,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003760951694819397,
      "loss": 4.9877,
      "mean_token_accuracy": 0.21144009977579117,
      "num_tokens": 78426138.0,
      "step": 42505
    },
    {
      "entropy": 5.395909833908081,
      "epoch": 3.5713925645872715,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003760684150889178,
      "loss": 4.8832,
      "mean_token_accuracy": 0.2135677307844162,
      "num_tokens": 78435116.0,
      "step": 42510
    },
    {
      "entropy": 5.550927066802979,
      "epoch": 3.5718126444024367,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037604165890552105,
      "loss": 5.1411,
      "mean_token_accuracy": 0.193967504799366,
      "num_tokens": 78444974.0,
      "step": 42515
    },
    {
      "entropy": 5.448810768127442,
      "epoch": 3.572232724217601,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037601490093222334,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21072245091199876,
      "num_tokens": 78453203.0,
      "step": 42520
    },
    {
      "entropy": 5.408879375457763,
      "epoch": 3.5726528040327663,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003759881411694986,
      "loss": 4.8587,
      "mean_token_accuracy": 0.21142291873693467,
      "num_tokens": 78462708.0,
      "step": 42525
    },
    {
      "entropy": 5.418021154403687,
      "epoch": 3.573072883847931,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037596137961782097,
      "loss": 4.9631,
      "mean_token_accuracy": 0.19835782200098037,
      "num_tokens": 78472036.0,
      "step": 42530
    },
    {
      "entropy": 5.513020038604736,
      "epoch": 3.573492963663096,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003759346162776645,
      "loss": 5.0206,
      "mean_token_accuracy": 0.20137048363685608,
      "num_tokens": 78481031.0,
      "step": 42535
    },
    {
      "entropy": 5.499466848373413,
      "epoch": 3.573913043478261,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003759078511495033,
      "loss": 5.0277,
      "mean_token_accuracy": 0.20623738765716554,
      "num_tokens": 78489721.0,
      "step": 42540
    },
    {
      "entropy": 5.445150756835938,
      "epoch": 3.574333123293426,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037588108423381144,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21246458739042282,
      "num_tokens": 78499479.0,
      "step": 42545
    },
    {
      "entropy": 5.376023817062378,
      "epoch": 3.5747532031085907,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037585431553106315,
      "loss": 4.909,
      "mean_token_accuracy": 0.2127481609582901,
      "num_tokens": 78508248.0,
      "step": 42550
    },
    {
      "entropy": 5.428846025466919,
      "epoch": 3.5751732829237555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003758275450417325,
      "loss": 4.9803,
      "mean_token_accuracy": 0.2054271876811981,
      "num_tokens": 78517415.0,
      "step": 42555
    },
    {
      "entropy": 5.543930864334106,
      "epoch": 3.5755933627389203,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037580077276629385,
      "loss": 5.1226,
      "mean_token_accuracy": 0.1907193258404732,
      "num_tokens": 78526989.0,
      "step": 42560
    },
    {
      "entropy": 5.430334186553955,
      "epoch": 3.5760134425540855,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003757739987052215,
      "loss": 4.9085,
      "mean_token_accuracy": 0.217022043466568,
      "num_tokens": 78536159.0,
      "step": 42565
    },
    {
      "entropy": 5.445780944824219,
      "epoch": 3.5764335223692503,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003757472228589896,
      "loss": 4.9461,
      "mean_token_accuracy": 0.20712506175041198,
      "num_tokens": 78545499.0,
      "step": 42570
    },
    {
      "entropy": 5.402340602874756,
      "epoch": 3.576853602184415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037572044522807247,
      "loss": 4.9094,
      "mean_token_accuracy": 0.20843905508518218,
      "num_tokens": 78554094.0,
      "step": 42575
    },
    {
      "entropy": 5.495287561416626,
      "epoch": 3.57727368199958,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003756936658129446,
      "loss": 5.0287,
      "mean_token_accuracy": 0.19908883273601533,
      "num_tokens": 78563642.0,
      "step": 42580
    },
    {
      "entropy": 5.553338861465454,
      "epoch": 3.5776937618147446,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037566688461408026,
      "loss": 5.1297,
      "mean_token_accuracy": 0.19212352186441423,
      "num_tokens": 78573475.0,
      "step": 42585
    },
    {
      "entropy": 5.456232166290283,
      "epoch": 3.57811384162991,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003756401016319539,
      "loss": 4.9465,
      "mean_token_accuracy": 0.20979465991258622,
      "num_tokens": 78582658.0,
      "step": 42590
    },
    {
      "entropy": 5.449310064315796,
      "epoch": 3.5785339214450747,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037561331686704,
      "loss": 4.8991,
      "mean_token_accuracy": 0.20618595629930497,
      "num_tokens": 78590964.0,
      "step": 42595
    },
    {
      "entropy": 5.39500207901001,
      "epoch": 3.5789540012602394,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003755865303198129,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21145949363708497,
      "num_tokens": 78599795.0,
      "step": 42600
    },
    {
      "entropy": 5.469981527328491,
      "epoch": 3.5793740810754042,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037555974199074725,
      "loss": 5.0634,
      "mean_token_accuracy": 0.20431071519851685,
      "num_tokens": 78609267.0,
      "step": 42605
    },
    {
      "entropy": 5.506409931182861,
      "epoch": 3.579794160890569,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003755329518803176,
      "loss": 5.0347,
      "mean_token_accuracy": 0.19673894643783568,
      "num_tokens": 78617704.0,
      "step": 42610
    },
    {
      "entropy": 5.479662752151489,
      "epoch": 3.5802142407057342,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037550615998899835,
      "loss": 4.9409,
      "mean_token_accuracy": 0.20217780768871307,
      "num_tokens": 78626461.0,
      "step": 42615
    },
    {
      "entropy": 5.47921929359436,
      "epoch": 3.580634320520899,
      "grad_norm": 1.125,
      "learning_rate": 0.00037547936631726435,
      "loss": 4.9287,
      "mean_token_accuracy": 0.20464848279953002,
      "num_tokens": 78635669.0,
      "step": 42620
    },
    {
      "entropy": 5.4613847732543945,
      "epoch": 3.581054400336064,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003754525708655901,
      "loss": 4.9485,
      "mean_token_accuracy": 0.20385267287492753,
      "num_tokens": 78645346.0,
      "step": 42625
    },
    {
      "entropy": 5.432756423950195,
      "epoch": 3.5814744801512286,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037542577363445026,
      "loss": 5.0426,
      "mean_token_accuracy": 0.20102910548448563,
      "num_tokens": 78654223.0,
      "step": 42630
    },
    {
      "entropy": 5.465823221206665,
      "epoch": 3.5818945599663934,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037539897462431955,
      "loss": 4.9993,
      "mean_token_accuracy": 0.20145900249481202,
      "num_tokens": 78662887.0,
      "step": 42635
    },
    {
      "entropy": 5.439722728729248,
      "epoch": 3.5823146397815586,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003753721738356727,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21544751524925232,
      "num_tokens": 78671997.0,
      "step": 42640
    },
    {
      "entropy": 5.451484346389771,
      "epoch": 3.5827347195967234,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003753453712689845,
      "loss": 4.9689,
      "mean_token_accuracy": 0.21120593547821045,
      "num_tokens": 78681089.0,
      "step": 42645
    },
    {
      "entropy": 5.513374900817871,
      "epoch": 3.583154799411888,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037531856692472974,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21758038252592088,
      "num_tokens": 78689605.0,
      "step": 42650
    },
    {
      "entropy": 5.387511587142944,
      "epoch": 3.583574879227053,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037529176080338317,
      "loss": 4.9032,
      "mean_token_accuracy": 0.21444493532180786,
      "num_tokens": 78699166.0,
      "step": 42655
    },
    {
      "entropy": 5.44193115234375,
      "epoch": 3.5839949590422178,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037526495290541965,
      "loss": 5.0169,
      "mean_token_accuracy": 0.20589222609996796,
      "num_tokens": 78708540.0,
      "step": 42660
    },
    {
      "entropy": 5.506751537322998,
      "epoch": 3.584415038857383,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037523814323131425,
      "loss": 4.9544,
      "mean_token_accuracy": 0.19784671813249588,
      "num_tokens": 78718138.0,
      "step": 42665
    },
    {
      "entropy": 5.562648677825928,
      "epoch": 3.584835118672548,
      "grad_norm": 1.125,
      "learning_rate": 0.00037521133178154166,
      "loss": 5.1316,
      "mean_token_accuracy": 0.19308862686157227,
      "num_tokens": 78727005.0,
      "step": 42670
    },
    {
      "entropy": 5.495909738540649,
      "epoch": 3.5852551984877126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037518451855657694,
      "loss": 5.0478,
      "mean_token_accuracy": 0.20326980501413344,
      "num_tokens": 78736888.0,
      "step": 42675
    },
    {
      "entropy": 5.537099742889405,
      "epoch": 3.585675278302878,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000375157703556895,
      "loss": 5.1298,
      "mean_token_accuracy": 0.19529138058423995,
      "num_tokens": 78747962.0,
      "step": 42680
    },
    {
      "entropy": 5.385582017898559,
      "epoch": 3.5860953581180426,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003751308867829711,
      "loss": 4.9362,
      "mean_token_accuracy": 0.20557620674371718,
      "num_tokens": 78756744.0,
      "step": 42685
    },
    {
      "entropy": 5.383376979827881,
      "epoch": 3.5865154379332074,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037510406823527997,
      "loss": 4.9194,
      "mean_token_accuracy": 0.21570584625005723,
      "num_tokens": 78765258.0,
      "step": 42690
    },
    {
      "entropy": 5.466443061828613,
      "epoch": 3.586935517748372,
      "grad_norm": 1.0,
      "learning_rate": 0.00037507724791429687,
      "loss": 4.9682,
      "mean_token_accuracy": 0.2075448215007782,
      "num_tokens": 78775823.0,
      "step": 42695
    },
    {
      "entropy": 5.419271802902221,
      "epoch": 3.587355597563537,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003750504258204968,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21130784302949907,
      "num_tokens": 78785415.0,
      "step": 42700
    },
    {
      "entropy": 5.426567029953003,
      "epoch": 3.587775677378702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037502360195435495,
      "loss": 5.006,
      "mean_token_accuracy": 0.2054946169257164,
      "num_tokens": 78795213.0,
      "step": 42705
    },
    {
      "entropy": 5.412803840637207,
      "epoch": 3.588195757193867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037499677631634655,
      "loss": 4.9052,
      "mean_token_accuracy": 0.21127236783504486,
      "num_tokens": 78803858.0,
      "step": 42710
    },
    {
      "entropy": 5.384954738616943,
      "epoch": 3.5886158370090318,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003749699489069467,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20310975909233092,
      "num_tokens": 78812192.0,
      "step": 42715
    },
    {
      "entropy": 5.46665325164795,
      "epoch": 3.5890359168241965,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037494311972663077,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20218470692634583,
      "num_tokens": 78821574.0,
      "step": 42720
    },
    {
      "entropy": 5.483032274246216,
      "epoch": 3.5894559966393613,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003749162887758738,
      "loss": 4.988,
      "mean_token_accuracy": 0.20188813507556916,
      "num_tokens": 78831247.0,
      "step": 42725
    },
    {
      "entropy": 5.428891897201538,
      "epoch": 3.5898760764545266,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003748894560551513,
      "loss": 4.9309,
      "mean_token_accuracy": 0.21573738604784012,
      "num_tokens": 78840484.0,
      "step": 42730
    },
    {
      "entropy": 5.378445672988891,
      "epoch": 3.5902961562696913,
      "grad_norm": 1.125,
      "learning_rate": 0.0003748626215649385,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20638231039047242,
      "num_tokens": 78848746.0,
      "step": 42735
    },
    {
      "entropy": 5.4697911739349365,
      "epoch": 3.590716236084856,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003748357853057108,
      "loss": 5.0245,
      "mean_token_accuracy": 0.20510830730199814,
      "num_tokens": 78858752.0,
      "step": 42740
    },
    {
      "entropy": 5.456854677200317,
      "epoch": 3.591136315900021,
      "grad_norm": 1.125,
      "learning_rate": 0.00037480894727794355,
      "loss": 4.9544,
      "mean_token_accuracy": 0.20185839980840684,
      "num_tokens": 78866996.0,
      "step": 42745
    },
    {
      "entropy": 5.443111276626587,
      "epoch": 3.5915563957151857,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037478210748211223,
      "loss": 4.9547,
      "mean_token_accuracy": 0.20327399522066117,
      "num_tokens": 78876415.0,
      "step": 42750
    },
    {
      "entropy": 5.514011001586914,
      "epoch": 3.591976475530351,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037475526591869223,
      "loss": 5.0709,
      "mean_token_accuracy": 0.19227631539106368,
      "num_tokens": 78886302.0,
      "step": 42755
    },
    {
      "entropy": 5.467429733276367,
      "epoch": 3.5923965553455157,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037472842258815913,
      "loss": 4.9664,
      "mean_token_accuracy": 0.20757054686546325,
      "num_tokens": 78894676.0,
      "step": 42760
    },
    {
      "entropy": 5.470418930053711,
      "epoch": 3.5928166351606805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003747015774909883,
      "loss": 5.0283,
      "mean_token_accuracy": 0.20459945350885392,
      "num_tokens": 78904607.0,
      "step": 42765
    },
    {
      "entropy": 5.419697952270508,
      "epoch": 3.5932367149758453,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003746747306276554,
      "loss": 4.9323,
      "mean_token_accuracy": 0.20574079304933549,
      "num_tokens": 78912942.0,
      "step": 42770
    },
    {
      "entropy": 5.389264678955078,
      "epoch": 3.59365679479101,
      "grad_norm": 1.109375,
      "learning_rate": 0.000374647881998636,
      "loss": 4.9268,
      "mean_token_accuracy": 0.2115115463733673,
      "num_tokens": 78921234.0,
      "step": 42775
    },
    {
      "entropy": 5.466720962524414,
      "epoch": 3.5940768746061753,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037462103160440557,
      "loss": 5.0462,
      "mean_token_accuracy": 0.19802819639444352,
      "num_tokens": 78931402.0,
      "step": 42780
    },
    {
      "entropy": 5.4647845268249515,
      "epoch": 3.59449695442134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037459417944543996,
      "loss": 4.9552,
      "mean_token_accuracy": 0.20760064125061034,
      "num_tokens": 78939537.0,
      "step": 42785
    },
    {
      "entropy": 5.55280499458313,
      "epoch": 3.594917034236505,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003745673255222147,
      "loss": 5.043,
      "mean_token_accuracy": 0.20486547499895097,
      "num_tokens": 78948363.0,
      "step": 42790
    },
    {
      "entropy": 5.471277284622192,
      "epoch": 3.5953371140516697,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037454046983520544,
      "loss": 5.0571,
      "mean_token_accuracy": 0.20150902271270751,
      "num_tokens": 78957355.0,
      "step": 42795
    },
    {
      "entropy": 5.400015783309937,
      "epoch": 3.5957571938668345,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037451361238488817,
      "loss": 5.0,
      "mean_token_accuracy": 0.20465621799230577,
      "num_tokens": 78966263.0,
      "step": 42800
    },
    {
      "entropy": 5.483988666534424,
      "epoch": 3.5961772736819997,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037448675317173846,
      "loss": 4.9943,
      "mean_token_accuracy": 0.20652904510498046,
      "num_tokens": 78975430.0,
      "step": 42805
    },
    {
      "entropy": 5.488620710372925,
      "epoch": 3.5965973534971645,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037445989219623215,
      "loss": 4.9145,
      "mean_token_accuracy": 0.20570864230394365,
      "num_tokens": 78985184.0,
      "step": 42810
    },
    {
      "entropy": 5.494660949707031,
      "epoch": 3.5970174333123293,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037443302945884506,
      "loss": 5.0869,
      "mean_token_accuracy": 0.19950158447027205,
      "num_tokens": 78994544.0,
      "step": 42815
    },
    {
      "entropy": 5.432946109771729,
      "epoch": 3.5974375131274945,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037440616496005303,
      "loss": 4.9356,
      "mean_token_accuracy": 0.20106886625289916,
      "num_tokens": 79003509.0,
      "step": 42820
    },
    {
      "entropy": 5.414330720901489,
      "epoch": 3.597857592942659,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000374379298700332,
      "loss": 4.9954,
      "mean_token_accuracy": 0.20584761798381807,
      "num_tokens": 79012279.0,
      "step": 42825
    },
    {
      "entropy": 5.465554714202881,
      "epoch": 3.598277672757824,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003743524306801578,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20596243739128112,
      "num_tokens": 79021550.0,
      "step": 42830
    },
    {
      "entropy": 5.487448644638062,
      "epoch": 3.598697752572989,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037432556090000657,
      "loss": 5.0071,
      "mean_token_accuracy": 0.20937809944152833,
      "num_tokens": 79030942.0,
      "step": 42835
    },
    {
      "entropy": 5.416200780868531,
      "epoch": 3.5991178323881536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003742986893603541,
      "loss": 4.7999,
      "mean_token_accuracy": 0.2206709563732147,
      "num_tokens": 79038788.0,
      "step": 42840
    },
    {
      "entropy": 5.462961196899414,
      "epoch": 3.599537912203319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003742718160616764,
      "loss": 5.0962,
      "mean_token_accuracy": 0.1979453146457672,
      "num_tokens": 79048904.0,
      "step": 42845
    },
    {
      "entropy": 5.3746253490448,
      "epoch": 3.5999579920184837,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003742449410044498,
      "loss": 4.8362,
      "mean_token_accuracy": 0.2167111411690712,
      "num_tokens": 79058157.0,
      "step": 42850
    },
    {
      "entropy": 5.487565517425537,
      "epoch": 3.6003780718336484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037421806418915,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2050819456577301,
      "num_tokens": 79067645.0,
      "step": 42855
    },
    {
      "entropy": 5.408398866653442,
      "epoch": 3.6007981516488132,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037419118561625333,
      "loss": 4.9054,
      "mean_token_accuracy": 0.2097861498594284,
      "num_tokens": 79076434.0,
      "step": 42860
    },
    {
      "entropy": 5.5407898902893065,
      "epoch": 3.601218231463978,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037416430528623587,
      "loss": 5.0706,
      "mean_token_accuracy": 0.194950233399868,
      "num_tokens": 79085231.0,
      "step": 42865
    },
    {
      "entropy": 5.411996984481812,
      "epoch": 3.6016383112791432,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003741374231995738,
      "loss": 4.9029,
      "mean_token_accuracy": 0.2118785113096237,
      "num_tokens": 79093976.0,
      "step": 42870
    },
    {
      "entropy": 5.400426387786865,
      "epoch": 3.602058391094308,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003741105393567434,
      "loss": 4.9696,
      "mean_token_accuracy": 0.20654178708791732,
      "num_tokens": 79103718.0,
      "step": 42875
    },
    {
      "entropy": 5.506963443756104,
      "epoch": 3.602478470909473,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003740836537582207,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20729449540376663,
      "num_tokens": 79112807.0,
      "step": 42880
    },
    {
      "entropy": 5.512052869796753,
      "epoch": 3.6028985507246376,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003740567664044822,
      "loss": 5.0898,
      "mean_token_accuracy": 0.20223396569490432,
      "num_tokens": 79121519.0,
      "step": 42885
    },
    {
      "entropy": 5.382924795150757,
      "epoch": 3.6033186305398024,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037402987729600413,
      "loss": 4.944,
      "mean_token_accuracy": 0.2111974060535431,
      "num_tokens": 79130221.0,
      "step": 42890
    },
    {
      "entropy": 5.416294193267822,
      "epoch": 3.6037387103549676,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037400298643326267,
      "loss": 4.89,
      "mean_token_accuracy": 0.21421585381031036,
      "num_tokens": 79139138.0,
      "step": 42895
    },
    {
      "entropy": 5.5245922088623045,
      "epoch": 3.6041587901701324,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003739760938167343,
      "loss": 4.9913,
      "mean_token_accuracy": 0.20613787770271302,
      "num_tokens": 79147826.0,
      "step": 42900
    },
    {
      "entropy": 5.543500804901123,
      "epoch": 3.604578869985297,
      "grad_norm": 0.96875,
      "learning_rate": 0.00037394919944689546,
      "loss": 5.1024,
      "mean_token_accuracy": 0.19744177907705307,
      "num_tokens": 79157526.0,
      "step": 42905
    },
    {
      "entropy": 5.569593334197998,
      "epoch": 3.604998949800462,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037392230332422246,
      "loss": 5.081,
      "mean_token_accuracy": 0.1987232893705368,
      "num_tokens": 79167591.0,
      "step": 42910
    },
    {
      "entropy": 5.4143767833709715,
      "epoch": 3.6054190296156268,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003738954054491918,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21068974882364272,
      "num_tokens": 79176883.0,
      "step": 42915
    },
    {
      "entropy": 5.511336469650269,
      "epoch": 3.605839109430792,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003738685058222799,
      "loss": 5.0777,
      "mean_token_accuracy": 0.19532744437456132,
      "num_tokens": 79186387.0,
      "step": 42920
    },
    {
      "entropy": 5.505622243881225,
      "epoch": 3.606259189245957,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003738416044439634,
      "loss": 5.0164,
      "mean_token_accuracy": 0.2067989245057106,
      "num_tokens": 79195342.0,
      "step": 42925
    },
    {
      "entropy": 5.4612620830535885,
      "epoch": 3.6066792690611216,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003738147013147188,
      "loss": 4.9828,
      "mean_token_accuracy": 0.2068938970565796,
      "num_tokens": 79204284.0,
      "step": 42930
    },
    {
      "entropy": 5.391713380813599,
      "epoch": 3.6070993488762864,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003737877964350225,
      "loss": 5.0126,
      "mean_token_accuracy": 0.20644478052854537,
      "num_tokens": 79213466.0,
      "step": 42935
    },
    {
      "entropy": 5.401545190811158,
      "epoch": 3.607519428691451,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037376088980535147,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21518438905477524,
      "num_tokens": 79222288.0,
      "step": 42940
    },
    {
      "entropy": 5.508369255065918,
      "epoch": 3.6079395085066164,
      "grad_norm": 1.1875,
      "learning_rate": 0.000373733981426182,
      "loss": 4.9582,
      "mean_token_accuracy": 0.2016034707427025,
      "num_tokens": 79231485.0,
      "step": 42945
    },
    {
      "entropy": 5.5276093006134035,
      "epoch": 3.608359588321781,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037370707129799087,
      "loss": 4.9855,
      "mean_token_accuracy": 0.20287186950445174,
      "num_tokens": 79240761.0,
      "step": 42950
    },
    {
      "entropy": 5.481391525268554,
      "epoch": 3.608779668136946,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037368015942125483,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20952762812376022,
      "num_tokens": 79250352.0,
      "step": 42955
    },
    {
      "entropy": 5.446796131134033,
      "epoch": 3.6091997479521107,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003736532457964506,
      "loss": 5.0103,
      "mean_token_accuracy": 0.203959558904171,
      "num_tokens": 79259531.0,
      "step": 42960
    },
    {
      "entropy": 5.52161340713501,
      "epoch": 3.6096198277672755,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037362633042405493,
      "loss": 5.0032,
      "mean_token_accuracy": 0.1984008088707924,
      "num_tokens": 79269671.0,
      "step": 42965
    },
    {
      "entropy": 5.511129474639892,
      "epoch": 3.6100399075824408,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00037359941330454457,
      "loss": 4.9936,
      "mean_token_accuracy": 0.200275519490242,
      "num_tokens": 79279898.0,
      "step": 42970
    },
    {
      "entropy": 5.437020349502563,
      "epoch": 3.6104599873976055,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037357249443839633,
      "loss": 4.9478,
      "mean_token_accuracy": 0.2062956362962723,
      "num_tokens": 79288545.0,
      "step": 42975
    },
    {
      "entropy": 5.469257116317749,
      "epoch": 3.6108800672127703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037354557382608715,
      "loss": 4.9899,
      "mean_token_accuracy": 0.20298028588294983,
      "num_tokens": 79297751.0,
      "step": 42980
    },
    {
      "entropy": 5.477612733840942,
      "epoch": 3.6113001470279356,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037351865146809386,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21034538894891738,
      "num_tokens": 79306439.0,
      "step": 42985
    },
    {
      "entropy": 5.42521104812622,
      "epoch": 3.6117202268431003,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037349172736489346,
      "loss": 4.994,
      "mean_token_accuracy": 0.20925893485546113,
      "num_tokens": 79315492.0,
      "step": 42990
    },
    {
      "entropy": 5.370365810394287,
      "epoch": 3.612140306658265,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003734648015169627,
      "loss": 4.8829,
      "mean_token_accuracy": 0.21726564764976503,
      "num_tokens": 79324494.0,
      "step": 42995
    },
    {
      "entropy": 5.456176233291626,
      "epoch": 3.61256038647343,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003734378739247788,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20104486495256424,
      "num_tokens": 79334112.0,
      "step": 43000
    },
    {
      "entropy": 5.422876834869385,
      "epoch": 3.6129804662885947,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003734109445888186,
      "loss": 4.9635,
      "mean_token_accuracy": 0.20451372414827346,
      "num_tokens": 79343266.0,
      "step": 43005
    },
    {
      "entropy": 5.4793336391448975,
      "epoch": 3.61340054610376,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037338401350955924,
      "loss": 4.9973,
      "mean_token_accuracy": 0.20426521450281143,
      "num_tokens": 79352347.0,
      "step": 43010
    },
    {
      "entropy": 5.470594882965088,
      "epoch": 3.6138206259189247,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003733570806874777,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21110848635435103,
      "num_tokens": 79361543.0,
      "step": 43015
    },
    {
      "entropy": 5.4380861759185795,
      "epoch": 3.6142407057340895,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003733301461230511,
      "loss": 4.9793,
      "mean_token_accuracy": 0.2080855295062065,
      "num_tokens": 79370221.0,
      "step": 43020
    },
    {
      "entropy": 5.442485809326172,
      "epoch": 3.6146607855492543,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003733032098167566,
      "loss": 5.0036,
      "mean_token_accuracy": 0.2043693631887436,
      "num_tokens": 79379788.0,
      "step": 43025
    },
    {
      "entropy": 5.514440250396729,
      "epoch": 3.615080865364419,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003732762717690714,
      "loss": 5.0674,
      "mean_token_accuracy": 0.20133667886257173,
      "num_tokens": 79389640.0,
      "step": 43030
    },
    {
      "entropy": 5.538638925552368,
      "epoch": 3.6155009451795843,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003732493319804726,
      "loss": 5.0065,
      "mean_token_accuracy": 0.20248111188411713,
      "num_tokens": 79398252.0,
      "step": 43035
    },
    {
      "entropy": 5.470621871948242,
      "epoch": 3.615921024994749,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003732223904514375,
      "loss": 5.0547,
      "mean_token_accuracy": 0.19824995547533036,
      "num_tokens": 79407828.0,
      "step": 43040
    },
    {
      "entropy": 5.37184100151062,
      "epoch": 3.616341104809914,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037319544718244336,
      "loss": 4.924,
      "mean_token_accuracy": 0.21892421841621398,
      "num_tokens": 79417948.0,
      "step": 43045
    },
    {
      "entropy": 5.479742431640625,
      "epoch": 3.6167611846250787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003731685021739674,
      "loss": 4.919,
      "mean_token_accuracy": 0.20915055572986602,
      "num_tokens": 79426644.0,
      "step": 43050
    },
    {
      "entropy": 5.474339962005615,
      "epoch": 3.6171812644402435,
      "grad_norm": 1.1875,
      "learning_rate": 0.000373141555426487,
      "loss": 4.9285,
      "mean_token_accuracy": 0.21254459470510484,
      "num_tokens": 79435541.0,
      "step": 43055
    },
    {
      "entropy": 5.471302175521851,
      "epoch": 3.6176013442554087,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003731146069404794,
      "loss": 5.0703,
      "mean_token_accuracy": 0.19624310433864595,
      "num_tokens": 79445737.0,
      "step": 43060
    },
    {
      "entropy": 5.382106018066406,
      "epoch": 3.6180214240705735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003730876567164221,
      "loss": 4.9138,
      "mean_token_accuracy": 0.20781280398368834,
      "num_tokens": 79454296.0,
      "step": 43065
    },
    {
      "entropy": 5.52345905303955,
      "epoch": 3.6184415038857383,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003730607047547925,
      "loss": 4.9255,
      "mean_token_accuracy": 0.2175481617450714,
      "num_tokens": 79462553.0,
      "step": 43070
    },
    {
      "entropy": 5.463076829910278,
      "epoch": 3.618861583700903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000373033751056068,
      "loss": 4.951,
      "mean_token_accuracy": 0.20615778714418412,
      "num_tokens": 79471323.0,
      "step": 43075
    },
    {
      "entropy": 5.478710031509399,
      "epoch": 3.619281663516068,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003730067956207261,
      "loss": 5.0714,
      "mean_token_accuracy": 0.20291665196418762,
      "num_tokens": 79479965.0,
      "step": 43080
    },
    {
      "entropy": 5.497102928161621,
      "epoch": 3.619701743331233,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037297983844924423,
      "loss": 4.99,
      "mean_token_accuracy": 0.20310520976781846,
      "num_tokens": 79488893.0,
      "step": 43085
    },
    {
      "entropy": 5.525641679763794,
      "epoch": 3.620121823146398,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003729528795421,
      "loss": 5.0537,
      "mean_token_accuracy": 0.20093268007040024,
      "num_tokens": 79497428.0,
      "step": 43090
    },
    {
      "entropy": 5.488678407669068,
      "epoch": 3.6205419029615626,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037292591889977097,
      "loss": 5.0409,
      "mean_token_accuracy": 0.20740772038698196,
      "num_tokens": 79506751.0,
      "step": 43095
    },
    {
      "entropy": 5.393803501129151,
      "epoch": 3.6209619827767274,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003728989565227347,
      "loss": 4.9821,
      "mean_token_accuracy": 0.205181884765625,
      "num_tokens": 79516559.0,
      "step": 43100
    },
    {
      "entropy": 5.5194621086120605,
      "epoch": 3.621382062591892,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003728719924114688,
      "loss": 5.0671,
      "mean_token_accuracy": 0.20011534839868544,
      "num_tokens": 79525188.0,
      "step": 43105
    },
    {
      "entropy": 5.480480718612671,
      "epoch": 3.6218021424070574,
      "grad_norm": 1.03125,
      "learning_rate": 0.000372845026566451,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20519713163375855,
      "num_tokens": 79534691.0,
      "step": 43110
    },
    {
      "entropy": 5.458273124694824,
      "epoch": 3.6222222222222222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003728180589881589,
      "loss": 5.0179,
      "mean_token_accuracy": 0.1982140064239502,
      "num_tokens": 79544372.0,
      "step": 43115
    },
    {
      "entropy": 5.425575256347656,
      "epoch": 3.622642302037387,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037279108967707027,
      "loss": 4.9556,
      "mean_token_accuracy": 0.20925856083631517,
      "num_tokens": 79554548.0,
      "step": 43120
    },
    {
      "entropy": 5.513405847549438,
      "epoch": 3.6230623818525522,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003727641186336628,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20791446417570114,
      "num_tokens": 79563831.0,
      "step": 43125
    },
    {
      "entropy": 5.459169530868531,
      "epoch": 3.6234824616677166,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037273714585841426,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21389023959636688,
      "num_tokens": 79573559.0,
      "step": 43130
    },
    {
      "entropy": 5.418894958496094,
      "epoch": 3.623902541482882,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037271017135180265,
      "loss": 4.9795,
      "mean_token_accuracy": 0.21484752297401427,
      "num_tokens": 79581903.0,
      "step": 43135
    },
    {
      "entropy": 5.429597043991089,
      "epoch": 3.6243226212980466,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003726831951143056,
      "loss": 4.9634,
      "mean_token_accuracy": 0.20999181717634202,
      "num_tokens": 79591023.0,
      "step": 43140
    },
    {
      "entropy": 5.486703205108642,
      "epoch": 3.6247427011132114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037265621714640103,
      "loss": 4.9155,
      "mean_token_accuracy": 0.2056356579065323,
      "num_tokens": 79600022.0,
      "step": 43145
    },
    {
      "entropy": 5.494822597503662,
      "epoch": 3.6251627809283766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003726292374485669,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21885409951210022,
      "num_tokens": 79608720.0,
      "step": 43150
    },
    {
      "entropy": 5.446161842346191,
      "epoch": 3.6255828607435414,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00037260225602128114,
      "loss": 4.969,
      "mean_token_accuracy": 0.21098690927028657,
      "num_tokens": 79618046.0,
      "step": 43155
    },
    {
      "entropy": 5.373832941055298,
      "epoch": 3.626002940558706,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037257527286502153,
      "loss": 4.9132,
      "mean_token_accuracy": 0.20801726579666138,
      "num_tokens": 79627347.0,
      "step": 43160
    },
    {
      "entropy": 5.452927732467652,
      "epoch": 3.626423020373871,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003725482879802663,
      "loss": 4.9864,
      "mean_token_accuracy": 0.2029452294111252,
      "num_tokens": 79635742.0,
      "step": 43165
    },
    {
      "entropy": 5.421720933914185,
      "epoch": 3.6268431001890358,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037252130136749336,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21079154163599015,
      "num_tokens": 79644795.0,
      "step": 43170
    },
    {
      "entropy": 5.4357280254364015,
      "epoch": 3.627263180004201,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003724943130271808,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20433695018291473,
      "num_tokens": 79653556.0,
      "step": 43175
    },
    {
      "entropy": 5.431610202789306,
      "epoch": 3.627683259819366,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037246732295980665,
      "loss": 4.9093,
      "mean_token_accuracy": 0.21042044460773468,
      "num_tokens": 79661958.0,
      "step": 43180
    },
    {
      "entropy": 5.385646629333496,
      "epoch": 3.6281033396345306,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003724403311658491,
      "loss": 4.9118,
      "mean_token_accuracy": 0.21061141937971115,
      "num_tokens": 79671596.0,
      "step": 43185
    },
    {
      "entropy": 5.550860261917114,
      "epoch": 3.6285234194496954,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00037241333764578626,
      "loss": 5.0453,
      "mean_token_accuracy": 0.20646344125270844,
      "num_tokens": 79680744.0,
      "step": 43190
    },
    {
      "entropy": 5.463724756240845,
      "epoch": 3.62894349926486,
      "grad_norm": 1.125,
      "learning_rate": 0.0003723863424000963,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21178326308727263,
      "num_tokens": 79690024.0,
      "step": 43195
    },
    {
      "entropy": 5.450302934646606,
      "epoch": 3.6293635790800254,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037235934542925736,
      "loss": 5.004,
      "mean_token_accuracy": 0.2008850246667862,
      "num_tokens": 79699176.0,
      "step": 43200
    },
    {
      "entropy": 5.454894733428955,
      "epoch": 3.62978365889519,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003723323467337479,
      "loss": 5.0237,
      "mean_token_accuracy": 0.2102269262075424,
      "num_tokens": 79707935.0,
      "step": 43205
    },
    {
      "entropy": 5.477397680282593,
      "epoch": 3.630203738710355,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037230534631404593,
      "loss": 5.0191,
      "mean_token_accuracy": 0.20088737308979035,
      "num_tokens": 79717423.0,
      "step": 43210
    },
    {
      "entropy": 5.562045478820801,
      "epoch": 3.6306238185255197,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037227834417062985,
      "loss": 5.0382,
      "mean_token_accuracy": 0.20667117983102798,
      "num_tokens": 79727416.0,
      "step": 43215
    },
    {
      "entropy": 5.503303956985474,
      "epoch": 3.6310438983406845,
      "grad_norm": 1.15625,
      "learning_rate": 0.000372251340303978,
      "loss": 4.9495,
      "mean_token_accuracy": 0.20527174770832063,
      "num_tokens": 79736110.0,
      "step": 43220
    },
    {
      "entropy": 5.515454816818237,
      "epoch": 3.6314639781558498,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003722243347145687,
      "loss": 4.9898,
      "mean_token_accuracy": 0.202990260720253,
      "num_tokens": 79746101.0,
      "step": 43225
    },
    {
      "entropy": 5.454017400741577,
      "epoch": 3.6318840579710145,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037219732740288047,
      "loss": 4.9648,
      "mean_token_accuracy": 0.20726645588874817,
      "num_tokens": 79754452.0,
      "step": 43230
    },
    {
      "entropy": 5.431678199768067,
      "epoch": 3.6323041377861793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037217031836939157,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21446488201618194,
      "num_tokens": 79762754.0,
      "step": 43235
    },
    {
      "entropy": 5.465512323379516,
      "epoch": 3.632724217601344,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037214330761458053,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20109750628471373,
      "num_tokens": 79772441.0,
      "step": 43240
    },
    {
      "entropy": 5.468526411056518,
      "epoch": 3.633144297416509,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037211629513892586,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20510171204805375,
      "num_tokens": 79781948.0,
      "step": 43245
    },
    {
      "entropy": 5.506252574920654,
      "epoch": 3.633564377231674,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037208928094290596,
      "loss": 5.0218,
      "mean_token_accuracy": 0.2079169273376465,
      "num_tokens": 79790581.0,
      "step": 43250
    },
    {
      "entropy": 5.533329963684082,
      "epoch": 3.633984457046839,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037206226502699947,
      "loss": 4.9923,
      "mean_token_accuracy": 0.2036940261721611,
      "num_tokens": 79799624.0,
      "step": 43255
    },
    {
      "entropy": 5.4490399837493895,
      "epoch": 3.6344045368620037,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037203524739168494,
      "loss": 5.0147,
      "mean_token_accuracy": 0.1996852546930313,
      "num_tokens": 79808996.0,
      "step": 43260
    },
    {
      "entropy": 5.398217010498047,
      "epoch": 3.6348246166771685,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003720082280374409,
      "loss": 4.9437,
      "mean_token_accuracy": 0.20986525267362593,
      "num_tokens": 79818744.0,
      "step": 43265
    },
    {
      "entropy": 5.527004814147949,
      "epoch": 3.6352446964923333,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037198120696474614,
      "loss": 5.0807,
      "mean_token_accuracy": 0.2001202866435051,
      "num_tokens": 79827946.0,
      "step": 43270
    },
    {
      "entropy": 5.521688413619995,
      "epoch": 3.6356647763074985,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003719541841740792,
      "loss": 5.0281,
      "mean_token_accuracy": 0.2070169672369957,
      "num_tokens": 79836585.0,
      "step": 43275
    },
    {
      "entropy": 5.427810573577881,
      "epoch": 3.6360848561226633,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003719271596659188,
      "loss": 4.939,
      "mean_token_accuracy": 0.21594007015228273,
      "num_tokens": 79845440.0,
      "step": 43280
    },
    {
      "entropy": 5.422017908096313,
      "epoch": 3.636504935937828,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037190013344074374,
      "loss": 4.949,
      "mean_token_accuracy": 0.2132036343216896,
      "num_tokens": 79854866.0,
      "step": 43285
    },
    {
      "entropy": 5.420138502120972,
      "epoch": 3.6369250157529933,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037187310549903263,
      "loss": 4.8671,
      "mean_token_accuracy": 0.22293345332145692,
      "num_tokens": 79864261.0,
      "step": 43290
    },
    {
      "entropy": 5.3965989112854,
      "epoch": 3.637345095568158,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003718460758412645,
      "loss": 4.9687,
      "mean_token_accuracy": 0.21118438243865967,
      "num_tokens": 79873106.0,
      "step": 43295
    },
    {
      "entropy": 5.462494945526123,
      "epoch": 3.637765175383323,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003718190444679178,
      "loss": 4.978,
      "mean_token_accuracy": 0.21088827252388,
      "num_tokens": 79881237.0,
      "step": 43300
    },
    {
      "entropy": 5.433235692977905,
      "epoch": 3.6381852551984877,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037179201137947166,
      "loss": 5.0104,
      "mean_token_accuracy": 0.20552735030651093,
      "num_tokens": 79891593.0,
      "step": 43305
    },
    {
      "entropy": 5.4784547805786135,
      "epoch": 3.6386053350136525,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003717649765764049,
      "loss": 5.0606,
      "mean_token_accuracy": 0.1997336506843567,
      "num_tokens": 79901769.0,
      "step": 43310
    },
    {
      "entropy": 5.5726964473724365,
      "epoch": 3.6390254148288177,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037173794005919636,
      "loss": 5.0393,
      "mean_token_accuracy": 0.2013656049966812,
      "num_tokens": 79910646.0,
      "step": 43315
    },
    {
      "entropy": 5.451874542236328,
      "epoch": 3.6394454946439825,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003717109018283251,
      "loss": 4.9442,
      "mean_token_accuracy": 0.21035551130771638,
      "num_tokens": 79920457.0,
      "step": 43320
    },
    {
      "entropy": 5.475772571563721,
      "epoch": 3.6398655744591473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037168386188426994,
      "loss": 5.0954,
      "mean_token_accuracy": 0.1949879914522171,
      "num_tokens": 79929635.0,
      "step": 43325
    },
    {
      "entropy": 5.551311492919922,
      "epoch": 3.640285654274312,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037165682022751,
      "loss": 5.0284,
      "mean_token_accuracy": 0.2068599671125412,
      "num_tokens": 79938883.0,
      "step": 43330
    },
    {
      "entropy": 5.4798883438110355,
      "epoch": 3.640705734089477,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003716297768585242,
      "loss": 4.9643,
      "mean_token_accuracy": 0.2014419212937355,
      "num_tokens": 79947897.0,
      "step": 43335
    },
    {
      "entropy": 5.476394844055176,
      "epoch": 3.641125813904642,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003716027317777917,
      "loss": 4.9342,
      "mean_token_accuracy": 0.20704191774129868,
      "num_tokens": 79957164.0,
      "step": 43340
    },
    {
      "entropy": 5.443697261810303,
      "epoch": 3.641545893719807,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003715756849857915,
      "loss": 4.9508,
      "mean_token_accuracy": 0.20858724415302277,
      "num_tokens": 79965179.0,
      "step": 43345
    },
    {
      "entropy": 5.509348726272583,
      "epoch": 3.6419659735349716,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037154863648300283,
      "loss": 5.0184,
      "mean_token_accuracy": 0.2083483800292015,
      "num_tokens": 79975457.0,
      "step": 43350
    },
    {
      "entropy": 5.426759910583496,
      "epoch": 3.6423860533501364,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003715215862699048,
      "loss": 4.9099,
      "mean_token_accuracy": 0.2115268364548683,
      "num_tokens": 79984812.0,
      "step": 43355
    },
    {
      "entropy": 5.426347875595093,
      "epoch": 3.642806133165301,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037149453434697654,
      "loss": 4.9038,
      "mean_token_accuracy": 0.21798973828554152,
      "num_tokens": 79993814.0,
      "step": 43360
    },
    {
      "entropy": 5.362350273132324,
      "epoch": 3.6432262129804664,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037146748071469727,
      "loss": 4.9226,
      "mean_token_accuracy": 0.20807077884674072,
      "num_tokens": 80002791.0,
      "step": 43365
    },
    {
      "entropy": 5.4551231384277346,
      "epoch": 3.6436462927956312,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003714404253735463,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21300409585237504,
      "num_tokens": 80011514.0,
      "step": 43370
    },
    {
      "entropy": 5.391065502166748,
      "epoch": 3.644066372610796,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003714133683240029,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21400960385799409,
      "num_tokens": 80020644.0,
      "step": 43375
    },
    {
      "entropy": 5.49539589881897,
      "epoch": 3.644486452425961,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003713863095665462,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20532139241695405,
      "num_tokens": 80029194.0,
      "step": 43380
    },
    {
      "entropy": 5.460585021972657,
      "epoch": 3.6449065322411256,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003713592491016557,
      "loss": 4.9819,
      "mean_token_accuracy": 0.20218820124864578,
      "num_tokens": 80038803.0,
      "step": 43385
    },
    {
      "entropy": 5.540037202835083,
      "epoch": 3.645326612056291,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037133218692981076,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20108207017183305,
      "num_tokens": 80047517.0,
      "step": 43390
    },
    {
      "entropy": 5.577211380004883,
      "epoch": 3.6457466918714556,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037130512305149065,
      "loss": 5.137,
      "mean_token_accuracy": 0.18895398527383805,
      "num_tokens": 80057530.0,
      "step": 43395
    },
    {
      "entropy": 5.50773401260376,
      "epoch": 3.6461667716866204,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003712780574671749,
      "loss": 5.0088,
      "mean_token_accuracy": 0.20435311198234557,
      "num_tokens": 80065891.0,
      "step": 43400
    },
    {
      "entropy": 5.483478736877442,
      "epoch": 3.646586851501785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000371250990177343,
      "loss": 5.021,
      "mean_token_accuracy": 0.20311090648174285,
      "num_tokens": 80074794.0,
      "step": 43405
    },
    {
      "entropy": 5.47559003829956,
      "epoch": 3.64700693131695,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003712239211824743,
      "loss": 4.9277,
      "mean_token_accuracy": 0.21168452650308608,
      "num_tokens": 80083935.0,
      "step": 43410
    },
    {
      "entropy": 5.411086702346802,
      "epoch": 3.647427011132115,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003711968504830484,
      "loss": 4.9739,
      "mean_token_accuracy": 0.2098028764128685,
      "num_tokens": 80092911.0,
      "step": 43415
    },
    {
      "entropy": 5.466084241867065,
      "epoch": 3.64784709094728,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003711697780795449,
      "loss": 4.9833,
      "mean_token_accuracy": 0.2020171359181404,
      "num_tokens": 80101806.0,
      "step": 43420
    },
    {
      "entropy": 5.477010917663574,
      "epoch": 3.6482671707624448,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003711427039724432,
      "loss": 4.97,
      "mean_token_accuracy": 0.21336399167776107,
      "num_tokens": 80109871.0,
      "step": 43425
    },
    {
      "entropy": 5.502029466629028,
      "epoch": 3.64868725057761,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037111562816222307,
      "loss": 4.9539,
      "mean_token_accuracy": 0.21492657363414763,
      "num_tokens": 80118636.0,
      "step": 43430
    },
    {
      "entropy": 5.4853034019470215,
      "epoch": 3.6491073303927744,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003710885506493641,
      "loss": 4.9635,
      "mean_token_accuracy": 0.2026304066181183,
      "num_tokens": 80128732.0,
      "step": 43435
    },
    {
      "entropy": 5.479530668258667,
      "epoch": 3.6495274102079396,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003710614714343459,
      "loss": 5.0349,
      "mean_token_accuracy": 0.2027772843837738,
      "num_tokens": 80138028.0,
      "step": 43440
    },
    {
      "entropy": 5.5303748607635494,
      "epoch": 3.6499474900231044,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003710343905176481,
      "loss": 5.0671,
      "mean_token_accuracy": 0.21250925809144974,
      "num_tokens": 80146975.0,
      "step": 43445
    },
    {
      "entropy": 5.465924453735352,
      "epoch": 3.650367569838269,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003710073078997507,
      "loss": 5.0403,
      "mean_token_accuracy": 0.20064905136823655,
      "num_tokens": 80155355.0,
      "step": 43450
    },
    {
      "entropy": 5.445980739593506,
      "epoch": 3.6507876496534344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003709802235811332,
      "loss": 4.9112,
      "mean_token_accuracy": 0.21700752675533294,
      "num_tokens": 80164029.0,
      "step": 43455
    },
    {
      "entropy": 5.452673816680909,
      "epoch": 3.651207729468599,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037095313756227546,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20553145408630372,
      "num_tokens": 80173288.0,
      "step": 43460
    },
    {
      "entropy": 5.513602542877197,
      "epoch": 3.651627809283764,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003709260498436573,
      "loss": 5.0428,
      "mean_token_accuracy": 0.19668288379907609,
      "num_tokens": 80182779.0,
      "step": 43465
    },
    {
      "entropy": 5.547500371932983,
      "epoch": 3.6520478890989287,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003708989604257585,
      "loss": 5.1015,
      "mean_token_accuracy": 0.19783726334571838,
      "num_tokens": 80192688.0,
      "step": 43470
    },
    {
      "entropy": 5.499807596206665,
      "epoch": 3.6524679689140935,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037087186930905903,
      "loss": 5.0495,
      "mean_token_accuracy": 0.1958880230784416,
      "num_tokens": 80202040.0,
      "step": 43475
    },
    {
      "entropy": 5.5254816055297855,
      "epoch": 3.6528880487292588,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003708447764940388,
      "loss": 5.0295,
      "mean_token_accuracy": 0.20113216936588288,
      "num_tokens": 80211762.0,
      "step": 43480
    },
    {
      "entropy": 5.4283853530883786,
      "epoch": 3.6533081285444235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037081768198117765,
      "loss": 4.9463,
      "mean_token_accuracy": 0.20804518312215806,
      "num_tokens": 80221157.0,
      "step": 43485
    },
    {
      "entropy": 5.495623445510864,
      "epoch": 3.6537282083595883,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003707905857709557,
      "loss": 5.0119,
      "mean_token_accuracy": 0.20514582544565202,
      "num_tokens": 80229748.0,
      "step": 43490
    },
    {
      "entropy": 5.527695178985596,
      "epoch": 3.654148288174753,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037076348786385275,
      "loss": 5.0104,
      "mean_token_accuracy": 0.2093670278787613,
      "num_tokens": 80238610.0,
      "step": 43495
    },
    {
      "entropy": 5.39469347000122,
      "epoch": 3.654568367989918,
      "grad_norm": 1.125,
      "learning_rate": 0.00037073638826034893,
      "loss": 4.952,
      "mean_token_accuracy": 0.21150965839624405,
      "num_tokens": 80247777.0,
      "step": 43500
    },
    {
      "entropy": 5.438847208023072,
      "epoch": 3.654988447805083,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037070928696092436,
      "loss": 4.9178,
      "mean_token_accuracy": 0.20963984876871108,
      "num_tokens": 80257666.0,
      "step": 43505
    },
    {
      "entropy": 5.4480881690979,
      "epoch": 3.655408527620248,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000370682183966059,
      "loss": 4.9327,
      "mean_token_accuracy": 0.20877716690301895,
      "num_tokens": 80266270.0,
      "step": 43510
    },
    {
      "entropy": 5.377590036392212,
      "epoch": 3.6558286074354127,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003706550792762331,
      "loss": 4.8371,
      "mean_token_accuracy": 0.21652536541223527,
      "num_tokens": 80275164.0,
      "step": 43515
    },
    {
      "entropy": 5.44387149810791,
      "epoch": 3.6562486872505775,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037062797289192663,
      "loss": 5.0042,
      "mean_token_accuracy": 0.19857332855463028,
      "num_tokens": 80284371.0,
      "step": 43520
    },
    {
      "entropy": 5.419881057739258,
      "epoch": 3.6566687670657423,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003706008648136199,
      "loss": 4.9435,
      "mean_token_accuracy": 0.20680492520332336,
      "num_tokens": 80293590.0,
      "step": 43525
    },
    {
      "entropy": 5.5214934825897215,
      "epoch": 3.6570888468809075,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003705737550417931,
      "loss": 5.0527,
      "mean_token_accuracy": 0.20412692427635193,
      "num_tokens": 80302001.0,
      "step": 43530
    },
    {
      "entropy": 5.491965818405151,
      "epoch": 3.6575089266960723,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037054664357692637,
      "loss": 5.019,
      "mean_token_accuracy": 0.20630670636892318,
      "num_tokens": 80311576.0,
      "step": 43535
    },
    {
      "entropy": 5.495136308670044,
      "epoch": 3.657929006511237,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003705195304195001,
      "loss": 4.9392,
      "mean_token_accuracy": 0.20368163883686066,
      "num_tokens": 80320410.0,
      "step": 43540
    },
    {
      "entropy": 5.50035662651062,
      "epoch": 3.658349086326402,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037049241556999453,
      "loss": 5.0377,
      "mean_token_accuracy": 0.2054095596075058,
      "num_tokens": 80330733.0,
      "step": 43545
    },
    {
      "entropy": 5.405598497390747,
      "epoch": 3.6587691661415667,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037046529902888997,
      "loss": 4.8925,
      "mean_token_accuracy": 0.21597631871700287,
      "num_tokens": 80339774.0,
      "step": 43550
    },
    {
      "entropy": 5.448401212692261,
      "epoch": 3.659189245956732,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003704381807966668,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20800121128559113,
      "num_tokens": 80348345.0,
      "step": 43555
    },
    {
      "entropy": 5.42056827545166,
      "epoch": 3.6596093257718967,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037041106087380536,
      "loss": 5.0353,
      "mean_token_accuracy": 0.19941325783729552,
      "num_tokens": 80357286.0,
      "step": 43560
    },
    {
      "entropy": 5.474565315246582,
      "epoch": 3.6600294055870615,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037038393926078623,
      "loss": 4.9648,
      "mean_token_accuracy": 0.20037793219089509,
      "num_tokens": 80367387.0,
      "step": 43565
    },
    {
      "entropy": 5.549097919464112,
      "epoch": 3.6604494854022267,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037035681595808963,
      "loss": 5.0484,
      "mean_token_accuracy": 0.2113879606127739,
      "num_tokens": 80377326.0,
      "step": 43570
    },
    {
      "entropy": 5.4779980182647705,
      "epoch": 3.660869565217391,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003703296909661962,
      "loss": 4.9942,
      "mean_token_accuracy": 0.2010681852698326,
      "num_tokens": 80386644.0,
      "step": 43575
    },
    {
      "entropy": 5.462807035446167,
      "epoch": 3.6612896450325563,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003703025642855863,
      "loss": 5.0409,
      "mean_token_accuracy": 0.19834935814142227,
      "num_tokens": 80396265.0,
      "step": 43580
    },
    {
      "entropy": 5.519050788879395,
      "epoch": 3.661709724847721,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003702754359167406,
      "loss": 5.0051,
      "mean_token_accuracy": 0.20182154327630997,
      "num_tokens": 80405031.0,
      "step": 43585
    },
    {
      "entropy": 5.469996356964112,
      "epoch": 3.662129804662886,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003702483058601396,
      "loss": 5.0568,
      "mean_token_accuracy": 0.20393055230379104,
      "num_tokens": 80414728.0,
      "step": 43590
    },
    {
      "entropy": 5.481759119033813,
      "epoch": 3.662549884478051,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037022117411626395,
      "loss": 5.057,
      "mean_token_accuracy": 0.2076714515686035,
      "num_tokens": 80424569.0,
      "step": 43595
    },
    {
      "entropy": 5.480972576141357,
      "epoch": 3.662969964293216,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003701940406855942,
      "loss": 4.9553,
      "mean_token_accuracy": 0.20780982524156572,
      "num_tokens": 80434240.0,
      "step": 43600
    },
    {
      "entropy": 5.436888742446899,
      "epoch": 3.6633900441083806,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037016690556861094,
      "loss": 5.0291,
      "mean_token_accuracy": 0.19818449765443802,
      "num_tokens": 80443083.0,
      "step": 43605
    },
    {
      "entropy": 5.4543462753295895,
      "epoch": 3.6638101239235454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037013976876579507,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20391096770763398,
      "num_tokens": 80452544.0,
      "step": 43610
    },
    {
      "entropy": 5.5873198986053465,
      "epoch": 3.6642302037387102,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003701126302776272,
      "loss": 5.0989,
      "mean_token_accuracy": 0.1991402730345726,
      "num_tokens": 80463026.0,
      "step": 43615
    },
    {
      "entropy": 5.528503656387329,
      "epoch": 3.6646502835538755,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037008549010458797,
      "loss": 4.9645,
      "mean_token_accuracy": 0.20475495457649232,
      "num_tokens": 80471964.0,
      "step": 43620
    },
    {
      "entropy": 5.45400972366333,
      "epoch": 3.6650703633690402,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00037005834824715826,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20552403330802918,
      "num_tokens": 80481810.0,
      "step": 43625
    },
    {
      "entropy": 5.458579158782959,
      "epoch": 3.665490443184205,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037003120470581897,
      "loss": 4.9853,
      "mean_token_accuracy": 0.20759291499853133,
      "num_tokens": 80491329.0,
      "step": 43630
    },
    {
      "entropy": 5.458689165115357,
      "epoch": 3.66591052299937,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003700040594810507,
      "loss": 5.0246,
      "mean_token_accuracy": 0.20503152757883072,
      "num_tokens": 80500717.0,
      "step": 43635
    },
    {
      "entropy": 5.428633832931519,
      "epoch": 3.6663306028145346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003699769125733345,
      "loss": 4.8665,
      "mean_token_accuracy": 0.21254194676876068,
      "num_tokens": 80510130.0,
      "step": 43640
    },
    {
      "entropy": 5.44412407875061,
      "epoch": 3.6667506826297,
      "grad_norm": 1.125,
      "learning_rate": 0.00036994976398315116,
      "loss": 5.0042,
      "mean_token_accuracy": 0.21375672072172164,
      "num_tokens": 80519852.0,
      "step": 43645
    },
    {
      "entropy": 5.478022289276123,
      "epoch": 3.6671707624448646,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003699226137109817,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2105574756860733,
      "num_tokens": 80528666.0,
      "step": 43650
    },
    {
      "entropy": 5.442919111251831,
      "epoch": 3.6675908422600294,
      "grad_norm": 1.046875,
      "learning_rate": 0.000369895461757307,
      "loss": 4.9601,
      "mean_token_accuracy": 0.2057576984167099,
      "num_tokens": 80538645.0,
      "step": 43655
    },
    {
      "entropy": 5.512905693054199,
      "epoch": 3.668010922075194,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000369868308122608,
      "loss": 5.1381,
      "mean_token_accuracy": 0.19102609157562256,
      "num_tokens": 80548068.0,
      "step": 43660
    },
    {
      "entropy": 5.419986867904663,
      "epoch": 3.668431001890359,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036984115280736587,
      "loss": 4.9871,
      "mean_token_accuracy": 0.2007022589445114,
      "num_tokens": 80556909.0,
      "step": 43665
    },
    {
      "entropy": 5.510308408737183,
      "epoch": 3.668851081705524,
      "grad_norm": 0.96875,
      "learning_rate": 0.00036981399581206137,
      "loss": 5.0275,
      "mean_token_accuracy": 0.2147708371281624,
      "num_tokens": 80566004.0,
      "step": 43670
    },
    {
      "entropy": 5.444189643859863,
      "epoch": 3.669271161520689,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003697868371371759,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20746919363737107,
      "num_tokens": 80574242.0,
      "step": 43675
    },
    {
      "entropy": 5.440886354446411,
      "epoch": 3.669691241335854,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036975967678319047,
      "loss": 4.9941,
      "mean_token_accuracy": 0.20866166651248932,
      "num_tokens": 80583302.0,
      "step": 43680
    },
    {
      "entropy": 5.4117532730102536,
      "epoch": 3.6701113211510186,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036973251475058607,
      "loss": 4.9236,
      "mean_token_accuracy": 0.2090457186102867,
      "num_tokens": 80594488.0,
      "step": 43685
    },
    {
      "entropy": 5.46555118560791,
      "epoch": 3.6705314009661834,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003697053510398439,
      "loss": 4.9734,
      "mean_token_accuracy": 0.2060657948255539,
      "num_tokens": 80604752.0,
      "step": 43690
    },
    {
      "entropy": 5.456528425216675,
      "epoch": 3.6709514807813486,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036967818565144535,
      "loss": 4.9642,
      "mean_token_accuracy": 0.2027345269918442,
      "num_tokens": 80613424.0,
      "step": 43695
    },
    {
      "entropy": 5.452835655212402,
      "epoch": 3.6713715605965134,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036965101858587137,
      "loss": 5.0214,
      "mean_token_accuracy": 0.20197518914937973,
      "num_tokens": 80622699.0,
      "step": 43700
    },
    {
      "entropy": 5.486165237426758,
      "epoch": 3.671791640411678,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003696238498436034,
      "loss": 5.0519,
      "mean_token_accuracy": 0.19796801805496217,
      "num_tokens": 80632749.0,
      "step": 43705
    },
    {
      "entropy": 5.48127851486206,
      "epoch": 3.672211720226843,
      "grad_norm": 0.96875,
      "learning_rate": 0.00036959667942512264,
      "loss": 5.0558,
      "mean_token_accuracy": 0.20016140043735503,
      "num_tokens": 80642886.0,
      "step": 43710
    },
    {
      "entropy": 5.4555220127105715,
      "epoch": 3.6726318000420077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036956950733091045,
      "loss": 4.9845,
      "mean_token_accuracy": 0.209897281229496,
      "num_tokens": 80652102.0,
      "step": 43715
    },
    {
      "entropy": 5.42604250907898,
      "epoch": 3.673051879857173,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000369542333561448,
      "loss": 4.9907,
      "mean_token_accuracy": 0.20527718514204024,
      "num_tokens": 80661120.0,
      "step": 43720
    },
    {
      "entropy": 5.496898984909057,
      "epoch": 3.6734719596723377,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003695151581172169,
      "loss": 5.0428,
      "mean_token_accuracy": 0.2117182731628418,
      "num_tokens": 80670813.0,
      "step": 43725
    },
    {
      "entropy": 5.485071611404419,
      "epoch": 3.6738920394875025,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003694879809986984,
      "loss": 4.996,
      "mean_token_accuracy": 0.20881824642419816,
      "num_tokens": 80680558.0,
      "step": 43730
    },
    {
      "entropy": 5.461761474609375,
      "epoch": 3.6743121193026678,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003694608022063741,
      "loss": 4.9639,
      "mean_token_accuracy": 0.2028776541352272,
      "num_tokens": 80690224.0,
      "step": 43735
    },
    {
      "entropy": 5.472381639480591,
      "epoch": 3.6747321991178326,
      "grad_norm": 0.96875,
      "learning_rate": 0.00036943362174072516,
      "loss": 5.044,
      "mean_token_accuracy": 0.2017589509487152,
      "num_tokens": 80700114.0,
      "step": 43740
    },
    {
      "entropy": 5.474211645126343,
      "epoch": 3.6751522789329973,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003694064396022333,
      "loss": 4.8644,
      "mean_token_accuracy": 0.21483421176671982,
      "num_tokens": 80708769.0,
      "step": 43745
    },
    {
      "entropy": 5.371218538284301,
      "epoch": 3.675572358748162,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036937925579137997,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2104051426053047,
      "num_tokens": 80717850.0,
      "step": 43750
    },
    {
      "entropy": 5.436564970016479,
      "epoch": 3.675992438563327,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003693520703086467,
      "loss": 4.9633,
      "mean_token_accuracy": 0.20016606003046036,
      "num_tokens": 80728018.0,
      "step": 43755
    },
    {
      "entropy": 5.416923093795776,
      "epoch": 3.676412518378492,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036932488315451515,
      "loss": 5.0038,
      "mean_token_accuracy": 0.2056306853890419,
      "num_tokens": 80737253.0,
      "step": 43760
    },
    {
      "entropy": 5.421985483169555,
      "epoch": 3.676832598193657,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003692976943294668,
      "loss": 4.9483,
      "mean_token_accuracy": 0.20120543241500854,
      "num_tokens": 80745971.0,
      "step": 43765
    },
    {
      "entropy": 5.398173856735229,
      "epoch": 3.6772526780088217,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003692705038339834,
      "loss": 4.8869,
      "mean_token_accuracy": 0.2099319189786911,
      "num_tokens": 80756013.0,
      "step": 43770
    },
    {
      "entropy": 5.469110441207886,
      "epoch": 3.6776727578239865,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003692433116685466,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20221807956695556,
      "num_tokens": 80765283.0,
      "step": 43775
    },
    {
      "entropy": 5.481357717514038,
      "epoch": 3.6780928376391513,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036921611783363793,
      "loss": 4.9757,
      "mean_token_accuracy": 0.21359534412622452,
      "num_tokens": 80774829.0,
      "step": 43780
    },
    {
      "entropy": 5.41036057472229,
      "epoch": 3.6785129174543165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036918892232973933,
      "loss": 4.951,
      "mean_token_accuracy": 0.2033068746328354,
      "num_tokens": 80783585.0,
      "step": 43785
    },
    {
      "entropy": 5.427263784408569,
      "epoch": 3.6789329972694813,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036916172515733247,
      "loss": 4.9543,
      "mean_token_accuracy": 0.20945417433977126,
      "num_tokens": 80792281.0,
      "step": 43790
    },
    {
      "entropy": 5.533624792098999,
      "epoch": 3.679353077084646,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036913452631689915,
      "loss": 5.0469,
      "mean_token_accuracy": 0.19927322268486022,
      "num_tokens": 80802191.0,
      "step": 43795
    },
    {
      "entropy": 5.499765253067016,
      "epoch": 3.679773156899811,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036910732580892116,
      "loss": 4.9884,
      "mean_token_accuracy": 0.20759883970022203,
      "num_tokens": 80811070.0,
      "step": 43800
    },
    {
      "entropy": 5.454195022583008,
      "epoch": 3.6801932367149757,
      "grad_norm": 1.125,
      "learning_rate": 0.00036908012363388035,
      "loss": 4.9872,
      "mean_token_accuracy": 0.20628122091293336,
      "num_tokens": 80820092.0,
      "step": 43805
    },
    {
      "entropy": 5.373696947097779,
      "epoch": 3.680613316530141,
      "grad_norm": 2.90625,
      "learning_rate": 0.00036905291979225856,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21046042144298555,
      "num_tokens": 80830139.0,
      "step": 43810
    },
    {
      "entropy": 5.4933404445648195,
      "epoch": 3.6810333963453057,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036902571428453775,
      "loss": 5.0335,
      "mean_token_accuracy": 0.1949485644698143,
      "num_tokens": 80839478.0,
      "step": 43815
    },
    {
      "entropy": 5.385309934616089,
      "epoch": 3.6814534761604705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003689985071111998,
      "loss": 4.9923,
      "mean_token_accuracy": 0.2058923900127411,
      "num_tokens": 80848850.0,
      "step": 43820
    },
    {
      "entropy": 5.476764822006226,
      "epoch": 3.6818735559756353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003689712982727267,
      "loss": 4.9858,
      "mean_token_accuracy": 0.2009149447083473,
      "num_tokens": 80858806.0,
      "step": 43825
    },
    {
      "entropy": 5.6159055709838865,
      "epoch": 3.6822936357908,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003689440877696005,
      "loss": 5.0726,
      "mean_token_accuracy": 0.2056381434202194,
      "num_tokens": 80867887.0,
      "step": 43830
    },
    {
      "entropy": 5.523929119110107,
      "epoch": 3.6827137156059653,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003689168756023032,
      "loss": 5.071,
      "mean_token_accuracy": 0.20336553752422332,
      "num_tokens": 80877782.0,
      "step": 43835
    },
    {
      "entropy": 5.56558141708374,
      "epoch": 3.68313379542113,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003688896617713167,
      "loss": 5.1099,
      "mean_token_accuracy": 0.19950517565011977,
      "num_tokens": 80887615.0,
      "step": 43840
    },
    {
      "entropy": 5.52865777015686,
      "epoch": 3.683553875236295,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036886244627712324,
      "loss": 5.0081,
      "mean_token_accuracy": 0.20601899921894073,
      "num_tokens": 80896537.0,
      "step": 43845
    },
    {
      "entropy": 5.475896501541138,
      "epoch": 3.6839739550514596,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003688352291202048,
      "loss": 5.0305,
      "mean_token_accuracy": 0.19881714433431624,
      "num_tokens": 80905004.0,
      "step": 43850
    },
    {
      "entropy": 5.402507162094116,
      "epoch": 3.6843940348666244,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036880801030104364,
      "loss": 4.9356,
      "mean_token_accuracy": 0.20869410485029222,
      "num_tokens": 80914190.0,
      "step": 43855
    },
    {
      "entropy": 5.44474196434021,
      "epoch": 3.6848141146817897,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000368780789820122,
      "loss": 4.9898,
      "mean_token_accuracy": 0.2028186872601509,
      "num_tokens": 80923643.0,
      "step": 43860
    },
    {
      "entropy": 5.545326375961304,
      "epoch": 3.6852341944969544,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003687535676779218,
      "loss": 5.097,
      "mean_token_accuracy": 0.20351426899433137,
      "num_tokens": 80933283.0,
      "step": 43865
    },
    {
      "entropy": 5.5134045600891115,
      "epoch": 3.6856542743121192,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003687263438749255,
      "loss": 4.9673,
      "mean_token_accuracy": 0.2033502861857414,
      "num_tokens": 80941937.0,
      "step": 43870
    },
    {
      "entropy": 5.507037210464477,
      "epoch": 3.6860743541272845,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003686991184116152,
      "loss": 5.0383,
      "mean_token_accuracy": 0.20919442921876907,
      "num_tokens": 80951089.0,
      "step": 43875
    },
    {
      "entropy": 5.395397472381592,
      "epoch": 3.686494433942449,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036867189128847336,
      "loss": 4.9552,
      "mean_token_accuracy": 0.19711624830961227,
      "num_tokens": 80960680.0,
      "step": 43880
    },
    {
      "entropy": 5.457663297653198,
      "epoch": 3.686914513757614,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036864466250598215,
      "loss": 4.9655,
      "mean_token_accuracy": 0.20683169215917588,
      "num_tokens": 80970232.0,
      "step": 43885
    },
    {
      "entropy": 5.512489318847656,
      "epoch": 3.687334593572779,
      "grad_norm": 1.015625,
      "learning_rate": 0.000368617432064624,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20018835216760636,
      "num_tokens": 80979909.0,
      "step": 43890
    },
    {
      "entropy": 5.471127653121949,
      "epoch": 3.6877546733879436,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003685901999648812,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20539110898971558,
      "num_tokens": 80989469.0,
      "step": 43895
    },
    {
      "entropy": 5.561388111114502,
      "epoch": 3.688174753203109,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036856296620723625,
      "loss": 5.0339,
      "mean_token_accuracy": 0.20450773388147353,
      "num_tokens": 80998384.0,
      "step": 43900
    },
    {
      "entropy": 5.52495436668396,
      "epoch": 3.6885948330182736,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003685357307921715,
      "loss": 4.9899,
      "mean_token_accuracy": 0.20111576467752457,
      "num_tokens": 81008045.0,
      "step": 43905
    },
    {
      "entropy": 5.460327005386352,
      "epoch": 3.6890149128334384,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036850849372016943,
      "loss": 4.9623,
      "mean_token_accuracy": 0.20933777391910552,
      "num_tokens": 81017416.0,
      "step": 43910
    },
    {
      "entropy": 5.445544004440308,
      "epoch": 3.689434992648603,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003684812549917125,
      "loss": 4.9624,
      "mean_token_accuracy": 0.20748942345380783,
      "num_tokens": 81026286.0,
      "step": 43915
    },
    {
      "entropy": 5.497039890289306,
      "epoch": 3.689855072463768,
      "grad_norm": 1.125,
      "learning_rate": 0.0003684540146072834,
      "loss": 5.0205,
      "mean_token_accuracy": 0.20447393357753754,
      "num_tokens": 81035460.0,
      "step": 43920
    },
    {
      "entropy": 5.436031818389893,
      "epoch": 3.690275152278933,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036842677256736445,
      "loss": 4.8475,
      "mean_token_accuracy": 0.2134431317448616,
      "num_tokens": 81044087.0,
      "step": 43925
    },
    {
      "entropy": 5.360692930221558,
      "epoch": 3.690695232094098,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036839952887243846,
      "loss": 4.8945,
      "mean_token_accuracy": 0.21340311616659163,
      "num_tokens": 81052430.0,
      "step": 43930
    },
    {
      "entropy": 5.452909660339356,
      "epoch": 3.691115311909263,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003683722835229878,
      "loss": 5.0256,
      "mean_token_accuracy": 0.1959852010011673,
      "num_tokens": 81060900.0,
      "step": 43935
    },
    {
      "entropy": 5.4697469711303714,
      "epoch": 3.6915353917244276,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003683450365194951,
      "loss": 4.8402,
      "mean_token_accuracy": 0.21621792763471603,
      "num_tokens": 81069748.0,
      "step": 43940
    },
    {
      "entropy": 5.446500062942505,
      "epoch": 3.6919554715395924,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003683177878624433,
      "loss": 4.9442,
      "mean_token_accuracy": 0.20803983360528946,
      "num_tokens": 81078844.0,
      "step": 43945
    },
    {
      "entropy": 5.402380561828613,
      "epoch": 3.6923755513547576,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036829053755231496,
      "loss": 4.9289,
      "mean_token_accuracy": 0.20973109006881713,
      "num_tokens": 81087306.0,
      "step": 43950
    },
    {
      "entropy": 5.474700689315796,
      "epoch": 3.6927956311699224,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036826328558959274,
      "loss": 5.0213,
      "mean_token_accuracy": 0.20157161504030227,
      "num_tokens": 81096214.0,
      "step": 43955
    },
    {
      "entropy": 5.472966527938842,
      "epoch": 3.693215710985087,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036823603197475946,
      "loss": 4.9659,
      "mean_token_accuracy": 0.20823789536952972,
      "num_tokens": 81104947.0,
      "step": 43960
    },
    {
      "entropy": 5.5297290802001955,
      "epoch": 3.693635790800252,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036820877670829785,
      "loss": 5.024,
      "mean_token_accuracy": 0.20242595821619033,
      "num_tokens": 81114445.0,
      "step": 43965
    },
    {
      "entropy": 5.486129188537598,
      "epoch": 3.6940558706154167,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036818151979069077,
      "loss": 5.0288,
      "mean_token_accuracy": 0.2021821454167366,
      "num_tokens": 81123144.0,
      "step": 43970
    },
    {
      "entropy": 5.440105533599853,
      "epoch": 3.694475950430582,
      "grad_norm": 1.0,
      "learning_rate": 0.00036815426122242107,
      "loss": 4.9272,
      "mean_token_accuracy": 0.2076609894633293,
      "num_tokens": 81133779.0,
      "step": 43975
    },
    {
      "entropy": 5.472083520889282,
      "epoch": 3.6948960302457468,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036812700100397164,
      "loss": 4.9433,
      "mean_token_accuracy": 0.2088090479373932,
      "num_tokens": 81142259.0,
      "step": 43980
    },
    {
      "entropy": 5.519091701507568,
      "epoch": 3.6953161100609115,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003680997391358252,
      "loss": 5.0754,
      "mean_token_accuracy": 0.19978614747524262,
      "num_tokens": 81151748.0,
      "step": 43985
    },
    {
      "entropy": 5.463827085494995,
      "epoch": 3.6957361898760763,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036807247561846496,
      "loss": 4.9302,
      "mean_token_accuracy": 0.2076859340071678,
      "num_tokens": 81161260.0,
      "step": 43990
    },
    {
      "entropy": 5.372696113586426,
      "epoch": 3.696156269691241,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003680452104523736,
      "loss": 4.9147,
      "mean_token_accuracy": 0.2085743084549904,
      "num_tokens": 81170076.0,
      "step": 43995
    },
    {
      "entropy": 5.483768892288208,
      "epoch": 3.6965763495064063,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036801794363803435,
      "loss": 5.0176,
      "mean_token_accuracy": 0.20122747868299484,
      "num_tokens": 81179366.0,
      "step": 44000
    },
    {
      "entropy": 5.494409275054932,
      "epoch": 3.696996429321571,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003679906751759301,
      "loss": 5.0207,
      "mean_token_accuracy": 0.2032885193824768,
      "num_tokens": 81188572.0,
      "step": 44005
    },
    {
      "entropy": 5.575102949142456,
      "epoch": 3.697416509136736,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003679634050665439,
      "loss": 5.0777,
      "mean_token_accuracy": 0.20721690505743026,
      "num_tokens": 81198387.0,
      "step": 44010
    },
    {
      "entropy": 5.390927124023437,
      "epoch": 3.6978365889519007,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003679361333103589,
      "loss": 4.8641,
      "mean_token_accuracy": 0.2113644525408745,
      "num_tokens": 81206757.0,
      "step": 44015
    },
    {
      "entropy": 5.48366527557373,
      "epoch": 3.6982566687670655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036790885990785804,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20118856728076934,
      "num_tokens": 81216210.0,
      "step": 44020
    },
    {
      "entropy": 5.520912551879883,
      "epoch": 3.6986767485822307,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003678815848595247,
      "loss": 5.0557,
      "mean_token_accuracy": 0.20090529322624207,
      "num_tokens": 81224891.0,
      "step": 44025
    },
    {
      "entropy": 5.4242040634155275,
      "epoch": 3.6990968283973955,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003678543081658419,
      "loss": 4.9612,
      "mean_token_accuracy": 0.2073383703827858,
      "num_tokens": 81234569.0,
      "step": 44030
    },
    {
      "entropy": 5.501741027832031,
      "epoch": 3.6995169082125603,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003678270298272928,
      "loss": 4.9997,
      "mean_token_accuracy": 0.19616715461015702,
      "num_tokens": 81243998.0,
      "step": 44035
    },
    {
      "entropy": 5.4839249610900875,
      "epoch": 3.6999369880277255,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003677997498443607,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20485631823539735,
      "num_tokens": 81252468.0,
      "step": 44040
    },
    {
      "entropy": 5.470072889328003,
      "epoch": 3.7003570678428903,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036777246821752876,
      "loss": 5.005,
      "mean_token_accuracy": 0.20318331718444824,
      "num_tokens": 81263270.0,
      "step": 44045
    },
    {
      "entropy": 5.464685678482056,
      "epoch": 3.700777147658055,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036774518494728036,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20916568636894226,
      "num_tokens": 81273028.0,
      "step": 44050
    },
    {
      "entropy": 5.479925584793091,
      "epoch": 3.70119722747322,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003677179000340988,
      "loss": 5.001,
      "mean_token_accuracy": 0.20448608845472335,
      "num_tokens": 81281746.0,
      "step": 44055
    },
    {
      "entropy": 5.587298488616943,
      "epoch": 3.7016173072883847,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003676906134784673,
      "loss": 5.1014,
      "mean_token_accuracy": 0.20773258358240126,
      "num_tokens": 81291652.0,
      "step": 44060
    },
    {
      "entropy": 5.471157646179199,
      "epoch": 3.70203738710355,
      "grad_norm": 1.125,
      "learning_rate": 0.0003676633252808693,
      "loss": 4.958,
      "mean_token_accuracy": 0.20222517400979995,
      "num_tokens": 81301030.0,
      "step": 44065
    },
    {
      "entropy": 5.375124454498291,
      "epoch": 3.7024574669187147,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036763603544178825,
      "loss": 4.8642,
      "mean_token_accuracy": 0.20096287578344346,
      "num_tokens": 81309621.0,
      "step": 44070
    },
    {
      "entropy": 5.404283332824707,
      "epoch": 3.7028775467338795,
      "grad_norm": 1.0,
      "learning_rate": 0.0003676087439617075,
      "loss": 4.9161,
      "mean_token_accuracy": 0.2060094729065895,
      "num_tokens": 81319536.0,
      "step": 44075
    },
    {
      "entropy": 5.465079355239868,
      "epoch": 3.7032976265490443,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003675814508411106,
      "loss": 4.9948,
      "mean_token_accuracy": 0.2111411765217781,
      "num_tokens": 81329272.0,
      "step": 44080
    },
    {
      "entropy": 5.400245189666748,
      "epoch": 3.703717706364209,
      "grad_norm": 1.09375,
      "learning_rate": 0.000367554156080481,
      "loss": 4.8702,
      "mean_token_accuracy": 0.21687754839658738,
      "num_tokens": 81338517.0,
      "step": 44085
    },
    {
      "entropy": 5.515996837615967,
      "epoch": 3.7041377861793743,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036752685968030206,
      "loss": 5.051,
      "mean_token_accuracy": 0.20169775038957596,
      "num_tokens": 81348321.0,
      "step": 44090
    },
    {
      "entropy": 5.438698625564575,
      "epoch": 3.704557865994539,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036749956164105753,
      "loss": 5.0712,
      "mean_token_accuracy": 0.20036207437515258,
      "num_tokens": 81357572.0,
      "step": 44095
    },
    {
      "entropy": 5.482440853118897,
      "epoch": 3.704977945809704,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036747226196323093,
      "loss": 5.0019,
      "mean_token_accuracy": 0.2118126004934311,
      "num_tokens": 81366633.0,
      "step": 44100
    },
    {
      "entropy": 5.487999391555786,
      "epoch": 3.7053980256248686,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036744496064730576,
      "loss": 5.0415,
      "mean_token_accuracy": 0.19959472864866257,
      "num_tokens": 81375438.0,
      "step": 44105
    },
    {
      "entropy": 5.508031892776489,
      "epoch": 3.7058181054400334,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036741765769376575,
      "loss": 4.9909,
      "mean_token_accuracy": 0.20175992250442504,
      "num_tokens": 81384642.0,
      "step": 44110
    },
    {
      "entropy": 5.432580614089966,
      "epoch": 3.7062381852551987,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003673903531030945,
      "loss": 4.897,
      "mean_token_accuracy": 0.21608383655548097,
      "num_tokens": 81393546.0,
      "step": 44115
    },
    {
      "entropy": 5.545366716384888,
      "epoch": 3.7066582650703634,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003673630468757757,
      "loss": 5.0334,
      "mean_token_accuracy": 0.20208530277013778,
      "num_tokens": 81402383.0,
      "step": 44120
    },
    {
      "entropy": 5.548081731796264,
      "epoch": 3.7070783448855282,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036733573901229314,
      "loss": 5.0601,
      "mean_token_accuracy": 0.2042398452758789,
      "num_tokens": 81411993.0,
      "step": 44125
    },
    {
      "entropy": 5.432401418685913,
      "epoch": 3.707498424700693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003673084295131304,
      "loss": 4.9547,
      "mean_token_accuracy": 0.20695390701293945,
      "num_tokens": 81421731.0,
      "step": 44130
    },
    {
      "entropy": 5.4163384437561035,
      "epoch": 3.707918504515858,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003672811183787715,
      "loss": 4.9784,
      "mean_token_accuracy": 0.20471849888563157,
      "num_tokens": 81430332.0,
      "step": 44135
    },
    {
      "entropy": 5.488544511795044,
      "epoch": 3.708338584331023,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003672538056097,
      "loss": 4.9518,
      "mean_token_accuracy": 0.20424726456403733,
      "num_tokens": 81440008.0,
      "step": 44140
    },
    {
      "entropy": 5.391430282592774,
      "epoch": 3.708758664146188,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036722649120639984,
      "loss": 4.8918,
      "mean_token_accuracy": 0.20776180922985077,
      "num_tokens": 81449129.0,
      "step": 44145
    },
    {
      "entropy": 5.426891422271728,
      "epoch": 3.7091787439613526,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003671991751693549,
      "loss": 4.8572,
      "mean_token_accuracy": 0.21960821449756623,
      "num_tokens": 81458013.0,
      "step": 44150
    },
    {
      "entropy": 5.468498945236206,
      "epoch": 3.7095988237765174,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003671718574990491,
      "loss": 5.0353,
      "mean_token_accuracy": 0.199109448492527,
      "num_tokens": 81466982.0,
      "step": 44155
    },
    {
      "entropy": 5.504359722137451,
      "epoch": 3.710018903591682,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003671445381959662,
      "loss": 5.0099,
      "mean_token_accuracy": 0.1998247817158699,
      "num_tokens": 81476737.0,
      "step": 44160
    },
    {
      "entropy": 5.454436159133911,
      "epoch": 3.7104389834068474,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003671172172605904,
      "loss": 4.9411,
      "mean_token_accuracy": 0.20876715928316117,
      "num_tokens": 81485347.0,
      "step": 44165
    },
    {
      "entropy": 5.43024697303772,
      "epoch": 3.710859063222012,
      "grad_norm": 1.125,
      "learning_rate": 0.00036708989469340543,
      "loss": 4.9763,
      "mean_token_accuracy": 0.20215524286031722,
      "num_tokens": 81495482.0,
      "step": 44170
    },
    {
      "entropy": 5.509263372421264,
      "epoch": 3.711279143037177,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036706257049489544,
      "loss": 4.97,
      "mean_token_accuracy": 0.20997652113437654,
      "num_tokens": 81504047.0,
      "step": 44175
    },
    {
      "entropy": 5.474911737442016,
      "epoch": 3.711699222852342,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003670352446655444,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20850794911384582,
      "num_tokens": 81513288.0,
      "step": 44180
    },
    {
      "entropy": 5.3182448387146,
      "epoch": 3.7121193026675066,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003670079172058364,
      "loss": 4.7894,
      "mean_token_accuracy": 0.2193761557340622,
      "num_tokens": 81522204.0,
      "step": 44185
    },
    {
      "entropy": 5.486349725723267,
      "epoch": 3.712539382482672,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036698058811625544,
      "loss": 5.0272,
      "mean_token_accuracy": 0.21065246760845185,
      "num_tokens": 81531050.0,
      "step": 44190
    },
    {
      "entropy": 5.470558118820191,
      "epoch": 3.7129594622978366,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003669532573972859,
      "loss": 5.0138,
      "mean_token_accuracy": 0.20937739610671996,
      "num_tokens": 81540370.0,
      "step": 44195
    },
    {
      "entropy": 5.5050657272338865,
      "epoch": 3.7133795421130014,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036692592504941157,
      "loss": 5.0605,
      "mean_token_accuracy": 0.20486190915107727,
      "num_tokens": 81549889.0,
      "step": 44200
    },
    {
      "entropy": 5.493495607376099,
      "epoch": 3.7137996219281666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003668985910731169,
      "loss": 4.98,
      "mean_token_accuracy": 0.2103472337126732,
      "num_tokens": 81558860.0,
      "step": 44205
    },
    {
      "entropy": 5.494503021240234,
      "epoch": 3.7142197017433314,
      "grad_norm": 1.0625,
      "learning_rate": 0.000366871255468886,
      "loss": 5.0147,
      "mean_token_accuracy": 0.20952652841806413,
      "num_tokens": 81568785.0,
      "step": 44210
    },
    {
      "entropy": 5.541560792922974,
      "epoch": 3.714639781558496,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036684391823720307,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20585637241601945,
      "num_tokens": 81577459.0,
      "step": 44215
    },
    {
      "entropy": 5.593892621994018,
      "epoch": 3.715059861373661,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003668165793785525,
      "loss": 5.2239,
      "mean_token_accuracy": 0.19593350291252137,
      "num_tokens": 81586984.0,
      "step": 44220
    },
    {
      "entropy": 5.473512601852417,
      "epoch": 3.7154799411888257,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036678923889341846,
      "loss": 4.9946,
      "mean_token_accuracy": 0.20263268649578095,
      "num_tokens": 81596020.0,
      "step": 44225
    },
    {
      "entropy": 5.518449831008911,
      "epoch": 3.715900021003991,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003667618967822854,
      "loss": 5.0439,
      "mean_token_accuracy": 0.20386722385883332,
      "num_tokens": 81605068.0,
      "step": 44230
    },
    {
      "entropy": 5.42605676651001,
      "epoch": 3.7163201008191558,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036673455304563746,
      "loss": 4.9543,
      "mean_token_accuracy": 0.20983586013317107,
      "num_tokens": 81614916.0,
      "step": 44235
    },
    {
      "entropy": 5.512793731689453,
      "epoch": 3.7167401806343205,
      "grad_norm": 1.125,
      "learning_rate": 0.0003667072076839592,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20465537011623383,
      "num_tokens": 81623693.0,
      "step": 44240
    },
    {
      "entropy": 5.4693763256073,
      "epoch": 3.7171602604494853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000366679860697735,
      "loss": 4.9528,
      "mean_token_accuracy": 0.20725513398647308,
      "num_tokens": 81632949.0,
      "step": 44245
    },
    {
      "entropy": 5.395737028121948,
      "epoch": 3.71758034026465,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036665251208744924,
      "loss": 4.9169,
      "mean_token_accuracy": 0.213603176176548,
      "num_tokens": 81642239.0,
      "step": 44250
    },
    {
      "entropy": 5.462114524841309,
      "epoch": 3.7180004200798153,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036662516185358645,
      "loss": 4.928,
      "mean_token_accuracy": 0.21850616186857225,
      "num_tokens": 81651718.0,
      "step": 44255
    },
    {
      "entropy": 5.41646409034729,
      "epoch": 3.71842049989498,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003665978099966311,
      "loss": 4.959,
      "mean_token_accuracy": 0.2083064764738083,
      "num_tokens": 81661801.0,
      "step": 44260
    },
    {
      "entropy": 5.4729170322418215,
      "epoch": 3.718840579710145,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036657045651706773,
      "loss": 5.0662,
      "mean_token_accuracy": 0.2083781659603119,
      "num_tokens": 81671667.0,
      "step": 44265
    },
    {
      "entropy": 5.500611877441406,
      "epoch": 3.7192606595253097,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003665431014153809,
      "loss": 5.0857,
      "mean_token_accuracy": 0.2031654953956604,
      "num_tokens": 81682534.0,
      "step": 44270
    },
    {
      "entropy": 5.588571166992187,
      "epoch": 3.7196807393404745,
      "grad_norm": 1.125,
      "learning_rate": 0.0003665157446920551,
      "loss": 5.1066,
      "mean_token_accuracy": 0.19717551320791243,
      "num_tokens": 81692031.0,
      "step": 44275
    },
    {
      "entropy": 5.454624938964844,
      "epoch": 3.7201008191556397,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036648838634757496,
      "loss": 5.035,
      "mean_token_accuracy": 0.1979657679796219,
      "num_tokens": 81700536.0,
      "step": 44280
    },
    {
      "entropy": 5.4547319412231445,
      "epoch": 3.7205208989708045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003664610263824252,
      "loss": 4.9562,
      "mean_token_accuracy": 0.2067284479737282,
      "num_tokens": 81709369.0,
      "step": 44285
    },
    {
      "entropy": 5.476762771606445,
      "epoch": 3.7209409787859693,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003664336647970905,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20934143513441086,
      "num_tokens": 81718995.0,
      "step": 44290
    },
    {
      "entropy": 5.416195631027222,
      "epoch": 3.721361058601134,
      "grad_norm": 1.125,
      "learning_rate": 0.00036640630159205553,
      "loss": 4.9136,
      "mean_token_accuracy": 0.20850517004728317,
      "num_tokens": 81728454.0,
      "step": 44295
    },
    {
      "entropy": 5.490900325775146,
      "epoch": 3.721781138416299,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003663789367678049,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21053385138511657,
      "num_tokens": 81738357.0,
      "step": 44300
    },
    {
      "entropy": 5.587497520446777,
      "epoch": 3.722201218231464,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036635157032482356,
      "loss": 5.0312,
      "mean_token_accuracy": 0.2016616776585579,
      "num_tokens": 81748050.0,
      "step": 44305
    },
    {
      "entropy": 5.3208150386810305,
      "epoch": 3.722621298046629,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036632420226359613,
      "loss": 4.8,
      "mean_token_accuracy": 0.21809208542108535,
      "num_tokens": 81756887.0,
      "step": 44310
    },
    {
      "entropy": 5.494419956207276,
      "epoch": 3.7230413778617937,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036629683258460754,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19556164592504502,
      "num_tokens": 81766198.0,
      "step": 44315
    },
    {
      "entropy": 5.42340087890625,
      "epoch": 3.7234614576769585,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036626946128834255,
      "loss": 4.886,
      "mean_token_accuracy": 0.21027726978063582,
      "num_tokens": 81775095.0,
      "step": 44320
    },
    {
      "entropy": 5.554959344863891,
      "epoch": 3.7238815374921233,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036624208837528595,
      "loss": 5.0342,
      "mean_token_accuracy": 0.20084683150053023,
      "num_tokens": 81784450.0,
      "step": 44325
    },
    {
      "entropy": 5.43545560836792,
      "epoch": 3.7243016173072885,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003662147138459228,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20430592745542525,
      "num_tokens": 81793659.0,
      "step": 44330
    },
    {
      "entropy": 5.4709546089172365,
      "epoch": 3.7247216971224533,
      "grad_norm": 1.0,
      "learning_rate": 0.00036618733770073806,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20378649830818177,
      "num_tokens": 81802815.0,
      "step": 44335
    },
    {
      "entropy": 5.551650524139404,
      "epoch": 3.725141776937618,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003661599599402165,
      "loss": 5.0333,
      "mean_token_accuracy": 0.2027538925409317,
      "num_tokens": 81812111.0,
      "step": 44340
    },
    {
      "entropy": 5.402217006683349,
      "epoch": 3.7255618567527833,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036613258056484313,
      "loss": 4.9767,
      "mean_token_accuracy": 0.21290911138057708,
      "num_tokens": 81821229.0,
      "step": 44345
    },
    {
      "entropy": 5.436000823974609,
      "epoch": 3.725981936567948,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036610519957510306,
      "loss": 5.0186,
      "mean_token_accuracy": 0.19494698643684388,
      "num_tokens": 81831011.0,
      "step": 44350
    },
    {
      "entropy": 5.552644443511963,
      "epoch": 3.726402016383113,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00036607781697148127,
      "loss": 5.0548,
      "mean_token_accuracy": 0.2119493916630745,
      "num_tokens": 81841865.0,
      "step": 44355
    },
    {
      "entropy": 5.5055256366729735,
      "epoch": 3.7268220961982776,
      "grad_norm": 1.046875,
      "learning_rate": 0.000366050432754463,
      "loss": 5.0274,
      "mean_token_accuracy": 0.19968254417181014,
      "num_tokens": 81851348.0,
      "step": 44360
    },
    {
      "entropy": 5.518359804153443,
      "epoch": 3.7272421760134424,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036602304692453304,
      "loss": 4.9833,
      "mean_token_accuracy": 0.20612193048000335,
      "num_tokens": 81860314.0,
      "step": 44365
    },
    {
      "entropy": 5.44634976387024,
      "epoch": 3.7276622558286077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003659956594821768,
      "loss": 4.9748,
      "mean_token_accuracy": 0.2050852119922638,
      "num_tokens": 81869526.0,
      "step": 44370
    },
    {
      "entropy": 5.496985054016113,
      "epoch": 3.7280823356437724,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00036596827042787924,
      "loss": 4.9286,
      "mean_token_accuracy": 0.21401968002319335,
      "num_tokens": 81878585.0,
      "step": 44375
    },
    {
      "entropy": 5.535724735260009,
      "epoch": 3.7285024154589372,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003659408797621256,
      "loss": 5.0278,
      "mean_token_accuracy": 0.20034208446741103,
      "num_tokens": 81888310.0,
      "step": 44380
    },
    {
      "entropy": 5.4230639934539795,
      "epoch": 3.728922495274102,
      "grad_norm": 1.125,
      "learning_rate": 0.0003659134874854011,
      "loss": 4.9535,
      "mean_token_accuracy": 0.20395065248012542,
      "num_tokens": 81897587.0,
      "step": 44385
    },
    {
      "entropy": 5.411967706680298,
      "epoch": 3.729342575089267,
      "grad_norm": 1.09375,
      "learning_rate": 0.000365886093598191,
      "loss": 4.9814,
      "mean_token_accuracy": 0.20269029587507248,
      "num_tokens": 81906189.0,
      "step": 44390
    },
    {
      "entropy": 5.384929275512695,
      "epoch": 3.729762654904432,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036585869810098055,
      "loss": 4.8706,
      "mean_token_accuracy": 0.21442365199327468,
      "num_tokens": 81915744.0,
      "step": 44395
    },
    {
      "entropy": 5.489529991149903,
      "epoch": 3.730182734719597,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000365831300994255,
      "loss": 5.0017,
      "mean_token_accuracy": 0.20427645295858382,
      "num_tokens": 81925775.0,
      "step": 44400
    },
    {
      "entropy": 5.489974164962769,
      "epoch": 3.7306028145347616,
      "grad_norm": 1.0,
      "learning_rate": 0.0003658039022784998,
      "loss": 4.9485,
      "mean_token_accuracy": 0.21523386389017105,
      "num_tokens": 81935717.0,
      "step": 44405
    },
    {
      "entropy": 5.513744878768921,
      "epoch": 3.7310228943499264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003657765019542002,
      "loss": 5.0043,
      "mean_token_accuracy": 0.2062036857008934,
      "num_tokens": 81944207.0,
      "step": 44410
    },
    {
      "entropy": 5.501269102096558,
      "epoch": 3.731442974165091,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003657491000218415,
      "loss": 5.0508,
      "mean_token_accuracy": 0.1961793214082718,
      "num_tokens": 81953060.0,
      "step": 44415
    },
    {
      "entropy": 5.441328763961792,
      "epoch": 3.7318630539802564,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003657216964819093,
      "loss": 4.9669,
      "mean_token_accuracy": 0.2063187390565872,
      "num_tokens": 81962852.0,
      "step": 44420
    },
    {
      "entropy": 5.436296796798706,
      "epoch": 3.732283133795421,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036569429133488906,
      "loss": 4.9872,
      "mean_token_accuracy": 0.21173696219921112,
      "num_tokens": 81971693.0,
      "step": 44425
    },
    {
      "entropy": 5.455056524276733,
      "epoch": 3.732703213610586,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036566688458126604,
      "loss": 4.9107,
      "mean_token_accuracy": 0.2114875376224518,
      "num_tokens": 81981259.0,
      "step": 44430
    },
    {
      "entropy": 5.393499565124512,
      "epoch": 3.7331232934257508,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036563947622152593,
      "loss": 4.9127,
      "mean_token_accuracy": 0.22251986116170883,
      "num_tokens": 81989111.0,
      "step": 44435
    },
    {
      "entropy": 5.4876962184906,
      "epoch": 3.7335433732409156,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003656120662561542,
      "loss": 4.9777,
      "mean_token_accuracy": 0.20585610419511796,
      "num_tokens": 81998351.0,
      "step": 44440
    },
    {
      "entropy": 5.433440828323365,
      "epoch": 3.733963453056081,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003655846546856363,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2008989378809929,
      "num_tokens": 82007952.0,
      "step": 44445
    },
    {
      "entropy": 5.445834636688232,
      "epoch": 3.7343835328712456,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036555724151045787,
      "loss": 4.9018,
      "mean_token_accuracy": 0.21257359087467192,
      "num_tokens": 82016520.0,
      "step": 44450
    },
    {
      "entropy": 5.4867910861969,
      "epoch": 3.7348036126864104,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036552982673110454,
      "loss": 4.9968,
      "mean_token_accuracy": 0.2059198260307312,
      "num_tokens": 82025015.0,
      "step": 44455
    },
    {
      "entropy": 5.490011930465698,
      "epoch": 3.735223692501575,
      "grad_norm": 1.015625,
      "learning_rate": 0.000365502410348062,
      "loss": 5.0765,
      "mean_token_accuracy": 0.19642151147127151,
      "num_tokens": 82035437.0,
      "step": 44460
    },
    {
      "entropy": 5.486982822418213,
      "epoch": 3.73564377231674,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036547499236181573,
      "loss": 5.0426,
      "mean_token_accuracy": 0.19949680417776108,
      "num_tokens": 82045034.0,
      "step": 44465
    },
    {
      "entropy": 5.579323148727417,
      "epoch": 3.736063852131905,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036544757277285174,
      "loss": 5.1048,
      "mean_token_accuracy": 0.1954721376299858,
      "num_tokens": 82054152.0,
      "step": 44470
    },
    {
      "entropy": 5.461007833480835,
      "epoch": 3.73648393194707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036542015158165547,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21411746889352798,
      "num_tokens": 82062920.0,
      "step": 44475
    },
    {
      "entropy": 5.412520790100098,
      "epoch": 3.7369040117622347,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036539272878871273,
      "loss": 4.9508,
      "mean_token_accuracy": 0.2108491137623787,
      "num_tokens": 82072100.0,
      "step": 44480
    },
    {
      "entropy": 5.426235580444336,
      "epoch": 3.7373240915774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036536530439450936,
      "loss": 5.0168,
      "mean_token_accuracy": 0.2069403722882271,
      "num_tokens": 82081471.0,
      "step": 44485
    },
    {
      "entropy": 5.5617235660552975,
      "epoch": 3.7377441713925643,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003653378783995312,
      "loss": 4.98,
      "mean_token_accuracy": 0.20566904544830322,
      "num_tokens": 82091277.0,
      "step": 44490
    },
    {
      "entropy": 5.473322248458862,
      "epoch": 3.7381642512077295,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003653104508042641,
      "loss": 4.9562,
      "mean_token_accuracy": 0.20745487064123153,
      "num_tokens": 82100767.0,
      "step": 44495
    },
    {
      "entropy": 5.42320442199707,
      "epoch": 3.7385843310228943,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003652830216091938,
      "loss": 4.9584,
      "mean_token_accuracy": 0.20895353853702545,
      "num_tokens": 82109698.0,
      "step": 44500
    },
    {
      "entropy": 5.4403304100036625,
      "epoch": 3.739004410838059,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036525559081480624,
      "loss": 4.9721,
      "mean_token_accuracy": 0.20352687537670136,
      "num_tokens": 82119469.0,
      "step": 44505
    },
    {
      "entropy": 5.458084630966186,
      "epoch": 3.7394244906532244,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036522815842158737,
      "loss": 4.9067,
      "mean_token_accuracy": 0.209610952436924,
      "num_tokens": 82128194.0,
      "step": 44510
    },
    {
      "entropy": 5.3791872501373295,
      "epoch": 3.739844570468389,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036520072443002317,
      "loss": 4.8386,
      "mean_token_accuracy": 0.22094373852014543,
      "num_tokens": 82137107.0,
      "step": 44515
    },
    {
      "entropy": 5.3567358493804935,
      "epoch": 3.740264650283554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003651732888405996,
      "loss": 4.9228,
      "mean_token_accuracy": 0.21629729121923447,
      "num_tokens": 82146911.0,
      "step": 44520
    },
    {
      "entropy": 5.49135193824768,
      "epoch": 3.7406847300987187,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003651458516538026,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21751806288957595,
      "num_tokens": 82155363.0,
      "step": 44525
    },
    {
      "entropy": 5.476598739624023,
      "epoch": 3.7411048099138835,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036511841287011824,
      "loss": 4.9576,
      "mean_token_accuracy": 0.2070927679538727,
      "num_tokens": 82163826.0,
      "step": 44530
    },
    {
      "entropy": 5.549552392959595,
      "epoch": 3.7415248897290487,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003650909724900327,
      "loss": 5.1462,
      "mean_token_accuracy": 0.19564874917268754,
      "num_tokens": 82173630.0,
      "step": 44535
    },
    {
      "entropy": 5.464767837524414,
      "epoch": 3.7419449695442135,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036506353051403185,
      "loss": 4.9267,
      "mean_token_accuracy": 0.2124742701649666,
      "num_tokens": 82184142.0,
      "step": 44540
    },
    {
      "entropy": 5.517882347106934,
      "epoch": 3.7423650493593783,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000365036086942602,
      "loss": 4.9267,
      "mean_token_accuracy": 0.20377090573310852,
      "num_tokens": 82192674.0,
      "step": 44545
    },
    {
      "entropy": 5.5627968311309814,
      "epoch": 3.742785129174543,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003650086417762292,
      "loss": 5.101,
      "mean_token_accuracy": 0.2019645169377327,
      "num_tokens": 82201714.0,
      "step": 44550
    },
    {
      "entropy": 5.434895610809326,
      "epoch": 3.743205208989708,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003649811950153996,
      "loss": 4.9458,
      "mean_token_accuracy": 0.2057600200176239,
      "num_tokens": 82211448.0,
      "step": 44555
    },
    {
      "entropy": 5.413529682159424,
      "epoch": 3.743625288804873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003649537466605996,
      "loss": 4.9494,
      "mean_token_accuracy": 0.20108400881290436,
      "num_tokens": 82220754.0,
      "step": 44560
    },
    {
      "entropy": 5.485084104537964,
      "epoch": 3.744045368620038,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003649262967123152,
      "loss": 5.0201,
      "mean_token_accuracy": 0.20175733119249345,
      "num_tokens": 82230015.0,
      "step": 44565
    },
    {
      "entropy": 5.514227867126465,
      "epoch": 3.7444654484352027,
      "grad_norm": 1.125,
      "learning_rate": 0.0003648988451710328,
      "loss": 5.0152,
      "mean_token_accuracy": 0.20471455603837968,
      "num_tokens": 82239897.0,
      "step": 44570
    },
    {
      "entropy": 5.3798911571502686,
      "epoch": 3.7448855282503675,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036487139203723873,
      "loss": 4.8212,
      "mean_token_accuracy": 0.21460749804973603,
      "num_tokens": 82248986.0,
      "step": 44575
    },
    {
      "entropy": 5.454190254211426,
      "epoch": 3.7453056080655323,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003648439373114191,
      "loss": 4.9885,
      "mean_token_accuracy": 0.20676469951868057,
      "num_tokens": 82258026.0,
      "step": 44580
    },
    {
      "entropy": 5.4305253505706785,
      "epoch": 3.7457256878806975,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036481648099406046,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21422784477472306,
      "num_tokens": 82266975.0,
      "step": 44585
    },
    {
      "entropy": 5.481308650970459,
      "epoch": 3.7461457676958623,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000364789023085649,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2064451679587364,
      "num_tokens": 82276628.0,
      "step": 44590
    },
    {
      "entropy": 5.437481498718261,
      "epoch": 3.746565847511027,
      "grad_norm": 1.125,
      "learning_rate": 0.0003647615635866713,
      "loss": 4.9387,
      "mean_token_accuracy": 0.21027205884456635,
      "num_tokens": 82284862.0,
      "step": 44595
    },
    {
      "entropy": 5.4458027362823485,
      "epoch": 3.746985927326192,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036473410249761377,
      "loss": 5.0031,
      "mean_token_accuracy": 0.201484876871109,
      "num_tokens": 82294010.0,
      "step": 44600
    },
    {
      "entropy": 5.499577188491822,
      "epoch": 3.7474060071413566,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003647066398189628,
      "loss": 4.9927,
      "mean_token_accuracy": 0.1982037678360939,
      "num_tokens": 82303081.0,
      "step": 44605
    },
    {
      "entropy": 5.575963687896729,
      "epoch": 3.747826086956522,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036467917555120484,
      "loss": 5.0592,
      "mean_token_accuracy": 0.19933711886405944,
      "num_tokens": 82311427.0,
      "step": 44610
    },
    {
      "entropy": 5.461174249649048,
      "epoch": 3.7482461667716866,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003646517096948265,
      "loss": 4.9976,
      "mean_token_accuracy": 0.2034676656126976,
      "num_tokens": 82322641.0,
      "step": 44615
    },
    {
      "entropy": 5.477996063232422,
      "epoch": 3.7486662465868514,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003646242422503143,
      "loss": 4.9878,
      "mean_token_accuracy": 0.20027432441711426,
      "num_tokens": 82331462.0,
      "step": 44620
    },
    {
      "entropy": 5.553630352020264,
      "epoch": 3.7490863264020162,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003645967732181548,
      "loss": 5.048,
      "mean_token_accuracy": 0.1994422748684883,
      "num_tokens": 82341776.0,
      "step": 44625
    },
    {
      "entropy": 5.457800817489624,
      "epoch": 3.749506406217181,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003645693025988345,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21110413521528243,
      "num_tokens": 82351784.0,
      "step": 44630
    },
    {
      "entropy": 5.464189100265503,
      "epoch": 3.7499264860323462,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036454183039284023,
      "loss": 4.974,
      "mean_token_accuracy": 0.2075361877679825,
      "num_tokens": 82361587.0,
      "step": 44635
    },
    {
      "entropy": 5.565709114074707,
      "epoch": 3.750346565847511,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036451435660065853,
      "loss": 5.0549,
      "mean_token_accuracy": 0.19681501984596253,
      "num_tokens": 82371836.0,
      "step": 44640
    },
    {
      "entropy": 5.453589963912964,
      "epoch": 3.750766645662676,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003644868812227761,
      "loss": 4.9469,
      "mean_token_accuracy": 0.20985159873962403,
      "num_tokens": 82381450.0,
      "step": 44645
    },
    {
      "entropy": 5.467759609222412,
      "epoch": 3.751186725477841,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003644594042596796,
      "loss": 5.0218,
      "mean_token_accuracy": 0.2026531994342804,
      "num_tokens": 82390960.0,
      "step": 44650
    },
    {
      "entropy": 5.393474960327149,
      "epoch": 3.751606805293006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003644319257118558,
      "loss": 4.8931,
      "mean_token_accuracy": 0.20722440928220748,
      "num_tokens": 82400683.0,
      "step": 44655
    },
    {
      "entropy": 5.433473682403564,
      "epoch": 3.7520268851081706,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036440444557979157,
      "loss": 4.9679,
      "mean_token_accuracy": 0.21431687027215957,
      "num_tokens": 82410729.0,
      "step": 44660
    },
    {
      "entropy": 5.508470726013184,
      "epoch": 3.7524469649233354,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036437696386397363,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20658175647258759,
      "num_tokens": 82418740.0,
      "step": 44665
    },
    {
      "entropy": 5.444218397140503,
      "epoch": 3.7528670447385,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036434948056488875,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2077540710568428,
      "num_tokens": 82427675.0,
      "step": 44670
    },
    {
      "entropy": 5.450279235839844,
      "epoch": 3.7532871245536654,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003643219956830239,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20810138136148454,
      "num_tokens": 82438433.0,
      "step": 44675
    },
    {
      "entropy": 5.431361818313599,
      "epoch": 3.75370720436883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036429450921886573,
      "loss": 4.9486,
      "mean_token_accuracy": 0.2047027125954628,
      "num_tokens": 82447699.0,
      "step": 44680
    },
    {
      "entropy": 5.480170106887817,
      "epoch": 3.754127284183995,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003642670211729014,
      "loss": 5.0083,
      "mean_token_accuracy": 0.21225569248199463,
      "num_tokens": 82457067.0,
      "step": 44685
    },
    {
      "entropy": 5.408309745788574,
      "epoch": 3.75454736399916,
      "grad_norm": 1.125,
      "learning_rate": 0.0003642395315456178,
      "loss": 4.8811,
      "mean_token_accuracy": 0.21132710427045823,
      "num_tokens": 82465160.0,
      "step": 44690
    },
    {
      "entropy": 5.462009334564209,
      "epoch": 3.7549674438143246,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003642120403375018,
      "loss": 4.9512,
      "mean_token_accuracy": 0.2081238731741905,
      "num_tokens": 82474238.0,
      "step": 44695
    },
    {
      "entropy": 5.4919085025787355,
      "epoch": 3.75538752362949,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036418454754904043,
      "loss": 5.0297,
      "mean_token_accuracy": 0.20913095921278,
      "num_tokens": 82483072.0,
      "step": 44700
    },
    {
      "entropy": 5.5137686252594,
      "epoch": 3.7558076034446546,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003641570531807207,
      "loss": 5.0766,
      "mean_token_accuracy": 0.19830048978328704,
      "num_tokens": 82492657.0,
      "step": 44705
    },
    {
      "entropy": 5.490079164505005,
      "epoch": 3.7562276832598194,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036412955723302983,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21121624410152434,
      "num_tokens": 82501462.0,
      "step": 44710
    },
    {
      "entropy": 5.455183601379394,
      "epoch": 3.756647763074984,
      "grad_norm": 1.59375,
      "learning_rate": 0.0003641020597064546,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20826236754655839,
      "num_tokens": 82510722.0,
      "step": 44715
    },
    {
      "entropy": 5.426322174072266,
      "epoch": 3.757067842890149,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003640745606014823,
      "loss": 4.9708,
      "mean_token_accuracy": 0.20567276924848557,
      "num_tokens": 82519092.0,
      "step": 44720
    },
    {
      "entropy": 5.402661085128784,
      "epoch": 3.757487922705314,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003640470599186,
      "loss": 4.8994,
      "mean_token_accuracy": 0.2100374311208725,
      "num_tokens": 82528199.0,
      "step": 44725
    },
    {
      "entropy": 5.4657848358154295,
      "epoch": 3.757908002520479,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036401955765829496,
      "loss": 4.9243,
      "mean_token_accuracy": 0.2059900000691414,
      "num_tokens": 82537051.0,
      "step": 44730
    },
    {
      "entropy": 5.443833732604981,
      "epoch": 3.7583280823356437,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036399205382105427,
      "loss": 5.0279,
      "mean_token_accuracy": 0.19930174052715302,
      "num_tokens": 82546707.0,
      "step": 44735
    },
    {
      "entropy": 5.447762489318848,
      "epoch": 3.7587481621508085,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036396454840736513,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2098611369729042,
      "num_tokens": 82555253.0,
      "step": 44740
    },
    {
      "entropy": 5.455058813095093,
      "epoch": 3.7591682419659733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003639370414177149,
      "loss": 4.875,
      "mean_token_accuracy": 0.2188514620065689,
      "num_tokens": 82564332.0,
      "step": 44745
    },
    {
      "entropy": 5.450922584533691,
      "epoch": 3.7595883217811386,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003639095328525907,
      "loss": 4.958,
      "mean_token_accuracy": 0.20204814821481704,
      "num_tokens": 82573398.0,
      "step": 44750
    },
    {
      "entropy": 5.46222825050354,
      "epoch": 3.7600084015963033,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003638820227124799,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20122946053743362,
      "num_tokens": 82582356.0,
      "step": 44755
    },
    {
      "entropy": 5.556595373153686,
      "epoch": 3.760428481411468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003638545109978699,
      "loss": 5.1055,
      "mean_token_accuracy": 0.20044035017490386,
      "num_tokens": 82591419.0,
      "step": 44760
    },
    {
      "entropy": 5.541141510009766,
      "epoch": 3.760848561226633,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036382699770924794,
      "loss": 5.0786,
      "mean_token_accuracy": 0.20022474080324174,
      "num_tokens": 82601816.0,
      "step": 44765
    },
    {
      "entropy": 5.4625612735748295,
      "epoch": 3.7612686410417977,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036379948284710154,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2087634190917015,
      "num_tokens": 82611428.0,
      "step": 44770
    },
    {
      "entropy": 5.553218793869019,
      "epoch": 3.761688720856963,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003637719664119179,
      "loss": 5.0972,
      "mean_token_accuracy": 0.19652532488107682,
      "num_tokens": 82621073.0,
      "step": 44775
    },
    {
      "entropy": 5.45903697013855,
      "epoch": 3.7621088006721277,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036374444840418465,
      "loss": 4.9931,
      "mean_token_accuracy": 0.20381225496530533,
      "num_tokens": 82630422.0,
      "step": 44780
    },
    {
      "entropy": 5.491938543319702,
      "epoch": 3.7625288804872925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036371692882438924,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20093890726566316,
      "num_tokens": 82639272.0,
      "step": 44785
    },
    {
      "entropy": 5.470127725601197,
      "epoch": 3.7629489603024577,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036368940767301903,
      "loss": 5.0075,
      "mean_token_accuracy": 0.20518796890974045,
      "num_tokens": 82648228.0,
      "step": 44790
    },
    {
      "entropy": 5.503501462936401,
      "epoch": 3.763369040117622,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003636618849505617,
      "loss": 4.9784,
      "mean_token_accuracy": 0.2060699164867401,
      "num_tokens": 82657335.0,
      "step": 44795
    },
    {
      "entropy": 5.586040925979614,
      "epoch": 3.7637891199327873,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003636343606575047,
      "loss": 5.0688,
      "mean_token_accuracy": 0.19594500213861465,
      "num_tokens": 82666786.0,
      "step": 44800
    },
    {
      "entropy": 5.411420011520386,
      "epoch": 3.764209199747952,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036360683479433564,
      "loss": 4.8621,
      "mean_token_accuracy": 0.216522653400898,
      "num_tokens": 82675046.0,
      "step": 44805
    },
    {
      "entropy": 5.534075593948364,
      "epoch": 3.764629279563117,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036357930736154214,
      "loss": 4.9834,
      "mean_token_accuracy": 0.2071324571967125,
      "num_tokens": 82684206.0,
      "step": 44810
    },
    {
      "entropy": 5.530736351013184,
      "epoch": 3.765049359378282,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003635517783596117,
      "loss": 5.0194,
      "mean_token_accuracy": 0.20489497631788253,
      "num_tokens": 82693913.0,
      "step": 44815
    },
    {
      "entropy": 5.432253885269165,
      "epoch": 3.765469439193447,
      "grad_norm": 1.125,
      "learning_rate": 0.0003635242477890323,
      "loss": 4.9493,
      "mean_token_accuracy": 0.21354881227016448,
      "num_tokens": 82703601.0,
      "step": 44820
    },
    {
      "entropy": 5.4097918510437015,
      "epoch": 3.7658895190086117,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036349671565029135,
      "loss": 4.9542,
      "mean_token_accuracy": 0.20868767499923707,
      "num_tokens": 82712857.0,
      "step": 44825
    },
    {
      "entropy": 5.444096565246582,
      "epoch": 3.7663095988237765,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036346918194387663,
      "loss": 4.9453,
      "mean_token_accuracy": 0.20811535120010377,
      "num_tokens": 82722151.0,
      "step": 44830
    },
    {
      "entropy": 5.440825653076172,
      "epoch": 3.7667296786389413,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036344164667027597,
      "loss": 4.9928,
      "mean_token_accuracy": 0.2110084593296051,
      "num_tokens": 82731371.0,
      "step": 44835
    },
    {
      "entropy": 5.399930095672607,
      "epoch": 3.7671497584541065,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036341410982997704,
      "loss": 4.9695,
      "mean_token_accuracy": 0.20392924100160598,
      "num_tokens": 82740629.0,
      "step": 44840
    },
    {
      "entropy": 5.4543358325958256,
      "epoch": 3.7675698382692713,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003633865714234677,
      "loss": 4.9786,
      "mean_token_accuracy": 0.21122181713581084,
      "num_tokens": 82751004.0,
      "step": 44845
    },
    {
      "entropy": 5.4545526027679445,
      "epoch": 3.767989918084436,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036335903145123566,
      "loss": 4.9265,
      "mean_token_accuracy": 0.21379536092281343,
      "num_tokens": 82761665.0,
      "step": 44850
    },
    {
      "entropy": 5.500798940658569,
      "epoch": 3.768409997899601,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000363331489913769,
      "loss": 5.0298,
      "mean_token_accuracy": 0.20475834608078003,
      "num_tokens": 82771546.0,
      "step": 44855
    },
    {
      "entropy": 5.534832000732422,
      "epoch": 3.7688300777147656,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003633039468115554,
      "loss": 4.9872,
      "mean_token_accuracy": 0.2062191277742386,
      "num_tokens": 82780557.0,
      "step": 44860
    },
    {
      "entropy": 5.59613528251648,
      "epoch": 3.769250157529931,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003632764021450829,
      "loss": 5.1356,
      "mean_token_accuracy": 0.19699543118476867,
      "num_tokens": 82789909.0,
      "step": 44865
    },
    {
      "entropy": 5.505459356307983,
      "epoch": 3.7696702373450957,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003632488559148394,
      "loss": 4.9633,
      "mean_token_accuracy": 0.20829520225524903,
      "num_tokens": 82799789.0,
      "step": 44870
    },
    {
      "entropy": 5.443078470230103,
      "epoch": 3.7700903171602604,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036322130812131275,
      "loss": 4.9468,
      "mean_token_accuracy": 0.2061211034655571,
      "num_tokens": 82808530.0,
      "step": 44875
    },
    {
      "entropy": 5.430437469482422,
      "epoch": 3.7705103969754252,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036319375876499115,
      "loss": 4.9665,
      "mean_token_accuracy": 0.21306073069572448,
      "num_tokens": 82817105.0,
      "step": 44880
    },
    {
      "entropy": 5.453491735458374,
      "epoch": 3.77093047679059,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036316620784636257,
      "loss": 4.938,
      "mean_token_accuracy": 0.20634080618619918,
      "num_tokens": 82826761.0,
      "step": 44885
    },
    {
      "entropy": 5.429092121124268,
      "epoch": 3.7713505566057552,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00036313865536591496,
      "loss": 4.9183,
      "mean_token_accuracy": 0.212340746819973,
      "num_tokens": 82836458.0,
      "step": 44890
    },
    {
      "entropy": 5.450584411621094,
      "epoch": 3.77177063642092,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036311110132413647,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2069968968629837,
      "num_tokens": 82845180.0,
      "step": 44895
    },
    {
      "entropy": 5.511199712753296,
      "epoch": 3.772190716236085,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036308354572151517,
      "loss": 4.9707,
      "mean_token_accuracy": 0.2135113149881363,
      "num_tokens": 82853416.0,
      "step": 44900
    },
    {
      "entropy": 5.420588970184326,
      "epoch": 3.7726107960512496,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003630559885585392,
      "loss": 4.9026,
      "mean_token_accuracy": 0.21532950103282927,
      "num_tokens": 82861994.0,
      "step": 44905
    },
    {
      "entropy": 5.4969391345977785,
      "epoch": 3.7730308758664144,
      "grad_norm": 1.125,
      "learning_rate": 0.00036302842983569676,
      "loss": 5.0028,
      "mean_token_accuracy": 0.21052107214927673,
      "num_tokens": 82871568.0,
      "step": 44910
    },
    {
      "entropy": 5.457197761535644,
      "epoch": 3.7734509556815796,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036300086955347603,
      "loss": 5.0079,
      "mean_token_accuracy": 0.20561204552650453,
      "num_tokens": 82880576.0,
      "step": 44915
    },
    {
      "entropy": 5.446472549438477,
      "epoch": 3.7738710354967444,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003629733077123652,
      "loss": 4.8983,
      "mean_token_accuracy": 0.20971639603376388,
      "num_tokens": 82889588.0,
      "step": 44920
    },
    {
      "entropy": 5.616317653656006,
      "epoch": 3.774291115311909,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003629457443128525,
      "loss": 5.1143,
      "mean_token_accuracy": 0.19757925868034362,
      "num_tokens": 82898649.0,
      "step": 44925
    },
    {
      "entropy": 5.471220827102661,
      "epoch": 3.774711195127074,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036291817935542623,
      "loss": 4.91,
      "mean_token_accuracy": 0.2100493848323822,
      "num_tokens": 82908362.0,
      "step": 44930
    },
    {
      "entropy": 5.421352338790894,
      "epoch": 3.7751312749422388,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036289061284057466,
      "loss": 4.915,
      "mean_token_accuracy": 0.21250669956207274,
      "num_tokens": 82918245.0,
      "step": 44935
    },
    {
      "entropy": 5.459627723693847,
      "epoch": 3.775551354757404,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036286304476878617,
      "loss": 5.0124,
      "mean_token_accuracy": 0.20957471281290055,
      "num_tokens": 82927345.0,
      "step": 44940
    },
    {
      "entropy": 5.507727718353271,
      "epoch": 3.775971434572569,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003628354751405491,
      "loss": 5.0623,
      "mean_token_accuracy": 0.20139686465263368,
      "num_tokens": 82936303.0,
      "step": 44945
    },
    {
      "entropy": 5.551179122924805,
      "epoch": 3.7763915143877336,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003628079039563518,
      "loss": 5.1018,
      "mean_token_accuracy": 0.1985434964299202,
      "num_tokens": 82945817.0,
      "step": 44950
    },
    {
      "entropy": 5.492087697982788,
      "epoch": 3.776811594202899,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003627803312166827,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20731378644704818,
      "num_tokens": 82954715.0,
      "step": 44955
    },
    {
      "entropy": 5.547045373916626,
      "epoch": 3.7772316740180636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036275275692203025,
      "loss": 5.1174,
      "mean_token_accuracy": 0.20193683207035065,
      "num_tokens": 82964425.0,
      "step": 44960
    },
    {
      "entropy": 5.4453246116638185,
      "epoch": 3.7776517538332284,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036272518107288286,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20602482110261916,
      "num_tokens": 82973227.0,
      "step": 44965
    },
    {
      "entropy": 5.515623617172241,
      "epoch": 3.778071833648393,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036269760366972915,
      "loss": 5.0501,
      "mean_token_accuracy": 0.20318607687950135,
      "num_tokens": 82982705.0,
      "step": 44970
    },
    {
      "entropy": 5.4563555240631105,
      "epoch": 3.778491913463558,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036267002471305744,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21166544109582902,
      "num_tokens": 82991921.0,
      "step": 44975
    },
    {
      "entropy": 5.463559150695801,
      "epoch": 3.778911993278723,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003626424442033564,
      "loss": 4.9656,
      "mean_token_accuracy": 0.20870956480503083,
      "num_tokens": 83001565.0,
      "step": 44980
    },
    {
      "entropy": 5.4462837219238285,
      "epoch": 3.779332073093888,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036261486214111467,
      "loss": 4.9955,
      "mean_token_accuracy": 0.20150898844003678,
      "num_tokens": 83012534.0,
      "step": 44985
    },
    {
      "entropy": 5.4707294464111325,
      "epoch": 3.7797521529090528,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036258727852682074,
      "loss": 4.9866,
      "mean_token_accuracy": 0.20761049687862396,
      "num_tokens": 83021469.0,
      "step": 44990
    },
    {
      "entropy": 5.435497188568116,
      "epoch": 3.7801722327242175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003625596933609632,
      "loss": 4.9613,
      "mean_token_accuracy": 0.20300392508506776,
      "num_tokens": 83030727.0,
      "step": 44995
    },
    {
      "entropy": 5.491104936599731,
      "epoch": 3.7805923125393823,
      "grad_norm": 1.125,
      "learning_rate": 0.0003625321066440308,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20678451657295227,
      "num_tokens": 83039452.0,
      "step": 45000
    },
    {
      "epoch": 3.7805923125393823,
      "eval_entropy": 5.25115816120008,
      "eval_loss": 5.068654537200928,
      "eval_mean_token_accuracy": 0.21079218018504134,
      "eval_num_tokens": 83039452.0,
      "eval_runtime": 27.3793,
      "eval_samples_per_second": 1364.751,
      "eval_steps_per_second": 170.603,
      "step": 45000
    },
    {
      "entropy": 5.554234790802002,
      "epoch": 3.7810123923545476,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003625045183765122,
      "loss": 5.0339,
      "mean_token_accuracy": 0.19985276013612746,
      "num_tokens": 83048346.0,
      "step": 45005
    },
    {
      "entropy": 5.522035026550293,
      "epoch": 3.7814324721697123,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036247692855889615,
      "loss": 5.0134,
      "mean_token_accuracy": 0.2071347340941429,
      "num_tokens": 83056970.0,
      "step": 45010
    },
    {
      "entropy": 5.432668685913086,
      "epoch": 3.781852551984877,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003624493371916714,
      "loss": 4.9173,
      "mean_token_accuracy": 0.20814963728189467,
      "num_tokens": 83065714.0,
      "step": 45015
    },
    {
      "entropy": 5.422872400283813,
      "epoch": 3.782272631800042,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036242174427532657,
      "loss": 4.9604,
      "mean_token_accuracy": 0.21443890184164047,
      "num_tokens": 83075257.0,
      "step": 45020
    },
    {
      "entropy": 5.520058441162109,
      "epoch": 3.7826927116152067,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036239414981035063,
      "loss": 5.0225,
      "mean_token_accuracy": 0.2004452496767044,
      "num_tokens": 83085112.0,
      "step": 45025
    },
    {
      "entropy": 5.549726963043213,
      "epoch": 3.783112791430372,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003623665537972323,
      "loss": 5.048,
      "mean_token_accuracy": 0.2061794176697731,
      "num_tokens": 83094029.0,
      "step": 45030
    },
    {
      "entropy": 5.488968420028686,
      "epoch": 3.7835328712455367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003623389562364605,
      "loss": 4.9653,
      "mean_token_accuracy": 0.20301298946142196,
      "num_tokens": 83102710.0,
      "step": 45035
    },
    {
      "entropy": 5.410503816604614,
      "epoch": 3.7839529510607015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036231135712852405,
      "loss": 4.8658,
      "mean_token_accuracy": 0.21809448748826982,
      "num_tokens": 83111579.0,
      "step": 45040
    },
    {
      "entropy": 5.417238664627075,
      "epoch": 3.7843730308758663,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036228375647391175,
      "loss": 4.9571,
      "mean_token_accuracy": 0.20512301921844484,
      "num_tokens": 83120321.0,
      "step": 45045
    },
    {
      "entropy": 5.4771064758300785,
      "epoch": 3.784793110691031,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036225615427311277,
      "loss": 5.0022,
      "mean_token_accuracy": 0.20689105093479157,
      "num_tokens": 83129445.0,
      "step": 45050
    },
    {
      "entropy": 5.431946420669556,
      "epoch": 3.7852131905061963,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003622285505266159,
      "loss": 4.9269,
      "mean_token_accuracy": 0.20818433910608292,
      "num_tokens": 83137876.0,
      "step": 45055
    },
    {
      "entropy": 5.539296865463257,
      "epoch": 3.785633270321361,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036220094523491017,
      "loss": 5.0371,
      "mean_token_accuracy": 0.2021270975470543,
      "num_tokens": 83147651.0,
      "step": 45060
    },
    {
      "entropy": 5.531162452697754,
      "epoch": 3.786053350136526,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036217333839848467,
      "loss": 4.9866,
      "mean_token_accuracy": 0.2010897234082222,
      "num_tokens": 83157339.0,
      "step": 45065
    },
    {
      "entropy": 5.54209532737732,
      "epoch": 3.7864734299516907,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003621457300178282,
      "loss": 5.0286,
      "mean_token_accuracy": 0.20260653644800186,
      "num_tokens": 83167012.0,
      "step": 45070
    },
    {
      "entropy": 5.456747674942017,
      "epoch": 3.7868935097668555,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003621181200934302,
      "loss": 4.913,
      "mean_token_accuracy": 0.21517449617385864,
      "num_tokens": 83175881.0,
      "step": 45075
    },
    {
      "entropy": 5.456646203994751,
      "epoch": 3.7873135895820207,
      "grad_norm": 1.125,
      "learning_rate": 0.0003620905086257795,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20703933537006378,
      "num_tokens": 83184338.0,
      "step": 45080
    },
    {
      "entropy": 5.444443416595459,
      "epoch": 3.7877336693971855,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003620628956153652,
      "loss": 4.8994,
      "mean_token_accuracy": 0.21452607065439225,
      "num_tokens": 83192839.0,
      "step": 45085
    },
    {
      "entropy": 5.404225254058838,
      "epoch": 3.7881537492123503,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036203528106267663,
      "loss": 4.9836,
      "mean_token_accuracy": 0.2091739818453789,
      "num_tokens": 83203481.0,
      "step": 45090
    },
    {
      "entropy": 5.360936975479126,
      "epoch": 3.7885738290275155,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003620076649682028,
      "loss": 4.855,
      "mean_token_accuracy": 0.20982296466827394,
      "num_tokens": 83211524.0,
      "step": 45095
    },
    {
      "entropy": 5.493238925933838,
      "epoch": 3.78899390884268,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003619800473324331,
      "loss": 5.0148,
      "mean_token_accuracy": 0.2033487930893898,
      "num_tokens": 83220473.0,
      "step": 45100
    },
    {
      "entropy": 5.473421049118042,
      "epoch": 3.789413988657845,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003619524281558565,
      "loss": 4.9827,
      "mean_token_accuracy": 0.20614763349294662,
      "num_tokens": 83229164.0,
      "step": 45105
    },
    {
      "entropy": 5.4912152767181395,
      "epoch": 3.78983406847301,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003619248074389625,
      "loss": 5.0473,
      "mean_token_accuracy": 0.20429817289114,
      "num_tokens": 83238416.0,
      "step": 45110
    },
    {
      "entropy": 5.493531608581543,
      "epoch": 3.7902541482881746,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003618971851822402,
      "loss": 4.9632,
      "mean_token_accuracy": 0.20993776470422745,
      "num_tokens": 83247416.0,
      "step": 45115
    },
    {
      "entropy": 5.533056402206421,
      "epoch": 3.79067422810334,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003618695613861791,
      "loss": 5.0604,
      "mean_token_accuracy": 0.1960168719291687,
      "num_tokens": 83257361.0,
      "step": 45120
    },
    {
      "entropy": 5.479019832611084,
      "epoch": 3.7910943079185047,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036184193605126846,
      "loss": 4.9342,
      "mean_token_accuracy": 0.2069164037704468,
      "num_tokens": 83266745.0,
      "step": 45125
    },
    {
      "entropy": 5.498305940628052,
      "epoch": 3.7915143877336694,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003618143091779976,
      "loss": 5.0063,
      "mean_token_accuracy": 0.2058342143893242,
      "num_tokens": 83276179.0,
      "step": 45130
    },
    {
      "entropy": 5.466190576553345,
      "epoch": 3.7919344675488342,
      "grad_norm": 1.03125,
      "learning_rate": 0.000361786680766856,
      "loss": 4.9392,
      "mean_token_accuracy": 0.20609024316072463,
      "num_tokens": 83286029.0,
      "step": 45135
    },
    {
      "entropy": 5.540278911590576,
      "epoch": 3.792354547363999,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000361759050818333,
      "loss": 5.0064,
      "mean_token_accuracy": 0.20611691176891328,
      "num_tokens": 83295325.0,
      "step": 45140
    },
    {
      "entropy": 5.55146484375,
      "epoch": 3.7927746271791642,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036173141933291806,
      "loss": 5.0914,
      "mean_token_accuracy": 0.19313549846410752,
      "num_tokens": 83304545.0,
      "step": 45145
    },
    {
      "entropy": 5.4750748634338375,
      "epoch": 3.793194706994329,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003617037863111008,
      "loss": 4.9327,
      "mean_token_accuracy": 0.20841884166002272,
      "num_tokens": 83314079.0,
      "step": 45150
    },
    {
      "entropy": 5.490170049667358,
      "epoch": 3.793614786809494,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036167615175337046,
      "loss": 4.948,
      "mean_token_accuracy": 0.2049674928188324,
      "num_tokens": 83322734.0,
      "step": 45155
    },
    {
      "entropy": 5.580669593811035,
      "epoch": 3.7940348666246586,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036164851566021684,
      "loss": 5.1713,
      "mean_token_accuracy": 0.19794250279664993,
      "num_tokens": 83332999.0,
      "step": 45160
    },
    {
      "entropy": 5.4213227272033695,
      "epoch": 3.7944549464398234,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003616208780321294,
      "loss": 4.8832,
      "mean_token_accuracy": 0.20897506028413773,
      "num_tokens": 83340734.0,
      "step": 45165
    },
    {
      "entropy": 5.444396448135376,
      "epoch": 3.7948750262549886,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036159323886959766,
      "loss": 4.9577,
      "mean_token_accuracy": 0.2134905532002449,
      "num_tokens": 83350057.0,
      "step": 45170
    },
    {
      "entropy": 5.551660108566284,
      "epoch": 3.7952951060701534,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003615655981731112,
      "loss": 5.008,
      "mean_token_accuracy": 0.21435191631317138,
      "num_tokens": 83359047.0,
      "step": 45175
    },
    {
      "entropy": 5.505976295471191,
      "epoch": 3.795715185885318,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003615379559431598,
      "loss": 4.9208,
      "mean_token_accuracy": 0.21825029700994492,
      "num_tokens": 83367985.0,
      "step": 45180
    },
    {
      "entropy": 5.45642876625061,
      "epoch": 3.796135265700483,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003615103121802331,
      "loss": 5.0063,
      "mean_token_accuracy": 0.20316114872694016,
      "num_tokens": 83377222.0,
      "step": 45185
    },
    {
      "entropy": 5.488523960113525,
      "epoch": 3.7965553455156478,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003614826668848208,
      "loss": 5.1106,
      "mean_token_accuracy": 0.1923283204436302,
      "num_tokens": 83387913.0,
      "step": 45190
    },
    {
      "entropy": 5.514877891540527,
      "epoch": 3.796975425330813,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003614550200574125,
      "loss": 5.0256,
      "mean_token_accuracy": 0.20552651435136796,
      "num_tokens": 83396992.0,
      "step": 45195
    },
    {
      "entropy": 5.514313411712647,
      "epoch": 3.797395505145978,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003614273716984981,
      "loss": 5.0022,
      "mean_token_accuracy": 0.206489197909832,
      "num_tokens": 83407212.0,
      "step": 45200
    },
    {
      "entropy": 5.582879972457886,
      "epoch": 3.7978155849611426,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003613997218085673,
      "loss": 5.1484,
      "mean_token_accuracy": 0.1945384219288826,
      "num_tokens": 83416498.0,
      "step": 45205
    },
    {
      "entropy": 5.51771821975708,
      "epoch": 3.7982356647763074,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003613720703881099,
      "loss": 4.9795,
      "mean_token_accuracy": 0.21023719757795334,
      "num_tokens": 83426278.0,
      "step": 45210
    },
    {
      "entropy": 5.492205858230591,
      "epoch": 3.798655744591472,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036134441743761577,
      "loss": 4.9689,
      "mean_token_accuracy": 0.20753728449344636,
      "num_tokens": 83435660.0,
      "step": 45215
    },
    {
      "entropy": 5.513283395767212,
      "epoch": 3.7990758244066374,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036131676295757476,
      "loss": 5.0854,
      "mean_token_accuracy": 0.20148292779922486,
      "num_tokens": 83444896.0,
      "step": 45220
    },
    {
      "entropy": 5.545602655410766,
      "epoch": 3.799495904221802,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036128910694847665,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20046530812978744,
      "num_tokens": 83454421.0,
      "step": 45225
    },
    {
      "entropy": 5.503959512710571,
      "epoch": 3.799915984036967,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003612614494108115,
      "loss": 4.8944,
      "mean_token_accuracy": 0.21748821884393693,
      "num_tokens": 83463140.0,
      "step": 45230
    },
    {
      "entropy": 5.400499105453491,
      "epoch": 3.8003360638521317,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003612337903450692,
      "loss": 4.9742,
      "mean_token_accuracy": 0.2067912697792053,
      "num_tokens": 83472434.0,
      "step": 45235
    },
    {
      "entropy": 5.440221929550171,
      "epoch": 3.8007561436672965,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003612061297517397,
      "loss": 4.9394,
      "mean_token_accuracy": 0.2046567067503929,
      "num_tokens": 83481049.0,
      "step": 45240
    },
    {
      "entropy": 5.430973672866822,
      "epoch": 3.8011762234824618,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000361178467631313,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21151076555252074,
      "num_tokens": 83490272.0,
      "step": 45245
    },
    {
      "entropy": 5.4692975044250485,
      "epoch": 3.8015963032976265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036115080398427913,
      "loss": 4.9892,
      "mean_token_accuracy": 0.20362784713506699,
      "num_tokens": 83499809.0,
      "step": 45250
    },
    {
      "entropy": 5.45586142539978,
      "epoch": 3.8020163831127913,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003611231388111282,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21573551893234252,
      "num_tokens": 83508306.0,
      "step": 45255
    },
    {
      "entropy": 5.432846355438232,
      "epoch": 3.8024364629279566,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003610954721123501,
      "loss": 4.9654,
      "mean_token_accuracy": 0.2075177773833275,
      "num_tokens": 83517384.0,
      "step": 45260
    },
    {
      "entropy": 5.427408695220947,
      "epoch": 3.8028565427431213,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036106780388843506,
      "loss": 5.0176,
      "mean_token_accuracy": 0.20237655490636824,
      "num_tokens": 83526493.0,
      "step": 45265
    },
    {
      "entropy": 5.6001543521881105,
      "epoch": 3.803276622558286,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003610401341398732,
      "loss": 5.0549,
      "mean_token_accuracy": 0.1975727841258049,
      "num_tokens": 83534972.0,
      "step": 45270
    },
    {
      "entropy": 5.513113355636596,
      "epoch": 3.803696702373451,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003610124628671547,
      "loss": 4.9601,
      "mean_token_accuracy": 0.21305454075336455,
      "num_tokens": 83543468.0,
      "step": 45275
    },
    {
      "entropy": 5.3981953144073485,
      "epoch": 3.8041167821886157,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003609847900707698,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22195508629083632,
      "num_tokens": 83552000.0,
      "step": 45280
    },
    {
      "entropy": 5.419692611694336,
      "epoch": 3.804536862003781,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036095711575120844,
      "loss": 4.9988,
      "mean_token_accuracy": 0.20758650302886963,
      "num_tokens": 83560711.0,
      "step": 45285
    },
    {
      "entropy": 5.434320878982544,
      "epoch": 3.8049569418189457,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036092943990896116,
      "loss": 5.0239,
      "mean_token_accuracy": 0.20058454275131227,
      "num_tokens": 83570223.0,
      "step": 45290
    },
    {
      "entropy": 5.507478952407837,
      "epoch": 3.8053770216341105,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036090176254451793,
      "loss": 4.9942,
      "mean_token_accuracy": 0.20416240245103837,
      "num_tokens": 83579006.0,
      "step": 45295
    },
    {
      "entropy": 5.513075351715088,
      "epoch": 3.8057971014492753,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036087408365836936,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20688366889953613,
      "num_tokens": 83586711.0,
      "step": 45300
    },
    {
      "entropy": 5.519097375869751,
      "epoch": 3.80621718126444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003608464032510056,
      "loss": 5.0428,
      "mean_token_accuracy": 0.19938103556632997,
      "num_tokens": 83596448.0,
      "step": 45305
    },
    {
      "entropy": 5.473328113555908,
      "epoch": 3.8066372610796053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000360818721322917,
      "loss": 5.0131,
      "mean_token_accuracy": 0.20169401913881302,
      "num_tokens": 83605410.0,
      "step": 45310
    },
    {
      "entropy": 5.526211500167847,
      "epoch": 3.80705734089477,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036079103787459393,
      "loss": 5.014,
      "mean_token_accuracy": 0.1993192583322525,
      "num_tokens": 83614582.0,
      "step": 45315
    },
    {
      "entropy": 5.447080898284912,
      "epoch": 3.807477420709935,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003607633529065268,
      "loss": 4.9161,
      "mean_token_accuracy": 0.2089477613568306,
      "num_tokens": 83622695.0,
      "step": 45320
    },
    {
      "entropy": 5.480102300643921,
      "epoch": 3.8078975005250997,
      "grad_norm": 1.125,
      "learning_rate": 0.000360735666419206,
      "loss": 5.0041,
      "mean_token_accuracy": 0.2033126413822174,
      "num_tokens": 83631226.0,
      "step": 45325
    },
    {
      "entropy": 5.593993473052978,
      "epoch": 3.8083175803402645,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036070797841312216,
      "loss": 5.1352,
      "mean_token_accuracy": 0.19658800959587097,
      "num_tokens": 83641319.0,
      "step": 45330
    },
    {
      "entropy": 5.518765544891357,
      "epoch": 3.8087376601554297,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003606802888887655,
      "loss": 5.0665,
      "mean_token_accuracy": 0.20172891318798064,
      "num_tokens": 83650729.0,
      "step": 45335
    },
    {
      "entropy": 5.467458724975586,
      "epoch": 3.8091577399705945,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036065259784662663,
      "loss": 4.8659,
      "mean_token_accuracy": 0.21940167844295502,
      "num_tokens": 83659789.0,
      "step": 45340
    },
    {
      "entropy": 5.480833244323731,
      "epoch": 3.8095778197857593,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036062490528719615,
      "loss": 4.945,
      "mean_token_accuracy": 0.2032385990023613,
      "num_tokens": 83668571.0,
      "step": 45345
    },
    {
      "entropy": 5.467881345748902,
      "epoch": 3.809997899600924,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036059721121096456,
      "loss": 4.8845,
      "mean_token_accuracy": 0.22240130454301835,
      "num_tokens": 83677360.0,
      "step": 45350
    },
    {
      "entropy": 5.420527601242066,
      "epoch": 3.810417979416089,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003605695156184224,
      "loss": 4.8899,
      "mean_token_accuracy": 0.21612813025712968,
      "num_tokens": 83686574.0,
      "step": 45355
    },
    {
      "entropy": 5.381431531906128,
      "epoch": 3.810838059231254,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003605418185100603,
      "loss": 4.97,
      "mean_token_accuracy": 0.20918099135160445,
      "num_tokens": 83695914.0,
      "step": 45360
    },
    {
      "entropy": 5.460322093963623,
      "epoch": 3.811258139046419,
      "grad_norm": 1.171875,
      "learning_rate": 0.000360514119886369,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2092622309923172,
      "num_tokens": 83704466.0,
      "step": 45365
    },
    {
      "entropy": 5.393676996231079,
      "epoch": 3.8116782188615836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036048641974783906,
      "loss": 4.791,
      "mean_token_accuracy": 0.2181530326604843,
      "num_tokens": 83713304.0,
      "step": 45370
    },
    {
      "entropy": 5.438168573379516,
      "epoch": 3.8120982986767484,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036045871809496123,
      "loss": 5.0089,
      "mean_token_accuracy": 0.21224124133586883,
      "num_tokens": 83723162.0,
      "step": 45375
    },
    {
      "entropy": 5.530148601531982,
      "epoch": 3.812518378491913,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003604310149282263,
      "loss": 5.0375,
      "mean_token_accuracy": 0.2117992475628853,
      "num_tokens": 83733407.0,
      "step": 45380
    },
    {
      "entropy": 5.568601989746094,
      "epoch": 3.8129384583070784,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00036040331024812485,
      "loss": 5.0486,
      "mean_token_accuracy": 0.19925611913204194,
      "num_tokens": 83742897.0,
      "step": 45385
    },
    {
      "entropy": 5.387993478775025,
      "epoch": 3.8133585381222432,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036037560405514766,
      "loss": 4.9292,
      "mean_token_accuracy": 0.210793174803257,
      "num_tokens": 83752472.0,
      "step": 45390
    },
    {
      "entropy": 5.4359277248382565,
      "epoch": 3.813778617937408,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003603478963497857,
      "loss": 4.9874,
      "mean_token_accuracy": 0.19957616329193115,
      "num_tokens": 83761881.0,
      "step": 45395
    },
    {
      "entropy": 5.552011728286743,
      "epoch": 3.8141986977525733,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003603201871325297,
      "loss": 4.9994,
      "mean_token_accuracy": 0.20571386069059372,
      "num_tokens": 83770986.0,
      "step": 45400
    },
    {
      "entropy": 5.5075263500213625,
      "epoch": 3.8146187775677376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003602924764038706,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20195199698209762,
      "num_tokens": 83779452.0,
      "step": 45405
    },
    {
      "entropy": 5.483094596862793,
      "epoch": 3.815038857382903,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003602647641642991,
      "loss": 5.0312,
      "mean_token_accuracy": 0.20351902544498443,
      "num_tokens": 83789214.0,
      "step": 45410
    },
    {
      "entropy": 5.563537311553955,
      "epoch": 3.8154589371980676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003602370504143062,
      "loss": 5.0442,
      "mean_token_accuracy": 0.20505354553461075,
      "num_tokens": 83798255.0,
      "step": 45415
    },
    {
      "entropy": 5.550776100158691,
      "epoch": 3.8158790170132324,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003602093351543828,
      "loss": 5.0627,
      "mean_token_accuracy": 0.19620354026556014,
      "num_tokens": 83808020.0,
      "step": 45420
    },
    {
      "entropy": 5.480927419662476,
      "epoch": 3.8162990968283976,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036018161838502,
      "loss": 4.9843,
      "mean_token_accuracy": 0.20612494200468062,
      "num_tokens": 83817640.0,
      "step": 45425
    },
    {
      "entropy": 5.540538215637207,
      "epoch": 3.8167191766435624,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036015390010670856,
      "loss": 5.1028,
      "mean_token_accuracy": 0.2027757003903389,
      "num_tokens": 83827032.0,
      "step": 45430
    },
    {
      "entropy": 5.542880868911743,
      "epoch": 3.817139256458727,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036012618031993975,
      "loss": 4.9845,
      "mean_token_accuracy": 0.21046866327524186,
      "num_tokens": 83836061.0,
      "step": 45435
    },
    {
      "entropy": 5.536638641357422,
      "epoch": 3.817559336273892,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003600984590252045,
      "loss": 5.0108,
      "mean_token_accuracy": 0.20642729550600053,
      "num_tokens": 83844947.0,
      "step": 45440
    },
    {
      "entropy": 5.505188846588135,
      "epoch": 3.8179794160890568,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036007073622299385,
      "loss": 5.074,
      "mean_token_accuracy": 0.20640522688627244,
      "num_tokens": 83854008.0,
      "step": 45445
    },
    {
      "entropy": 5.513962030410767,
      "epoch": 3.818399495904222,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003600430119137988,
      "loss": 4.9451,
      "mean_token_accuracy": 0.211969056725502,
      "num_tokens": 83862552.0,
      "step": 45450
    },
    {
      "entropy": 5.598603248596191,
      "epoch": 3.818819575719387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003600152860981107,
      "loss": 5.1035,
      "mean_token_accuracy": 0.19747521132230758,
      "num_tokens": 83872314.0,
      "step": 45455
    },
    {
      "entropy": 5.428586530685425,
      "epoch": 3.8192396555345516,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035998755877642057,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2110186830163002,
      "num_tokens": 83882366.0,
      "step": 45460
    },
    {
      "entropy": 5.465607070922852,
      "epoch": 3.8196597353497164,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003599598299492196,
      "loss": 4.9432,
      "mean_token_accuracy": 0.2105890229344368,
      "num_tokens": 83891446.0,
      "step": 45465
    },
    {
      "entropy": 5.448253011703491,
      "epoch": 3.820079815164881,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035993209961699904,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2063508927822113,
      "num_tokens": 83900219.0,
      "step": 45470
    },
    {
      "entropy": 5.407076358795166,
      "epoch": 3.8204998949800464,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035990436778024997,
      "loss": 4.8614,
      "mean_token_accuracy": 0.21575928181409837,
      "num_tokens": 83908747.0,
      "step": 45475
    },
    {
      "entropy": 5.520316886901855,
      "epoch": 3.820919974795211,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035987663443946377,
      "loss": 5.0524,
      "mean_token_accuracy": 0.20427699387073517,
      "num_tokens": 83917813.0,
      "step": 45480
    },
    {
      "entropy": 5.547455453872681,
      "epoch": 3.821340054610376,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003598488995951317,
      "loss": 4.9984,
      "mean_token_accuracy": 0.20185241252183914,
      "num_tokens": 83927388.0,
      "step": 45485
    },
    {
      "entropy": 5.506508636474609,
      "epoch": 3.8217601344255407,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003598211632477451,
      "loss": 4.9789,
      "mean_token_accuracy": 0.2048666149377823,
      "num_tokens": 83936340.0,
      "step": 45490
    },
    {
      "entropy": 5.447493219375611,
      "epoch": 3.8221802142407055,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035979342539779524,
      "loss": 5.0234,
      "mean_token_accuracy": 0.21021209210157393,
      "num_tokens": 83945935.0,
      "step": 45495
    },
    {
      "entropy": 5.394006967544556,
      "epoch": 3.8226002940558708,
      "grad_norm": 1.125,
      "learning_rate": 0.0003597656860457735,
      "loss": 4.8839,
      "mean_token_accuracy": 0.21502096056938172,
      "num_tokens": 83954973.0,
      "step": 45500
    },
    {
      "entropy": 5.485113906860351,
      "epoch": 3.8230203738710355,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003597379451921712,
      "loss": 5.0063,
      "mean_token_accuracy": 0.20568332225084304,
      "num_tokens": 83964223.0,
      "step": 45505
    },
    {
      "entropy": 5.44661922454834,
      "epoch": 3.8234404536862003,
      "grad_norm": 1.125,
      "learning_rate": 0.00035971020283747995,
      "loss": 4.9839,
      "mean_token_accuracy": 0.20384721159935,
      "num_tokens": 83973403.0,
      "step": 45510
    },
    {
      "entropy": 5.510671329498291,
      "epoch": 3.823860533501365,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035968245898219107,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20912400484085084,
      "num_tokens": 83982045.0,
      "step": 45515
    },
    {
      "entropy": 5.512541055679321,
      "epoch": 3.82428061331653,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035965471362679605,
      "loss": 5.038,
      "mean_token_accuracy": 0.20020679086446763,
      "num_tokens": 83991243.0,
      "step": 45520
    },
    {
      "entropy": 5.465027666091919,
      "epoch": 3.824700693131695,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003596269667717863,
      "loss": 4.8793,
      "mean_token_accuracy": 0.22106309831142426,
      "num_tokens": 84000333.0,
      "step": 45525
    },
    {
      "entropy": 5.5206298828125,
      "epoch": 3.82512077294686,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003595992184176534,
      "loss": 5.0186,
      "mean_token_accuracy": 0.2047865495085716,
      "num_tokens": 84009294.0,
      "step": 45530
    },
    {
      "entropy": 5.464115810394287,
      "epoch": 3.8255408527620247,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035957146856488903,
      "loss": 4.9543,
      "mean_token_accuracy": 0.20732540935277938,
      "num_tokens": 84019934.0,
      "step": 45535
    },
    {
      "entropy": 5.469796895980835,
      "epoch": 3.82596093257719,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035954371721398453,
      "loss": 4.9947,
      "mean_token_accuracy": 0.20839039236307144,
      "num_tokens": 84028468.0,
      "step": 45540
    },
    {
      "entropy": 5.484241580963134,
      "epoch": 3.8263810123923543,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035951596436543167,
      "loss": 4.9401,
      "mean_token_accuracy": 0.21506357789039612,
      "num_tokens": 84037335.0,
      "step": 45545
    },
    {
      "entropy": 5.441501235961914,
      "epoch": 3.8268010922075195,
      "grad_norm": 1.109375,
      "learning_rate": 0.000359488210019722,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20773301124572754,
      "num_tokens": 84047484.0,
      "step": 45550
    },
    {
      "entropy": 5.391167783737183,
      "epoch": 3.8272211720226843,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003594604541773472,
      "loss": 4.9145,
      "mean_token_accuracy": 0.20812318623065948,
      "num_tokens": 84058115.0,
      "step": 45555
    },
    {
      "entropy": 5.48285322189331,
      "epoch": 3.827641251837849,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035943269683879893,
      "loss": 4.9614,
      "mean_token_accuracy": 0.20212495774030687,
      "num_tokens": 84066524.0,
      "step": 45560
    },
    {
      "entropy": 5.532769727706909,
      "epoch": 3.8280613316530143,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035940493800456884,
      "loss": 5.0407,
      "mean_token_accuracy": 0.2073017969727516,
      "num_tokens": 84076482.0,
      "step": 45565
    },
    {
      "entropy": 5.459160470962525,
      "epoch": 3.828481411468179,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035937717767514887,
      "loss": 4.9713,
      "mean_token_accuracy": 0.2002065122127533,
      "num_tokens": 84085851.0,
      "step": 45570
    },
    {
      "entropy": 5.402078628540039,
      "epoch": 3.828901491283344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035934941585103055,
      "loss": 4.8655,
      "mean_token_accuracy": 0.21752362102270126,
      "num_tokens": 84094981.0,
      "step": 45575
    },
    {
      "entropy": 5.444327068328858,
      "epoch": 3.8293215710985087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035932165253270583,
      "loss": 4.8496,
      "mean_token_accuracy": 0.21415830850601197,
      "num_tokens": 84103745.0,
      "step": 45580
    },
    {
      "entropy": 5.397628545761108,
      "epoch": 3.8297416509136735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035929388772066646,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21159014403820037,
      "num_tokens": 84112428.0,
      "step": 45585
    },
    {
      "entropy": 5.394280052185058,
      "epoch": 3.8301617307288387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003592661214154042,
      "loss": 4.8747,
      "mean_token_accuracy": 0.21409597396850585,
      "num_tokens": 84121529.0,
      "step": 45590
    },
    {
      "entropy": 5.431778001785278,
      "epoch": 3.8305818105440035,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003592383536174111,
      "loss": 4.8956,
      "mean_token_accuracy": 0.2119567096233368,
      "num_tokens": 84130195.0,
      "step": 45595
    },
    {
      "entropy": 5.471905612945557,
      "epoch": 3.8310018903591683,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003592105843271788,
      "loss": 5.0192,
      "mean_token_accuracy": 0.19988992661237717,
      "num_tokens": 84139191.0,
      "step": 45600
    },
    {
      "entropy": 5.5143664360046385,
      "epoch": 3.831421970174333,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035918281354519946,
      "loss": 4.9454,
      "mean_token_accuracy": 0.21341679841279984,
      "num_tokens": 84148572.0,
      "step": 45605
    },
    {
      "entropy": 5.429778289794922,
      "epoch": 3.831842049989498,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003591550412719649,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20235745459795,
      "num_tokens": 84157489.0,
      "step": 45610
    },
    {
      "entropy": 5.509183120727539,
      "epoch": 3.832262129804663,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003591272675079671,
      "loss": 5.0403,
      "mean_token_accuracy": 0.19850964695215226,
      "num_tokens": 84166330.0,
      "step": 45615
    },
    {
      "entropy": 5.438375568389892,
      "epoch": 3.832682209619828,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003590994922536981,
      "loss": 4.9184,
      "mean_token_accuracy": 0.21116819828748704,
      "num_tokens": 84176255.0,
      "step": 45620
    },
    {
      "entropy": 5.462744998931885,
      "epoch": 3.8331022894349926,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003590717155096499,
      "loss": 4.9048,
      "mean_token_accuracy": 0.20908809453248978,
      "num_tokens": 84185860.0,
      "step": 45625
    },
    {
      "entropy": 5.293156957626342,
      "epoch": 3.8335223692501574,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035904393727631453,
      "loss": 4.7903,
      "mean_token_accuracy": 0.22343350648880006,
      "num_tokens": 84195016.0,
      "step": 45630
    },
    {
      "entropy": 5.533899545669556,
      "epoch": 3.8339424490653222,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035901615755418403,
      "loss": 5.1542,
      "mean_token_accuracy": 0.19846926480531693,
      "num_tokens": 84205003.0,
      "step": 45635
    },
    {
      "entropy": 5.527728128433227,
      "epoch": 3.8343625288804875,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003589883763437506,
      "loss": 5.0161,
      "mean_token_accuracy": 0.20708086490631103,
      "num_tokens": 84215256.0,
      "step": 45640
    },
    {
      "entropy": 5.416045904159546,
      "epoch": 3.8347826086956522,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003589605936455064,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21381970196962358,
      "num_tokens": 84224498.0,
      "step": 45645
    },
    {
      "entropy": 5.437884712219239,
      "epoch": 3.835202688510817,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003589328094599435,
      "loss": 4.9324,
      "mean_token_accuracy": 0.213506843149662,
      "num_tokens": 84233288.0,
      "step": 45650
    },
    {
      "entropy": 5.4661448955535885,
      "epoch": 3.835622768325982,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003589050237875541,
      "loss": 4.9278,
      "mean_token_accuracy": 0.2082259863615036,
      "num_tokens": 84242546.0,
      "step": 45655
    },
    {
      "entropy": 5.344748449325562,
      "epoch": 3.8360428481411466,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035887723662883037,
      "loss": 4.7997,
      "mean_token_accuracy": 0.2216522440314293,
      "num_tokens": 84251134.0,
      "step": 45660
    },
    {
      "entropy": 5.432834672927856,
      "epoch": 3.836462927956312,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003588494479842646,
      "loss": 4.9648,
      "mean_token_accuracy": 0.20597554445266725,
      "num_tokens": 84260637.0,
      "step": 45665
    },
    {
      "entropy": 5.482111406326294,
      "epoch": 3.8368830077714766,
      "grad_norm": 1.015625,
      "learning_rate": 0.000358821657854349,
      "loss": 4.9725,
      "mean_token_accuracy": 0.21304788589477539,
      "num_tokens": 84269572.0,
      "step": 45670
    },
    {
      "entropy": 5.424595737457276,
      "epoch": 3.8373030875866414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035879386623957594,
      "loss": 4.9454,
      "mean_token_accuracy": 0.20572424679994583,
      "num_tokens": 84278701.0,
      "step": 45675
    },
    {
      "entropy": 5.496418905258179,
      "epoch": 3.837723167401806,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035876607314043766,
      "loss": 5.0103,
      "mean_token_accuracy": 0.21113006621599198,
      "num_tokens": 84287955.0,
      "step": 45680
    },
    {
      "entropy": 5.513541126251221,
      "epoch": 3.838143247216971,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003587382785574266,
      "loss": 5.0304,
      "mean_token_accuracy": 0.20229301899671553,
      "num_tokens": 84297933.0,
      "step": 45685
    },
    {
      "entropy": 5.43252420425415,
      "epoch": 3.838563327032136,
      "grad_norm": 1.03125,
      "learning_rate": 0.000358710482491035,
      "loss": 5.0016,
      "mean_token_accuracy": 0.20851094871759415,
      "num_tokens": 84306945.0,
      "step": 45690
    },
    {
      "entropy": 5.468151569366455,
      "epoch": 3.838983406847301,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035868268494175526,
      "loss": 5.0204,
      "mean_token_accuracy": 0.19528701305389404,
      "num_tokens": 84316168.0,
      "step": 45695
    },
    {
      "entropy": 5.51449408531189,
      "epoch": 3.839403486662466,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003586548859100798,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20573558658361435,
      "num_tokens": 84325330.0,
      "step": 45700
    },
    {
      "entropy": 5.507280540466309,
      "epoch": 3.839823566477631,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035862708539650127,
      "loss": 5.033,
      "mean_token_accuracy": 0.19885323345661163,
      "num_tokens": 84333957.0,
      "step": 45705
    },
    {
      "entropy": 5.390732669830323,
      "epoch": 3.8402436462927954,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035859928340151187,
      "loss": 4.8872,
      "mean_token_accuracy": 0.2152610346674919,
      "num_tokens": 84343755.0,
      "step": 45710
    },
    {
      "entropy": 5.399772071838379,
      "epoch": 3.8406637261079606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035857147992560427,
      "loss": 4.9538,
      "mean_token_accuracy": 0.2046172007918358,
      "num_tokens": 84353596.0,
      "step": 45715
    },
    {
      "entropy": 5.457753849029541,
      "epoch": 3.8410838059231254,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003585436749692709,
      "loss": 4.9739,
      "mean_token_accuracy": 0.21149407774209977,
      "num_tokens": 84362763.0,
      "step": 45720
    },
    {
      "entropy": 5.522842741012573,
      "epoch": 3.84150388573829,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035851586853300437,
      "loss": 5.0037,
      "mean_token_accuracy": 0.19931488037109374,
      "num_tokens": 84371649.0,
      "step": 45725
    },
    {
      "entropy": 5.425795078277588,
      "epoch": 3.8419239655534554,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003584880606172973,
      "loss": 4.8681,
      "mean_token_accuracy": 0.2167248010635376,
      "num_tokens": 84379875.0,
      "step": 45730
    },
    {
      "entropy": 5.467283630371094,
      "epoch": 3.84234404536862,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003584602512226421,
      "loss": 4.9928,
      "mean_token_accuracy": 0.20523156374692916,
      "num_tokens": 84388804.0,
      "step": 45735
    },
    {
      "entropy": 5.497636079788208,
      "epoch": 3.842764125183785,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003584324403495316,
      "loss": 5.0467,
      "mean_token_accuracy": 0.20941767692565919,
      "num_tokens": 84398236.0,
      "step": 45740
    },
    {
      "entropy": 5.531971073150634,
      "epoch": 3.8431842049989497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035840462799845835,
      "loss": 5.0297,
      "mean_token_accuracy": 0.20330409705638885,
      "num_tokens": 84408315.0,
      "step": 45745
    },
    {
      "entropy": 5.513595247268677,
      "epoch": 3.8436042848141145,
      "grad_norm": 1.125,
      "learning_rate": 0.00035837681416991507,
      "loss": 4.945,
      "mean_token_accuracy": 0.2063590407371521,
      "num_tokens": 84417437.0,
      "step": 45750
    },
    {
      "entropy": 5.471863079071045,
      "epoch": 3.8440243646292798,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003583489988643945,
      "loss": 4.9778,
      "mean_token_accuracy": 0.20947963297367095,
      "num_tokens": 84426003.0,
      "step": 45755
    },
    {
      "entropy": 5.407818412780761,
      "epoch": 3.8444444444444446,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003583211820823893,
      "loss": 4.9758,
      "mean_token_accuracy": 0.21177362650632858,
      "num_tokens": 84436038.0,
      "step": 45760
    },
    {
      "entropy": 5.439835691452027,
      "epoch": 3.8448645242596093,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003582933638243922,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21065943092107772,
      "num_tokens": 84445708.0,
      "step": 45765
    },
    {
      "entropy": 5.492251396179199,
      "epoch": 3.845284604074774,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003582655440908961,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20299653857946395,
      "num_tokens": 84454626.0,
      "step": 45770
    },
    {
      "entropy": 5.433672714233398,
      "epoch": 3.845704683889939,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035823772288239383,
      "loss": 4.9089,
      "mean_token_accuracy": 0.21697840392589568,
      "num_tokens": 84463596.0,
      "step": 45775
    },
    {
      "entropy": 5.5022824764251705,
      "epoch": 3.846124763705104,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035820990019937805,
      "loss": 4.982,
      "mean_token_accuracy": 0.1995992586016655,
      "num_tokens": 84473645.0,
      "step": 45780
    },
    {
      "entropy": 5.406074810028076,
      "epoch": 3.846544843520269,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003581820760423418,
      "loss": 4.9792,
      "mean_token_accuracy": 0.20623093396425246,
      "num_tokens": 84482976.0,
      "step": 45785
    },
    {
      "entropy": 5.4643285274505615,
      "epoch": 3.8469649233354337,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003581542504117779,
      "loss": 5.0754,
      "mean_token_accuracy": 0.19825708270072936,
      "num_tokens": 84492821.0,
      "step": 45790
    },
    {
      "entropy": 5.480861616134644,
      "epoch": 3.8473850031505985,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035812642330817926,
      "loss": 4.9575,
      "mean_token_accuracy": 0.20843838900327682,
      "num_tokens": 84502868.0,
      "step": 45795
    },
    {
      "entropy": 5.533238744735717,
      "epoch": 3.8478050829657633,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003580985947320389,
      "loss": 5.0978,
      "mean_token_accuracy": 0.1923863559961319,
      "num_tokens": 84513682.0,
      "step": 45800
    },
    {
      "entropy": 5.551296138763428,
      "epoch": 3.8482251627809285,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003580707646838496,
      "loss": 5.1195,
      "mean_token_accuracy": 0.20432278513908386,
      "num_tokens": 84523473.0,
      "step": 45805
    },
    {
      "entropy": 5.52145791053772,
      "epoch": 3.8486452425960933,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035804293316410457,
      "loss": 4.9956,
      "mean_token_accuracy": 0.21568486243486404,
      "num_tokens": 84533761.0,
      "step": 45810
    },
    {
      "entropy": 5.477621698379517,
      "epoch": 3.849065322411258,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003580151001732967,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21176396757364274,
      "num_tokens": 84543288.0,
      "step": 45815
    },
    {
      "entropy": 5.407479333877563,
      "epoch": 3.849485402226423,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003579872657119191,
      "loss": 4.9629,
      "mean_token_accuracy": 0.19637032151222228,
      "num_tokens": 84552611.0,
      "step": 45820
    },
    {
      "entropy": 5.5142693519592285,
      "epoch": 3.8499054820415877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035795942978046483,
      "loss": 4.9908,
      "mean_token_accuracy": 0.20575396865606307,
      "num_tokens": 84562202.0,
      "step": 45825
    },
    {
      "entropy": 5.508435153961182,
      "epoch": 3.850325561856753,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003579315923794269,
      "loss": 4.9944,
      "mean_token_accuracy": 0.20434859544038772,
      "num_tokens": 84571601.0,
      "step": 45830
    },
    {
      "entropy": 5.409083414077759,
      "epoch": 3.8507456416719177,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003579037535092985,
      "loss": 4.8832,
      "mean_token_accuracy": 0.20960214287042617,
      "num_tokens": 84580163.0,
      "step": 45835
    },
    {
      "entropy": 5.43950047492981,
      "epoch": 3.8511657214870825,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003578759131705729,
      "loss": 5.0613,
      "mean_token_accuracy": 0.20355124771595,
      "num_tokens": 84589445.0,
      "step": 45840
    },
    {
      "entropy": 5.435727548599243,
      "epoch": 3.8515858013022477,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035784807136374317,
      "loss": 4.9494,
      "mean_token_accuracy": 0.20762757658958436,
      "num_tokens": 84599021.0,
      "step": 45845
    },
    {
      "entropy": 5.450001955032349,
      "epoch": 3.852005881117412,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003578202280893025,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21690785884857178,
      "num_tokens": 84609538.0,
      "step": 45850
    },
    {
      "entropy": 5.466797351837158,
      "epoch": 3.8524259609325773,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035779238334774407,
      "loss": 4.9474,
      "mean_token_accuracy": 0.20790819227695465,
      "num_tokens": 84618930.0,
      "step": 45855
    },
    {
      "entropy": 5.52763090133667,
      "epoch": 3.852846040747742,
      "grad_norm": 1.125,
      "learning_rate": 0.00035776453713956124,
      "loss": 5.0297,
      "mean_token_accuracy": 0.19960333704948424,
      "num_tokens": 84627386.0,
      "step": 45860
    },
    {
      "entropy": 5.416000652313232,
      "epoch": 3.853266120562907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003577366894652473,
      "loss": 4.9023,
      "mean_token_accuracy": 0.215285225212574,
      "num_tokens": 84636891.0,
      "step": 45865
    },
    {
      "entropy": 5.431692457199096,
      "epoch": 3.853686200378072,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003577088403252955,
      "loss": 4.894,
      "mean_token_accuracy": 0.21356598734855653,
      "num_tokens": 84646033.0,
      "step": 45870
    },
    {
      "entropy": 5.497993230819702,
      "epoch": 3.854106280193237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003576809897201991,
      "loss": 5.0342,
      "mean_token_accuracy": 0.20311199575662614,
      "num_tokens": 84655446.0,
      "step": 45875
    },
    {
      "entropy": 5.4350197315216064,
      "epoch": 3.8545263600084017,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003576531376504517,
      "loss": 4.8711,
      "mean_token_accuracy": 0.21902871280908584,
      "num_tokens": 84664174.0,
      "step": 45880
    },
    {
      "entropy": 5.38310399055481,
      "epoch": 3.8549464398235664,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003576252841165464,
      "loss": 4.9183,
      "mean_token_accuracy": 0.2146786242723465,
      "num_tokens": 84672634.0,
      "step": 45885
    },
    {
      "entropy": 5.493039655685425,
      "epoch": 3.8553665196387312,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003575974291189767,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20718649476766587,
      "num_tokens": 84682131.0,
      "step": 45890
    },
    {
      "entropy": 5.560459280014038,
      "epoch": 3.8557865994538965,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003575695726582361,
      "loss": 5.0427,
      "mean_token_accuracy": 0.20666738003492355,
      "num_tokens": 84690600.0,
      "step": 45895
    },
    {
      "entropy": 5.576205730438232,
      "epoch": 3.8562066792690612,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003575417147348181,
      "loss": 5.0625,
      "mean_token_accuracy": 0.20698314011096955,
      "num_tokens": 84699403.0,
      "step": 45900
    },
    {
      "entropy": 5.49653115272522,
      "epoch": 3.856626759084226,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003575138553492161,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20493181496858598,
      "num_tokens": 84708429.0,
      "step": 45905
    },
    {
      "entropy": 5.504670286178589,
      "epoch": 3.857046838899391,
      "grad_norm": 1.125,
      "learning_rate": 0.0003574859945019236,
      "loss": 5.0076,
      "mean_token_accuracy": 0.20168862491846085,
      "num_tokens": 84716947.0,
      "step": 45910
    },
    {
      "entropy": 5.432905769348144,
      "epoch": 3.8574669187145556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035745813219343426,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20607048869132996,
      "num_tokens": 84725439.0,
      "step": 45915
    },
    {
      "entropy": 5.483608293533325,
      "epoch": 3.857886998529721,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035743026842424155,
      "loss": 5.05,
      "mean_token_accuracy": 0.19552904069423677,
      "num_tokens": 84734836.0,
      "step": 45920
    },
    {
      "entropy": 5.439286804199218,
      "epoch": 3.8583070783448856,
      "grad_norm": 1.3125,
      "learning_rate": 0.000357402403194839,
      "loss": 4.9132,
      "mean_token_accuracy": 0.2102233424782753,
      "num_tokens": 84743638.0,
      "step": 45925
    },
    {
      "entropy": 5.583609580993652,
      "epoch": 3.8587271581600504,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003573745365057204,
      "loss": 5.1049,
      "mean_token_accuracy": 0.1984722062945366,
      "num_tokens": 84753117.0,
      "step": 45930
    },
    {
      "entropy": 5.507986545562744,
      "epoch": 3.859147237975215,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003573466683573793,
      "loss": 5.0061,
      "mean_token_accuracy": 0.20421230494976045,
      "num_tokens": 84762818.0,
      "step": 45935
    },
    {
      "entropy": 5.447854042053223,
      "epoch": 3.85956731779038,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003573187987503093,
      "loss": 4.9593,
      "mean_token_accuracy": 0.2082419440150261,
      "num_tokens": 84772464.0,
      "step": 45940
    },
    {
      "entropy": 5.409816837310791,
      "epoch": 3.859987397605545,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035729092768500424,
      "loss": 4.965,
      "mean_token_accuracy": 0.20856286585330963,
      "num_tokens": 84781235.0,
      "step": 45945
    },
    {
      "entropy": 5.433842754364013,
      "epoch": 3.86040747742071,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035726305516195776,
      "loss": 4.9357,
      "mean_token_accuracy": 0.20745178908109665,
      "num_tokens": 84790171.0,
      "step": 45950
    },
    {
      "entropy": 5.451541090011597,
      "epoch": 3.860827557235875,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003572351811816636,
      "loss": 5.0203,
      "mean_token_accuracy": 0.2029075488448143,
      "num_tokens": 84799862.0,
      "step": 45955
    },
    {
      "entropy": 5.52808632850647,
      "epoch": 3.8612476370510396,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003572073057446155,
      "loss": 5.0096,
      "mean_token_accuracy": 0.2043764978647232,
      "num_tokens": 84808473.0,
      "step": 45960
    },
    {
      "entropy": 5.467395973205567,
      "epoch": 3.8616677168662044,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003571794288513074,
      "loss": 5.0112,
      "mean_token_accuracy": 0.20447684377431868,
      "num_tokens": 84818500.0,
      "step": 45965
    },
    {
      "entropy": 5.452766847610474,
      "epoch": 3.8620877966813696,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035715155050223293,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20561027973890306,
      "num_tokens": 84827885.0,
      "step": 45970
    },
    {
      "entropy": 5.524042797088623,
      "epoch": 3.8625078764965344,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003571236706978861,
      "loss": 4.9709,
      "mean_token_accuracy": 0.20545393377542495,
      "num_tokens": 84836582.0,
      "step": 45975
    },
    {
      "entropy": 5.53323221206665,
      "epoch": 3.862927956311699,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003570957894387607,
      "loss": 5.0558,
      "mean_token_accuracy": 0.20150783956050872,
      "num_tokens": 84846020.0,
      "step": 45980
    },
    {
      "entropy": 5.5418397903442385,
      "epoch": 3.863348036126864,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035706790672535076,
      "loss": 5.0443,
      "mean_token_accuracy": 0.19879111498594285,
      "num_tokens": 84855115.0,
      "step": 45985
    },
    {
      "entropy": 5.429363107681274,
      "epoch": 3.8637681159420287,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035704002255815005,
      "loss": 4.8671,
      "mean_token_accuracy": 0.21366912871599197,
      "num_tokens": 84864804.0,
      "step": 45990
    },
    {
      "entropy": 5.404430627822876,
      "epoch": 3.864188195757194,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035701213693765255,
      "loss": 4.883,
      "mean_token_accuracy": 0.2203861728310585,
      "num_tokens": 84873581.0,
      "step": 45995
    },
    {
      "entropy": 5.4988898754119875,
      "epoch": 3.8646082755723588,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003569842498643524,
      "loss": 5.0421,
      "mean_token_accuracy": 0.20689714848995208,
      "num_tokens": 84882749.0,
      "step": 46000
    },
    {
      "entropy": 5.488219833374023,
      "epoch": 3.8650283553875235,
      "grad_norm": 1.0,
      "learning_rate": 0.00035695636133874334,
      "loss": 5.0171,
      "mean_token_accuracy": 0.20502312630414962,
      "num_tokens": 84892193.0,
      "step": 46005
    },
    {
      "entropy": 5.470421552658081,
      "epoch": 3.8654484352026888,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035692847136131956,
      "loss": 4.9742,
      "mean_token_accuracy": 0.21091794818639756,
      "num_tokens": 84901683.0,
      "step": 46010
    },
    {
      "entropy": 5.522643375396728,
      "epoch": 3.8658685150178536,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003569005799325751,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20621045380830766,
      "num_tokens": 84910133.0,
      "step": 46015
    },
    {
      "entropy": 5.470486354827881,
      "epoch": 3.8662885948330183,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035687268705300406,
      "loss": 4.9507,
      "mean_token_accuracy": 0.21065186858177185,
      "num_tokens": 84918606.0,
      "step": 46020
    },
    {
      "entropy": 5.439245128631592,
      "epoch": 3.866708674648183,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035684479272310053,
      "loss": 4.8664,
      "mean_token_accuracy": 0.21114060580730437,
      "num_tokens": 84926949.0,
      "step": 46025
    },
    {
      "entropy": 5.434114456176758,
      "epoch": 3.867128754463348,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003568168969433587,
      "loss": 4.9847,
      "mean_token_accuracy": 0.20514514297246933,
      "num_tokens": 84936494.0,
      "step": 46030
    },
    {
      "entropy": 5.491300487518311,
      "epoch": 3.867548834278513,
      "grad_norm": 1.125,
      "learning_rate": 0.0003567889997142726,
      "loss": 5.0115,
      "mean_token_accuracy": 0.2080220729112625,
      "num_tokens": 84944673.0,
      "step": 46035
    },
    {
      "entropy": 5.469391107559204,
      "epoch": 3.867968914093678,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035676110103633645,
      "loss": 4.9744,
      "mean_token_accuracy": 0.21116062253713608,
      "num_tokens": 84953895.0,
      "step": 46040
    },
    {
      "entropy": 5.454863500595093,
      "epoch": 3.8683889939088427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003567332009100446,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20349213629961013,
      "num_tokens": 84963334.0,
      "step": 46045
    },
    {
      "entropy": 5.483425235748291,
      "epoch": 3.8688090737240075,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035670529933589106,
      "loss": 5.0,
      "mean_token_accuracy": 0.21248696446418763,
      "num_tokens": 84972582.0,
      "step": 46050
    },
    {
      "entropy": 5.600112009048462,
      "epoch": 3.8692291535391723,
      "grad_norm": 1.125,
      "learning_rate": 0.00035667739631437033,
      "loss": 5.1541,
      "mean_token_accuracy": 0.20592549741268157,
      "num_tokens": 84981907.0,
      "step": 46055
    },
    {
      "entropy": 5.403918504714966,
      "epoch": 3.8696492333543375,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003566494918459765,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21597886979579925,
      "num_tokens": 84991252.0,
      "step": 46060
    },
    {
      "entropy": 5.419702434539795,
      "epoch": 3.8700693131695023,
      "grad_norm": 1.125,
      "learning_rate": 0.00035662158593120405,
      "loss": 4.8886,
      "mean_token_accuracy": 0.2171688586473465,
      "num_tokens": 84999808.0,
      "step": 46065
    },
    {
      "entropy": 5.46660852432251,
      "epoch": 3.870489392984667,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035659367857054713,
      "loss": 5.0418,
      "mean_token_accuracy": 0.19700057655572892,
      "num_tokens": 85009827.0,
      "step": 46070
    },
    {
      "entropy": 5.47761082649231,
      "epoch": 3.870909472799832,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003565657697645003,
      "loss": 5.0059,
      "mean_token_accuracy": 0.20587164610624314,
      "num_tokens": 85019493.0,
      "step": 46075
    },
    {
      "entropy": 5.349361705780029,
      "epoch": 3.8713295526149967,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003565378595135578,
      "loss": 4.813,
      "mean_token_accuracy": 0.2221221461892128,
      "num_tokens": 85028405.0,
      "step": 46080
    },
    {
      "entropy": 5.46967625617981,
      "epoch": 3.871749632430162,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035650994781821413,
      "loss": 5.0038,
      "mean_token_accuracy": 0.20998059064149857,
      "num_tokens": 85036810.0,
      "step": 46085
    },
    {
      "entropy": 5.468034839630127,
      "epoch": 3.8721697122453267,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003564820346789637,
      "loss": 4.9601,
      "mean_token_accuracy": 0.20391137152910233,
      "num_tokens": 85046473.0,
      "step": 46090
    },
    {
      "entropy": 5.401295232772827,
      "epoch": 3.8725897920604915,
      "grad_norm": 1.15625,
      "learning_rate": 0.000356454120096301,
      "loss": 4.9223,
      "mean_token_accuracy": 0.20771552622318268,
      "num_tokens": 85055204.0,
      "step": 46095
    },
    {
      "entropy": 5.412056398391724,
      "epoch": 3.8730098718756563,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035642620407072047,
      "loss": 4.8733,
      "mean_token_accuracy": 0.21630623191595078,
      "num_tokens": 85063787.0,
      "step": 46100
    },
    {
      "entropy": 5.4299506664276125,
      "epoch": 3.873429951690821,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035639828660271667,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21527125388383866,
      "num_tokens": 85073426.0,
      "step": 46105
    },
    {
      "entropy": 5.3968870639801025,
      "epoch": 3.8738500315059863,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035637036769278416,
      "loss": 4.8755,
      "mean_token_accuracy": 0.22006455957889556,
      "num_tokens": 85082148.0,
      "step": 46110
    },
    {
      "entropy": 5.521368837356567,
      "epoch": 3.874270111321151,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035634244734141743,
      "loss": 5.0674,
      "mean_token_accuracy": 0.202420474588871,
      "num_tokens": 85091594.0,
      "step": 46115
    },
    {
      "entropy": 5.477326107025147,
      "epoch": 3.874690191136316,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035631452554911116,
      "loss": 5.0141,
      "mean_token_accuracy": 0.2032351866364479,
      "num_tokens": 85100809.0,
      "step": 46120
    },
    {
      "entropy": 5.501087236404419,
      "epoch": 3.8751102709514806,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035628660231635997,
      "loss": 4.982,
      "mean_token_accuracy": 0.20632068067789078,
      "num_tokens": 85109726.0,
      "step": 46125
    },
    {
      "entropy": 5.470291233062744,
      "epoch": 3.8755303507666454,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035625867764365845,
      "loss": 4.961,
      "mean_token_accuracy": 0.21033154129981996,
      "num_tokens": 85118600.0,
      "step": 46130
    },
    {
      "entropy": 5.4690546035766605,
      "epoch": 3.8759504305818107,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035623075153150133,
      "loss": 5.001,
      "mean_token_accuracy": 0.2056131735444069,
      "num_tokens": 85128292.0,
      "step": 46135
    },
    {
      "entropy": 5.552813386917114,
      "epoch": 3.8763705103969754,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035620282398038324,
      "loss": 5.0704,
      "mean_token_accuracy": 0.20240238904953003,
      "num_tokens": 85138272.0,
      "step": 46140
    },
    {
      "entropy": 5.51073145866394,
      "epoch": 3.8767905902121402,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035617489499079893,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2050074204802513,
      "num_tokens": 85147833.0,
      "step": 46145
    },
    {
      "entropy": 5.514710426330566,
      "epoch": 3.8772106700273055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035614696456324314,
      "loss": 5.0879,
      "mean_token_accuracy": 0.19431883245706558,
      "num_tokens": 85157619.0,
      "step": 46150
    },
    {
      "entropy": 5.415058183670044,
      "epoch": 3.87763074984247,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035611903269821074,
      "loss": 4.8958,
      "mean_token_accuracy": 0.2124162197113037,
      "num_tokens": 85166801.0,
      "step": 46155
    },
    {
      "entropy": 5.501477527618408,
      "epoch": 3.878050829657635,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003560910993961964,
      "loss": 5.0257,
      "mean_token_accuracy": 0.20628631711006165,
      "num_tokens": 85175576.0,
      "step": 46160
    },
    {
      "entropy": 5.468538618087768,
      "epoch": 3.8784709094728,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035606316465769497,
      "loss": 4.9164,
      "mean_token_accuracy": 0.20459400564432145,
      "num_tokens": 85184347.0,
      "step": 46165
    },
    {
      "entropy": 5.478347969055176,
      "epoch": 3.8788909892879646,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003560352284832013,
      "loss": 4.9,
      "mean_token_accuracy": 0.2156287685036659,
      "num_tokens": 85193531.0,
      "step": 46170
    },
    {
      "entropy": 5.471164703369141,
      "epoch": 3.87931106910313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035600729087321027,
      "loss": 5.025,
      "mean_token_accuracy": 0.21018706858158112,
      "num_tokens": 85202921.0,
      "step": 46175
    },
    {
      "entropy": 5.535336923599243,
      "epoch": 3.8797311489182946,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035597935182821687,
      "loss": 5.1327,
      "mean_token_accuracy": 0.1973871499300003,
      "num_tokens": 85212242.0,
      "step": 46180
    },
    {
      "entropy": 5.485603904724121,
      "epoch": 3.8801512287334594,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035595141134871596,
      "loss": 4.9625,
      "mean_token_accuracy": 0.2040325105190277,
      "num_tokens": 85222150.0,
      "step": 46185
    },
    {
      "entropy": 5.513502788543701,
      "epoch": 3.880571308548624,
      "grad_norm": 1.125,
      "learning_rate": 0.0003559234694352024,
      "loss": 5.0641,
      "mean_token_accuracy": 0.1981515631079674,
      "num_tokens": 85231184.0,
      "step": 46190
    },
    {
      "entropy": 5.500349140167236,
      "epoch": 3.880991388363789,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035589552608817135,
      "loss": 5.0413,
      "mean_token_accuracy": 0.19998303651809693,
      "num_tokens": 85241015.0,
      "step": 46195
    },
    {
      "entropy": 5.477121734619141,
      "epoch": 3.881411468178954,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035586758130811764,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20529753416776658,
      "num_tokens": 85251256.0,
      "step": 46200
    },
    {
      "entropy": 5.468198251724243,
      "epoch": 3.881831547994119,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035583963509553643,
      "loss": 4.9739,
      "mean_token_accuracy": 0.20576083660125732,
      "num_tokens": 85260214.0,
      "step": 46205
    },
    {
      "entropy": 5.479732036590576,
      "epoch": 3.882251627809284,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003558116874509226,
      "loss": 4.9514,
      "mean_token_accuracy": 0.21007027775049208,
      "num_tokens": 85268643.0,
      "step": 46210
    },
    {
      "entropy": 5.427042388916016,
      "epoch": 3.8826717076244486,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003557837383747714,
      "loss": 4.8614,
      "mean_token_accuracy": 0.21467699855566025,
      "num_tokens": 85277045.0,
      "step": 46215
    },
    {
      "entropy": 5.462422609329224,
      "epoch": 3.8830917874396134,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035575578786757784,
      "loss": 4.9238,
      "mean_token_accuracy": 0.20984500348567964,
      "num_tokens": 85285398.0,
      "step": 46220
    },
    {
      "entropy": 5.321726274490357,
      "epoch": 3.8835118672547786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035572783592983714,
      "loss": 4.8974,
      "mean_token_accuracy": 0.2126734271645546,
      "num_tokens": 85294367.0,
      "step": 46225
    },
    {
      "entropy": 5.492943572998047,
      "epoch": 3.8839319470699434,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003556998825620444,
      "loss": 5.0467,
      "mean_token_accuracy": 0.2072865277528763,
      "num_tokens": 85302530.0,
      "step": 46230
    },
    {
      "entropy": 5.491476154327392,
      "epoch": 3.884352026885108,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035567192776469477,
      "loss": 4.949,
      "mean_token_accuracy": 0.20553590208292008,
      "num_tokens": 85311634.0,
      "step": 46235
    },
    {
      "entropy": 5.485853433609009,
      "epoch": 3.884772106700273,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003556439715382834,
      "loss": 5.0488,
      "mean_token_accuracy": 0.20228105932474136,
      "num_tokens": 85321069.0,
      "step": 46240
    },
    {
      "entropy": 5.39499683380127,
      "epoch": 3.8851921865154377,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035561601388330564,
      "loss": 4.9454,
      "mean_token_accuracy": 0.20915793478488923,
      "num_tokens": 85330129.0,
      "step": 46245
    },
    {
      "entropy": 5.485314750671387,
      "epoch": 3.885612266330603,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003555880548002568,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20689328759908676,
      "num_tokens": 85339341.0,
      "step": 46250
    },
    {
      "entropy": 5.356500673294067,
      "epoch": 3.8860323461457678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000355560094289632,
      "loss": 4.8574,
      "mean_token_accuracy": 0.21669452786445617,
      "num_tokens": 85349094.0,
      "step": 46255
    },
    {
      "entropy": 5.4659552574157715,
      "epoch": 3.8864524259609325,
      "grad_norm": 1.125,
      "learning_rate": 0.0003555321323519266,
      "loss": 4.9713,
      "mean_token_accuracy": 0.20754244178533554,
      "num_tokens": 85358344.0,
      "step": 46260
    },
    {
      "entropy": 5.478508710861206,
      "epoch": 3.8868725057760973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003555041689876359,
      "loss": 4.959,
      "mean_token_accuracy": 0.21071195751428604,
      "num_tokens": 85367803.0,
      "step": 46265
    },
    {
      "entropy": 5.465407991409302,
      "epoch": 3.887292585591262,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003554762041972555,
      "loss": 4.9454,
      "mean_token_accuracy": 0.20805760324001313,
      "num_tokens": 85376476.0,
      "step": 46270
    },
    {
      "entropy": 5.429696893692016,
      "epoch": 3.8877126654064273,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035544823798128037,
      "loss": 4.9276,
      "mean_token_accuracy": 0.21688983589410782,
      "num_tokens": 85384968.0,
      "step": 46275
    },
    {
      "entropy": 5.432178211212158,
      "epoch": 3.888132745221592,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003554202703402062,
      "loss": 4.9323,
      "mean_token_accuracy": 0.2153972640633583,
      "num_tokens": 85393970.0,
      "step": 46280
    },
    {
      "entropy": 5.489073133468628,
      "epoch": 3.888552825036757,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003553923012745284,
      "loss": 5.0577,
      "mean_token_accuracy": 0.20362791270017624,
      "num_tokens": 85404276.0,
      "step": 46285
    },
    {
      "entropy": 5.509395027160645,
      "epoch": 3.8889729048519217,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003553643307847423,
      "loss": 5.0103,
      "mean_token_accuracy": 0.20743401050567628,
      "num_tokens": 85414029.0,
      "step": 46290
    },
    {
      "entropy": 5.46131706237793,
      "epoch": 3.8893929846670865,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003553363588713435,
      "loss": 4.948,
      "mean_token_accuracy": 0.20224383771419524,
      "num_tokens": 85423783.0,
      "step": 46295
    },
    {
      "entropy": 5.443582439422608,
      "epoch": 3.8898130644822517,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035530838553482746,
      "loss": 4.9477,
      "mean_token_accuracy": 0.21392393112182617,
      "num_tokens": 85432665.0,
      "step": 46300
    },
    {
      "entropy": 5.45094928741455,
      "epoch": 3.8902331442974165,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035528041077568975,
      "loss": 4.9456,
      "mean_token_accuracy": 0.20821668207645416,
      "num_tokens": 85442093.0,
      "step": 46305
    },
    {
      "entropy": 5.450428056716919,
      "epoch": 3.8906532241125813,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035525243459442573,
      "loss": 4.9441,
      "mean_token_accuracy": 0.20341835170984268,
      "num_tokens": 85450773.0,
      "step": 46310
    },
    {
      "entropy": 5.504981660842896,
      "epoch": 3.8910733039277465,
      "grad_norm": 1.0,
      "learning_rate": 0.0003552244569915313,
      "loss": 4.9824,
      "mean_token_accuracy": 0.21026832312345506,
      "num_tokens": 85460405.0,
      "step": 46315
    },
    {
      "entropy": 5.535412120819092,
      "epoch": 3.8914933837429113,
      "grad_norm": 1.03125,
      "learning_rate": 0.000355196477967502,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20552974045276642,
      "num_tokens": 85470495.0,
      "step": 46320
    },
    {
      "entropy": 5.446189212799072,
      "epoch": 3.891913463558076,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003551684975228333,
      "loss": 4.971,
      "mean_token_accuracy": 0.20504421144723892,
      "num_tokens": 85480161.0,
      "step": 46325
    },
    {
      "entropy": 5.4348773002624515,
      "epoch": 3.892333543373241,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003551405156580209,
      "loss": 4.9351,
      "mean_token_accuracy": 0.20404609590768813,
      "num_tokens": 85489144.0,
      "step": 46330
    },
    {
      "entropy": 5.579488277435303,
      "epoch": 3.8927536231884057,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035511253237356064,
      "loss": 5.1345,
      "mean_token_accuracy": 0.20105453878641127,
      "num_tokens": 85498853.0,
      "step": 46335
    },
    {
      "entropy": 5.6026276588439945,
      "epoch": 3.893173703003571,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035508454766994805,
      "loss": 5.0417,
      "mean_token_accuracy": 0.20838421881198882,
      "num_tokens": 85507909.0,
      "step": 46340
    },
    {
      "entropy": 5.485991621017456,
      "epoch": 3.8935937828187357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035505656154767904,
      "loss": 5.0166,
      "mean_token_accuracy": 0.2102612718939781,
      "num_tokens": 85516520.0,
      "step": 46345
    },
    {
      "entropy": 5.518639183044433,
      "epoch": 3.8940138626339005,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035502857400724914,
      "loss": 5.0596,
      "mean_token_accuracy": 0.20047266334295272,
      "num_tokens": 85526160.0,
      "step": 46350
    },
    {
      "entropy": 5.566875886917114,
      "epoch": 3.8944339424490653,
      "grad_norm": 1.125,
      "learning_rate": 0.00035500058504915426,
      "loss": 4.9788,
      "mean_token_accuracy": 0.2087848037481308,
      "num_tokens": 85535792.0,
      "step": 46355
    },
    {
      "entropy": 5.558043909072876,
      "epoch": 3.89485402226423,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035497259467389027,
      "loss": 5.0041,
      "mean_token_accuracy": 0.19892854243516922,
      "num_tokens": 85544113.0,
      "step": 46360
    },
    {
      "entropy": 5.51569185256958,
      "epoch": 3.8952741020793953,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003549446028819529,
      "loss": 4.9832,
      "mean_token_accuracy": 0.20966526120901108,
      "num_tokens": 85552943.0,
      "step": 46365
    },
    {
      "entropy": 5.481766605377198,
      "epoch": 3.89569418189456,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035491660967383815,
      "loss": 4.9941,
      "mean_token_accuracy": 0.2077951565384865,
      "num_tokens": 85562114.0,
      "step": 46370
    },
    {
      "entropy": 5.466526365280151,
      "epoch": 3.896114261709725,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003548886150500417,
      "loss": 4.9415,
      "mean_token_accuracy": 0.2054185226559639,
      "num_tokens": 85570473.0,
      "step": 46375
    },
    {
      "entropy": 5.503040313720703,
      "epoch": 3.8965343415248896,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003548606190110596,
      "loss": 4.9466,
      "mean_token_accuracy": 0.2099194958806038,
      "num_tokens": 85578977.0,
      "step": 46380
    },
    {
      "entropy": 5.391363573074341,
      "epoch": 3.8969544213400544,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003548326215573878,
      "loss": 4.8683,
      "mean_token_accuracy": 0.20664979368448258,
      "num_tokens": 85587877.0,
      "step": 46385
    },
    {
      "entropy": 5.445726728439331,
      "epoch": 3.8973745011552197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035480462268952214,
      "loss": 4.9277,
      "mean_token_accuracy": 0.2133335530757904,
      "num_tokens": 85596540.0,
      "step": 46390
    },
    {
      "entropy": 5.412168788909912,
      "epoch": 3.8977945809703844,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035477662240795873,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20981003046035768,
      "num_tokens": 85605803.0,
      "step": 46395
    },
    {
      "entropy": 5.459699726104736,
      "epoch": 3.8982146607855492,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003547486207131935,
      "loss": 4.991,
      "mean_token_accuracy": 0.2059156894683838,
      "num_tokens": 85614675.0,
      "step": 46400
    },
    {
      "entropy": 5.4821107387542725,
      "epoch": 3.898634740600714,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035472061760572263,
      "loss": 4.9536,
      "mean_token_accuracy": 0.21043010652065278,
      "num_tokens": 85623605.0,
      "step": 46405
    },
    {
      "entropy": 5.57649073600769,
      "epoch": 3.899054820415879,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035469261308604193,
      "loss": 5.0874,
      "mean_token_accuracy": 0.19891486763954164,
      "num_tokens": 85632760.0,
      "step": 46410
    },
    {
      "entropy": 5.509580993652344,
      "epoch": 3.899474900231044,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003546646071546477,
      "loss": 4.9713,
      "mean_token_accuracy": 0.2082989141345024,
      "num_tokens": 85641885.0,
      "step": 46415
    },
    {
      "entropy": 5.559713220596313,
      "epoch": 3.899894980046209,
      "grad_norm": 1.09375,
      "learning_rate": 0.000354636599812036,
      "loss": 5.0513,
      "mean_token_accuracy": 0.20355720818042755,
      "num_tokens": 85650447.0,
      "step": 46420
    },
    {
      "entropy": 5.569453573226928,
      "epoch": 3.9003150598613736,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035460859105870287,
      "loss": 5.0431,
      "mean_token_accuracy": 0.2083310678601265,
      "num_tokens": 85660769.0,
      "step": 46425
    },
    {
      "entropy": 5.472656106948852,
      "epoch": 3.9007351396765384,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035458058089514455,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20489879250526427,
      "num_tokens": 85669684.0,
      "step": 46430
    },
    {
      "entropy": 5.479836702346802,
      "epoch": 3.901155219491703,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035455256932185724,
      "loss": 4.9734,
      "mean_token_accuracy": 0.20127388685941697,
      "num_tokens": 85678889.0,
      "step": 46435
    },
    {
      "entropy": 5.465476655960083,
      "epoch": 3.9015752993068684,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035452455633933717,
      "loss": 4.8907,
      "mean_token_accuracy": 0.21283770948648453,
      "num_tokens": 85688190.0,
      "step": 46440
    },
    {
      "entropy": 5.539303731918335,
      "epoch": 3.901995379122033,
      "grad_norm": 1.125,
      "learning_rate": 0.0003544965419480805,
      "loss": 5.0557,
      "mean_token_accuracy": 0.2043405294418335,
      "num_tokens": 85697166.0,
      "step": 46445
    },
    {
      "entropy": 5.46233172416687,
      "epoch": 3.902415458937198,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035446852614858345,
      "loss": 4.9832,
      "mean_token_accuracy": 0.2071922391653061,
      "num_tokens": 85706226.0,
      "step": 46450
    },
    {
      "entropy": 5.5532551288604735,
      "epoch": 3.902835538752363,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035444050894134244,
      "loss": 5.056,
      "mean_token_accuracy": 0.20415450632572174,
      "num_tokens": 85715372.0,
      "step": 46455
    },
    {
      "entropy": 5.489009046554566,
      "epoch": 3.9032556185675276,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035441249032685374,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21077962070703507,
      "num_tokens": 85724183.0,
      "step": 46460
    },
    {
      "entropy": 5.402665710449218,
      "epoch": 3.903675698382693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035438447030561364,
      "loss": 4.9098,
      "mean_token_accuracy": 0.20953487306833268,
      "num_tokens": 85733602.0,
      "step": 46465
    },
    {
      "entropy": 5.380085515975952,
      "epoch": 3.9040957781978576,
      "grad_norm": 1.125,
      "learning_rate": 0.00035435644887811855,
      "loss": 4.7939,
      "mean_token_accuracy": 0.2227449044585228,
      "num_tokens": 85742485.0,
      "step": 46470
    },
    {
      "entropy": 5.475286912918091,
      "epoch": 3.9045158580130224,
      "grad_norm": 1.125,
      "learning_rate": 0.0003543284260448648,
      "loss": 4.9584,
      "mean_token_accuracy": 0.2128822088241577,
      "num_tokens": 85752216.0,
      "step": 46475
    },
    {
      "entropy": 5.479429435729981,
      "epoch": 3.9049359378281876,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003543004018063488,
      "loss": 4.9776,
      "mean_token_accuracy": 0.2092825159430504,
      "num_tokens": 85761578.0,
      "step": 46480
    },
    {
      "entropy": 5.414155673980713,
      "epoch": 3.9053560176433524,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003542723761630671,
      "loss": 4.9089,
      "mean_token_accuracy": 0.21386807709932326,
      "num_tokens": 85771060.0,
      "step": 46485
    },
    {
      "entropy": 5.459785795211792,
      "epoch": 3.905776097458517,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035424434911551604,
      "loss": 4.9924,
      "mean_token_accuracy": 0.20062051117420196,
      "num_tokens": 85780505.0,
      "step": 46490
    },
    {
      "entropy": 5.436279153823852,
      "epoch": 3.906196177273682,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035421632066419216,
      "loss": 4.8894,
      "mean_token_accuracy": 0.2127431258559227,
      "num_tokens": 85790053.0,
      "step": 46495
    },
    {
      "entropy": 5.4500631332397464,
      "epoch": 3.9066162570888467,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003541882908095919,
      "loss": 4.9504,
      "mean_token_accuracy": 0.2062917724251747,
      "num_tokens": 85799492.0,
      "step": 46500
    },
    {
      "entropy": 5.448004245758057,
      "epoch": 3.907036336904012,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003541602595522119,
      "loss": 4.9706,
      "mean_token_accuracy": 0.21518565714359283,
      "num_tokens": 85809156.0,
      "step": 46505
    },
    {
      "entropy": 5.541571187973022,
      "epoch": 3.9074564167191768,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035413222689254864,
      "loss": 5.0581,
      "mean_token_accuracy": 0.20061065554618834,
      "num_tokens": 85818683.0,
      "step": 46510
    },
    {
      "entropy": 5.515251207351684,
      "epoch": 3.9078764965343415,
      "grad_norm": 1.125,
      "learning_rate": 0.0003541041928310987,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21164946556091307,
      "num_tokens": 85827809.0,
      "step": 46515
    },
    {
      "entropy": 5.537328672409058,
      "epoch": 3.9082965763495063,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003540761573683587,
      "loss": 5.016,
      "mean_token_accuracy": 0.20407619923353196,
      "num_tokens": 85836713.0,
      "step": 46520
    },
    {
      "entropy": 5.531600475311279,
      "epoch": 3.908716656164671,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035404812050482534,
      "loss": 5.0207,
      "mean_token_accuracy": 0.2131367489695549,
      "num_tokens": 85845422.0,
      "step": 46525
    },
    {
      "entropy": 5.514397859573364,
      "epoch": 3.9091367359798364,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003540200822409952,
      "loss": 4.9493,
      "mean_token_accuracy": 0.21537592113018036,
      "num_tokens": 85854493.0,
      "step": 46530
    },
    {
      "entropy": 5.38711462020874,
      "epoch": 3.909556815795001,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035399204257736505,
      "loss": 4.9123,
      "mean_token_accuracy": 0.2137945219874382,
      "num_tokens": 85863599.0,
      "step": 46535
    },
    {
      "entropy": 5.488154029846191,
      "epoch": 3.909976895610166,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035396400151443155,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20583684146404266,
      "num_tokens": 85872988.0,
      "step": 46540
    },
    {
      "entropy": 5.5540886402130125,
      "epoch": 3.9103969754253307,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003539359590526913,
      "loss": 5.0555,
      "mean_token_accuracy": 0.20333480536937715,
      "num_tokens": 85881503.0,
      "step": 46545
    },
    {
      "entropy": 5.440164279937744,
      "epoch": 3.9108170552404955,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003539079151926413,
      "loss": 4.8696,
      "mean_token_accuracy": 0.22320139557123184,
      "num_tokens": 85889801.0,
      "step": 46550
    },
    {
      "entropy": 5.366296911239624,
      "epoch": 3.9112371350556607,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003538798699347781,
      "loss": 4.9853,
      "mean_token_accuracy": 0.2034000039100647,
      "num_tokens": 85899514.0,
      "step": 46555
    },
    {
      "entropy": 5.477692365646362,
      "epoch": 3.9116572148708255,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035385182327959876,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21468746066093444,
      "num_tokens": 85908698.0,
      "step": 46560
    },
    {
      "entropy": 5.657209348678589,
      "epoch": 3.9120772946859903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035382377522759985,
      "loss": 5.0615,
      "mean_token_accuracy": 0.20698479562997818,
      "num_tokens": 85918261.0,
      "step": 46565
    },
    {
      "entropy": 5.567676591873169,
      "epoch": 3.912497374501155,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035379572577927843,
      "loss": 4.9785,
      "mean_token_accuracy": 0.20806486010551453,
      "num_tokens": 85926910.0,
      "step": 46570
    },
    {
      "entropy": 5.4907618999481205,
      "epoch": 3.91291745431632,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035376767493513133,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2105403184890747,
      "num_tokens": 85936991.0,
      "step": 46575
    },
    {
      "entropy": 5.504250764846802,
      "epoch": 3.913337534131485,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003537396226956553,
      "loss": 5.0356,
      "mean_token_accuracy": 0.2105438306927681,
      "num_tokens": 85947731.0,
      "step": 46580
    },
    {
      "entropy": 5.414878559112549,
      "epoch": 3.91375761394665,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035371156906134746,
      "loss": 4.906,
      "mean_token_accuracy": 0.21396509408950806,
      "num_tokens": 85956539.0,
      "step": 46585
    },
    {
      "entropy": 5.54533166885376,
      "epoch": 3.9141776937618147,
      "grad_norm": 1.25,
      "learning_rate": 0.00035368351403270473,
      "loss": 5.0188,
      "mean_token_accuracy": 0.20603854209184647,
      "num_tokens": 85965033.0,
      "step": 46590
    },
    {
      "entropy": 5.49982476234436,
      "epoch": 3.9145977735769795,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035365545761022405,
      "loss": 4.9716,
      "mean_token_accuracy": 0.2133754625916481,
      "num_tokens": 85973680.0,
      "step": 46595
    },
    {
      "entropy": 5.538028764724731,
      "epoch": 3.9150178533921443,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035362739979440247,
      "loss": 5.0448,
      "mean_token_accuracy": 0.20091884434223176,
      "num_tokens": 85983137.0,
      "step": 46600
    },
    {
      "entropy": 5.487932395935059,
      "epoch": 3.9154379332073095,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035359934058573693,
      "loss": 4.9435,
      "mean_token_accuracy": 0.2110178455710411,
      "num_tokens": 85991898.0,
      "step": 46605
    },
    {
      "entropy": 5.487998676300049,
      "epoch": 3.9158580130224743,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035357127998472454,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21072068363428115,
      "num_tokens": 86002313.0,
      "step": 46610
    },
    {
      "entropy": 5.438195991516113,
      "epoch": 3.916278092837639,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035354321799186246,
      "loss": 4.9472,
      "mean_token_accuracy": 0.2113578051328659,
      "num_tokens": 86011004.0,
      "step": 46615
    },
    {
      "entropy": 5.494868040084839,
      "epoch": 3.9166981726528043,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035351515460764764,
      "loss": 4.9184,
      "mean_token_accuracy": 0.20874702632427217,
      "num_tokens": 86020258.0,
      "step": 46620
    },
    {
      "entropy": 5.496315670013428,
      "epoch": 3.917118252467969,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003534870898325774,
      "loss": 4.9789,
      "mean_token_accuracy": 0.20758676081895827,
      "num_tokens": 86029235.0,
      "step": 46625
    },
    {
      "entropy": 5.491160917282104,
      "epoch": 3.917538332283134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003534590236671487,
      "loss": 4.9965,
      "mean_token_accuracy": 0.2127762645483017,
      "num_tokens": 86038548.0,
      "step": 46630
    },
    {
      "entropy": 5.485054779052734,
      "epoch": 3.9179584120982986,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035343095611185875,
      "loss": 4.8813,
      "mean_token_accuracy": 0.21811001151800155,
      "num_tokens": 86046842.0,
      "step": 46635
    },
    {
      "entropy": 5.427455615997315,
      "epoch": 3.9183784919134634,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003534028871672049,
      "loss": 4.8884,
      "mean_token_accuracy": 0.20867461413145066,
      "num_tokens": 86056009.0,
      "step": 46640
    },
    {
      "entropy": 5.446566486358643,
      "epoch": 3.9187985717286287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035337481683368424,
      "loss": 4.9545,
      "mean_token_accuracy": 0.2048921212553978,
      "num_tokens": 86065870.0,
      "step": 46645
    },
    {
      "entropy": 5.569565725326538,
      "epoch": 3.9192186515437935,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003533467451117939,
      "loss": 5.1556,
      "mean_token_accuracy": 0.18881577402353286,
      "num_tokens": 86075534.0,
      "step": 46650
    },
    {
      "entropy": 5.518320083618164,
      "epoch": 3.9196387313589582,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003533186720020315,
      "loss": 4.9877,
      "mean_token_accuracy": 0.20722126215696335,
      "num_tokens": 86084663.0,
      "step": 46655
    },
    {
      "entropy": 5.478625059127808,
      "epoch": 3.920058811174123,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035329059750489406,
      "loss": 5.0031,
      "mean_token_accuracy": 0.2067624270915985,
      "num_tokens": 86094588.0,
      "step": 46660
    },
    {
      "entropy": 5.525388622283936,
      "epoch": 3.920478890989288,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035326252162087905,
      "loss": 5.0522,
      "mean_token_accuracy": 0.20164544582366944,
      "num_tokens": 86105354.0,
      "step": 46665
    },
    {
      "entropy": 5.485063219070435,
      "epoch": 3.920898970804453,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003532344443504838,
      "loss": 4.9592,
      "mean_token_accuracy": 0.20526355504989624,
      "num_tokens": 86114413.0,
      "step": 46670
    },
    {
      "entropy": 5.493837022781372,
      "epoch": 3.921319050619618,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00035320636569420556,
      "loss": 4.9453,
      "mean_token_accuracy": 0.2073501691222191,
      "num_tokens": 86124315.0,
      "step": 46675
    },
    {
      "entropy": 5.463754177093506,
      "epoch": 3.9217391304347826,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003531782856525419,
      "loss": 4.987,
      "mean_token_accuracy": 0.2072965145111084,
      "num_tokens": 86135025.0,
      "step": 46680
    },
    {
      "entropy": 5.545755243301391,
      "epoch": 3.9221592102499474,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035315020422599017,
      "loss": 5.1058,
      "mean_token_accuracy": 0.19516551345586777,
      "num_tokens": 86144412.0,
      "step": 46685
    },
    {
      "entropy": 5.52220983505249,
      "epoch": 3.922579290065112,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003531221214150478,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2190067857503891,
      "num_tokens": 86153363.0,
      "step": 46690
    },
    {
      "entropy": 5.466607189178466,
      "epoch": 3.9229993698802774,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035309403722021224,
      "loss": 4.9239,
      "mean_token_accuracy": 0.20557209998369216,
      "num_tokens": 86161801.0,
      "step": 46695
    },
    {
      "entropy": 5.4466956615447994,
      "epoch": 3.923419449695442,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003530659516419811,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20848296582698822,
      "num_tokens": 86170350.0,
      "step": 46700
    },
    {
      "entropy": 5.51560606956482,
      "epoch": 3.923839529510607,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003530378646808517,
      "loss": 5.0244,
      "mean_token_accuracy": 0.2121352657675743,
      "num_tokens": 86179417.0,
      "step": 46705
    },
    {
      "entropy": 5.4792663097381595,
      "epoch": 3.924259609325772,
      "grad_norm": 1.0,
      "learning_rate": 0.0003530097763373218,
      "loss": 4.9674,
      "mean_token_accuracy": 0.21109657883644103,
      "num_tokens": 86188328.0,
      "step": 46710
    },
    {
      "entropy": 5.4547961235046385,
      "epoch": 3.9246796891409366,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035298168661188886,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2148043096065521,
      "num_tokens": 86198284.0,
      "step": 46715
    },
    {
      "entropy": 5.427550458908081,
      "epoch": 3.925099768956102,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035295359550505047,
      "loss": 4.9097,
      "mean_token_accuracy": 0.21311294585466384,
      "num_tokens": 86206789.0,
      "step": 46720
    },
    {
      "entropy": 5.58595838546753,
      "epoch": 3.9255198487712666,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003529255030173043,
      "loss": 5.0922,
      "mean_token_accuracy": 0.19424692690372466,
      "num_tokens": 86216912.0,
      "step": 46725
    },
    {
      "entropy": 5.510176134109497,
      "epoch": 3.9259399285864314,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035289740914914796,
      "loss": 4.9654,
      "mean_token_accuracy": 0.20191288739442825,
      "num_tokens": 86226284.0,
      "step": 46730
    },
    {
      "entropy": 5.4963068008422855,
      "epoch": 3.926360008401596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003528693139010792,
      "loss": 5.0036,
      "mean_token_accuracy": 0.21071763932704926,
      "num_tokens": 86235865.0,
      "step": 46735
    },
    {
      "entropy": 5.45797290802002,
      "epoch": 3.926780088216761,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035284121727359555,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21413241624832152,
      "num_tokens": 86244912.0,
      "step": 46740
    },
    {
      "entropy": 5.510669469833374,
      "epoch": 3.927200168031926,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035281311926719486,
      "loss": 5.0147,
      "mean_token_accuracy": 0.20471911132335663,
      "num_tokens": 86254671.0,
      "step": 46745
    },
    {
      "entropy": 5.598507976531982,
      "epoch": 3.927620247847091,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035278501988237476,
      "loss": 5.0298,
      "mean_token_accuracy": 0.2032340422272682,
      "num_tokens": 86263417.0,
      "step": 46750
    },
    {
      "entropy": 5.484240341186523,
      "epoch": 3.9280403276622557,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003527569191196332,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21345054507255554,
      "num_tokens": 86271494.0,
      "step": 46755
    },
    {
      "entropy": 5.410030937194824,
      "epoch": 3.928460407477421,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003527288169794678,
      "loss": 4.9171,
      "mean_token_accuracy": 0.21099962294101715,
      "num_tokens": 86280420.0,
      "step": 46760
    },
    {
      "entropy": 5.430241918563842,
      "epoch": 3.9288804872925853,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003527007134623764,
      "loss": 4.8952,
      "mean_token_accuracy": 0.21794813126325607,
      "num_tokens": 86289456.0,
      "step": 46765
    },
    {
      "entropy": 5.420575904846191,
      "epoch": 3.9293005671077506,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035267260856885693,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21505170166492463,
      "num_tokens": 86299214.0,
      "step": 46770
    },
    {
      "entropy": 5.441368246078492,
      "epoch": 3.9297206469229153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003526445022994072,
      "loss": 4.9387,
      "mean_token_accuracy": 0.2113505095243454,
      "num_tokens": 86308198.0,
      "step": 46775
    },
    {
      "entropy": 5.485731220245361,
      "epoch": 3.93014072673808,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035261639465452503,
      "loss": 5.0265,
      "mean_token_accuracy": 0.20308487117290497,
      "num_tokens": 86317245.0,
      "step": 46780
    },
    {
      "entropy": 5.4054283618927,
      "epoch": 3.9305608065532454,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003525882856347084,
      "loss": 4.8824,
      "mean_token_accuracy": 0.21769367307424545,
      "num_tokens": 86325895.0,
      "step": 46785
    },
    {
      "entropy": 5.37275710105896,
      "epoch": 3.93098088636841,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035256017524045526,
      "loss": 4.8693,
      "mean_token_accuracy": 0.22045747190713882,
      "num_tokens": 86335660.0,
      "step": 46790
    },
    {
      "entropy": 5.630298185348511,
      "epoch": 3.931400966183575,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003525320634722636,
      "loss": 5.1118,
      "mean_token_accuracy": 0.20289239585399627,
      "num_tokens": 86344888.0,
      "step": 46795
    },
    {
      "entropy": 5.471981334686279,
      "epoch": 3.9318210459987397,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035250395033063124,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2056514322757721,
      "num_tokens": 86355189.0,
      "step": 46800
    },
    {
      "entropy": 5.530629682540893,
      "epoch": 3.9322411258139045,
      "grad_norm": 0.96875,
      "learning_rate": 0.00035247583581605644,
      "loss": 5.012,
      "mean_token_accuracy": 0.2053530901670456,
      "num_tokens": 86364798.0,
      "step": 46805
    },
    {
      "entropy": 5.454830312728882,
      "epoch": 3.9326612056290697,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035244771992903705,
      "loss": 4.8505,
      "mean_token_accuracy": 0.21227440387010574,
      "num_tokens": 86373571.0,
      "step": 46810
    },
    {
      "entropy": 5.479228734970093,
      "epoch": 3.9330812854442345,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035241960267007116,
      "loss": 5.0479,
      "mean_token_accuracy": 0.1997309774160385,
      "num_tokens": 86382954.0,
      "step": 46815
    },
    {
      "entropy": 5.453389644622803,
      "epoch": 3.9335013652593993,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003523914840396569,
      "loss": 4.9831,
      "mean_token_accuracy": 0.20999592244625093,
      "num_tokens": 86392205.0,
      "step": 46820
    },
    {
      "entropy": 5.385837841033935,
      "epoch": 3.933921445074564,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035236336403829236,
      "loss": 4.9302,
      "mean_token_accuracy": 0.219074085354805,
      "num_tokens": 86401440.0,
      "step": 46825
    },
    {
      "entropy": 5.548935651779175,
      "epoch": 3.934341524889729,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003523352426664757,
      "loss": 5.006,
      "mean_token_accuracy": 0.20849962681531906,
      "num_tokens": 86411271.0,
      "step": 46830
    },
    {
      "entropy": 5.475886631011963,
      "epoch": 3.934761604704894,
      "grad_norm": 1.046875,
      "learning_rate": 0.000352307119924705,
      "loss": 4.8968,
      "mean_token_accuracy": 0.20910808593034744,
      "num_tokens": 86420879.0,
      "step": 46835
    },
    {
      "entropy": 5.392709159851075,
      "epoch": 3.935181684520059,
      "grad_norm": 1.0,
      "learning_rate": 0.0003522789958134784,
      "loss": 4.8743,
      "mean_token_accuracy": 0.21746811419725418,
      "num_tokens": 86429880.0,
      "step": 46840
    },
    {
      "entropy": 5.484234428405761,
      "epoch": 3.9356017643352237,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003522508703332943,
      "loss": 5.0453,
      "mean_token_accuracy": 0.19959573447704315,
      "num_tokens": 86439184.0,
      "step": 46845
    },
    {
      "entropy": 5.416522455215454,
      "epoch": 3.9360218441503885,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003522227434846507,
      "loss": 4.8589,
      "mean_token_accuracy": 0.21779343634843826,
      "num_tokens": 86447405.0,
      "step": 46850
    },
    {
      "entropy": 5.482292938232422,
      "epoch": 3.9364419239655533,
      "grad_norm": 1.125,
      "learning_rate": 0.00035219461526804605,
      "loss": 4.9909,
      "mean_token_accuracy": 0.21433460265398024,
      "num_tokens": 86456503.0,
      "step": 46855
    },
    {
      "entropy": 5.495529985427856,
      "epoch": 3.9368620037807185,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035216648568397844,
      "loss": 4.9202,
      "mean_token_accuracy": 0.20931622684001921,
      "num_tokens": 86465219.0,
      "step": 46860
    },
    {
      "entropy": 5.511257696151733,
      "epoch": 3.9372820835958833,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035213835473294635,
      "loss": 4.9902,
      "mean_token_accuracy": 0.20556780099868774,
      "num_tokens": 86473853.0,
      "step": 46865
    },
    {
      "entropy": 5.424131298065186,
      "epoch": 3.937702163411048,
      "grad_norm": 1.078125,
      "learning_rate": 0.000352110222415448,
      "loss": 4.946,
      "mean_token_accuracy": 0.20938239842653275,
      "num_tokens": 86482969.0,
      "step": 46870
    },
    {
      "entropy": 5.5683228969573975,
      "epoch": 3.938122243226213,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003520820887319818,
      "loss": 5.1036,
      "mean_token_accuracy": 0.20390065908432006,
      "num_tokens": 86492804.0,
      "step": 46875
    },
    {
      "entropy": 5.537280082702637,
      "epoch": 3.9385423230413776,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003520539536830462,
      "loss": 4.9871,
      "mean_token_accuracy": 0.2011071264743805,
      "num_tokens": 86501644.0,
      "step": 46880
    },
    {
      "entropy": 5.507141256332398,
      "epoch": 3.938962402856543,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003520258172691393,
      "loss": 4.97,
      "mean_token_accuracy": 0.21245099157094954,
      "num_tokens": 86511094.0,
      "step": 46885
    },
    {
      "entropy": 5.434385347366333,
      "epoch": 3.9393824826717077,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003519976794907597,
      "loss": 4.8141,
      "mean_token_accuracy": 0.2174760952591896,
      "num_tokens": 86519053.0,
      "step": 46890
    },
    {
      "entropy": 5.343194961547852,
      "epoch": 3.9398025624868724,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035196954034840593,
      "loss": 4.9467,
      "mean_token_accuracy": 0.20477036386728287,
      "num_tokens": 86528272.0,
      "step": 46895
    },
    {
      "entropy": 5.436627578735352,
      "epoch": 3.9402226423020372,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003519413998425764,
      "loss": 4.9671,
      "mean_token_accuracy": 0.20658211559057235,
      "num_tokens": 86537691.0,
      "step": 46900
    },
    {
      "entropy": 5.523180627822876,
      "epoch": 3.940642722117202,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003519132579737696,
      "loss": 5.01,
      "mean_token_accuracy": 0.2059602588415146,
      "num_tokens": 86547125.0,
      "step": 46905
    },
    {
      "entropy": 5.486487007141113,
      "epoch": 3.9410628019323672,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00035188511474248404,
      "loss": 4.9778,
      "mean_token_accuracy": 0.2034011095762253,
      "num_tokens": 86556849.0,
      "step": 46910
    },
    {
      "entropy": 5.518333768844604,
      "epoch": 3.941482881747532,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003518569701492182,
      "loss": 5.0136,
      "mean_token_accuracy": 0.2103480577468872,
      "num_tokens": 86566290.0,
      "step": 46915
    },
    {
      "entropy": 5.547968530654908,
      "epoch": 3.941902961562697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003518288241944707,
      "loss": 5.0763,
      "mean_token_accuracy": 0.20185637027025222,
      "num_tokens": 86575268.0,
      "step": 46920
    },
    {
      "entropy": 5.528519582748413,
      "epoch": 3.942323041377862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035180067687874016,
      "loss": 5.0027,
      "mean_token_accuracy": 0.20844482630491257,
      "num_tokens": 86584651.0,
      "step": 46925
    },
    {
      "entropy": 5.5036862850189205,
      "epoch": 3.942743121193027,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035177252820252526,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20943604409694672,
      "num_tokens": 86595344.0,
      "step": 46930
    },
    {
      "entropy": 5.49790244102478,
      "epoch": 3.9431632010081916,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003517443781663245,
      "loss": 4.8751,
      "mean_token_accuracy": 0.21614048928022384,
      "num_tokens": 86604341.0,
      "step": 46935
    },
    {
      "entropy": 5.435718631744384,
      "epoch": 3.9435832808233564,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003517162267706366,
      "loss": 4.9069,
      "mean_token_accuracy": 0.21091383397579194,
      "num_tokens": 86613308.0,
      "step": 46940
    },
    {
      "entropy": 5.4267542362213135,
      "epoch": 3.944003360638521,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003516880740159602,
      "loss": 4.9518,
      "mean_token_accuracy": 0.2151872918009758,
      "num_tokens": 86621890.0,
      "step": 46945
    },
    {
      "entropy": 5.468363428115845,
      "epoch": 3.9444234404536864,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003516599199027941,
      "loss": 5.0411,
      "mean_token_accuracy": 0.20536495745182037,
      "num_tokens": 86632032.0,
      "step": 46950
    },
    {
      "entropy": 5.403777742385865,
      "epoch": 3.944843520268851,
      "grad_norm": 1.03125,
      "learning_rate": 0.000351631764431637,
      "loss": 4.8343,
      "mean_token_accuracy": 0.21666285991668702,
      "num_tokens": 86641053.0,
      "step": 46955
    },
    {
      "entropy": 5.509826278686523,
      "epoch": 3.945263600084016,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035160360760298765,
      "loss": 4.9507,
      "mean_token_accuracy": 0.21322753131389618,
      "num_tokens": 86650682.0,
      "step": 46960
    },
    {
      "entropy": 5.461205053329468,
      "epoch": 3.945683679899181,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035157544941734483,
      "loss": 4.9738,
      "mean_token_accuracy": 0.21338551342487336,
      "num_tokens": 86660646.0,
      "step": 46965
    },
    {
      "entropy": 5.477558422088623,
      "epoch": 3.9461037597143456,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035154728987520735,
      "loss": 4.9773,
      "mean_token_accuracy": 0.21150304824113847,
      "num_tokens": 86669692.0,
      "step": 46970
    },
    {
      "entropy": 5.400142621994019,
      "epoch": 3.946523839529511,
      "grad_norm": 1.125,
      "learning_rate": 0.00035151912897707404,
      "loss": 4.8554,
      "mean_token_accuracy": 0.2223715901374817,
      "num_tokens": 86678593.0,
      "step": 46975
    },
    {
      "entropy": 5.49374623298645,
      "epoch": 3.9469439193446756,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035149096672344385,
      "loss": 5.0595,
      "mean_token_accuracy": 0.20099097788333892,
      "num_tokens": 86688180.0,
      "step": 46980
    },
    {
      "entropy": 5.533324718475342,
      "epoch": 3.9473639991598404,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00035146280311481535,
      "loss": 5.0373,
      "mean_token_accuracy": 0.2000677540898323,
      "num_tokens": 86698061.0,
      "step": 46985
    },
    {
      "entropy": 5.4538108825683596,
      "epoch": 3.947784078975005,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003514346381516879,
      "loss": 4.8816,
      "mean_token_accuracy": 0.21276432871818543,
      "num_tokens": 86707537.0,
      "step": 46990
    },
    {
      "entropy": 5.494065141677856,
      "epoch": 3.94820415879017,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035140647183456005,
      "loss": 4.9283,
      "mean_token_accuracy": 0.20263529419898987,
      "num_tokens": 86716426.0,
      "step": 46995
    },
    {
      "entropy": 5.486700582504272,
      "epoch": 3.948624238605335,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003513783041639309,
      "loss": 4.8927,
      "mean_token_accuracy": 0.20905016213655472,
      "num_tokens": 86725830.0,
      "step": 47000
    },
    {
      "entropy": 5.449172019958496,
      "epoch": 3.9490443184205,
      "grad_norm": 1.125,
      "learning_rate": 0.0003513501351402995,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21017161905765533,
      "num_tokens": 86734891.0,
      "step": 47005
    },
    {
      "entropy": 5.401081514358521,
      "epoch": 3.9494643982356648,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003513219647641648,
      "loss": 4.9401,
      "mean_token_accuracy": 0.2133332222700119,
      "num_tokens": 86744215.0,
      "step": 47010
    },
    {
      "entropy": 5.4586015224456785,
      "epoch": 3.9498844780508295,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035129379303602565,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2151724085211754,
      "num_tokens": 86752337.0,
      "step": 47015
    },
    {
      "entropy": 5.381815814971924,
      "epoch": 3.9503045578659943,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035126561995638134,
      "loss": 4.8712,
      "mean_token_accuracy": 0.21858957558870315,
      "num_tokens": 86760507.0,
      "step": 47020
    },
    {
      "entropy": 5.459724187850952,
      "epoch": 3.9507246376811596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035123744552573085,
      "loss": 4.9777,
      "mean_token_accuracy": 0.20323836505413057,
      "num_tokens": 86770281.0,
      "step": 47025
    },
    {
      "entropy": 5.454979467391968,
      "epoch": 3.9511447174963243,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035120926974457325,
      "loss": 4.9296,
      "mean_token_accuracy": 0.2121742218732834,
      "num_tokens": 86779974.0,
      "step": 47030
    },
    {
      "entropy": 5.537815904617309,
      "epoch": 3.951564797311489,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035118109261340766,
      "loss": 5.1043,
      "mean_token_accuracy": 0.19733913391828536,
      "num_tokens": 86790187.0,
      "step": 47035
    },
    {
      "entropy": 5.555784416198731,
      "epoch": 3.951984877126654,
      "grad_norm": 1.25,
      "learning_rate": 0.00035115291413273324,
      "loss": 4.9827,
      "mean_token_accuracy": 0.21244454234838486,
      "num_tokens": 86799524.0,
      "step": 47040
    },
    {
      "entropy": 5.441116046905518,
      "epoch": 3.9524049569418187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003511247343030492,
      "loss": 4.9096,
      "mean_token_accuracy": 0.22012429386377336,
      "num_tokens": 86808387.0,
      "step": 47045
    },
    {
      "entropy": 5.463719272613526,
      "epoch": 3.952825036756984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003510965531248547,
      "loss": 4.9594,
      "mean_token_accuracy": 0.21061501950025557,
      "num_tokens": 86817994.0,
      "step": 47050
    },
    {
      "entropy": 5.48538031578064,
      "epoch": 3.9532451165721487,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035106837059864884,
      "loss": 5.0684,
      "mean_token_accuracy": 0.20758747905492783,
      "num_tokens": 86827297.0,
      "step": 47055
    },
    {
      "entropy": 5.636958265304566,
      "epoch": 3.9536651963873135,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035104018672493104,
      "loss": 5.0809,
      "mean_token_accuracy": 0.20387664586305618,
      "num_tokens": 86836434.0,
      "step": 47060
    },
    {
      "entropy": 5.477658796310425,
      "epoch": 3.9540852762024787,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003510120015042004,
      "loss": 4.96,
      "mean_token_accuracy": 0.20897431373596193,
      "num_tokens": 86845844.0,
      "step": 47065
    },
    {
      "entropy": 5.472079753875732,
      "epoch": 3.954505356017643,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035098381493695643,
      "loss": 4.993,
      "mean_token_accuracy": 0.20114928632974624,
      "num_tokens": 86855051.0,
      "step": 47070
    },
    {
      "entropy": 5.483374309539795,
      "epoch": 3.9549254358328083,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003509556270236983,
      "loss": 5.0062,
      "mean_token_accuracy": 0.21139614135026932,
      "num_tokens": 86864395.0,
      "step": 47075
    },
    {
      "entropy": 5.495914316177368,
      "epoch": 3.955345515647973,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035092743776492523,
      "loss": 4.9684,
      "mean_token_accuracy": 0.20030052065849305,
      "num_tokens": 86874098.0,
      "step": 47080
    },
    {
      "entropy": 5.535921049118042,
      "epoch": 3.955765595463138,
      "grad_norm": 1.125,
      "learning_rate": 0.00035089924716113683,
      "loss": 4.9424,
      "mean_token_accuracy": 0.2118617370724678,
      "num_tokens": 86883087.0,
      "step": 47085
    },
    {
      "entropy": 5.463775587081909,
      "epoch": 3.956185675278303,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035087105521283224,
      "loss": 4.9642,
      "mean_token_accuracy": 0.2093803659081459,
      "num_tokens": 86891635.0,
      "step": 47090
    },
    {
      "entropy": 5.361511468887329,
      "epoch": 3.956605755093468,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000350842861920511,
      "loss": 4.9175,
      "mean_token_accuracy": 0.21302872002124787,
      "num_tokens": 86900296.0,
      "step": 47095
    },
    {
      "entropy": 5.424446678161621,
      "epoch": 3.9570258349086327,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003508146672846725,
      "loss": 4.9256,
      "mean_token_accuracy": 0.20731855630874635,
      "num_tokens": 86909339.0,
      "step": 47100
    },
    {
      "entropy": 5.571498394012451,
      "epoch": 3.9574459147237975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035078647130581627,
      "loss": 4.9812,
      "mean_token_accuracy": 0.20439632833003998,
      "num_tokens": 86918668.0,
      "step": 47105
    },
    {
      "entropy": 5.4715300559997555,
      "epoch": 3.9578659945389623,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003507582739844417,
      "loss": 4.9184,
      "mean_token_accuracy": 0.20577338039875032,
      "num_tokens": 86927999.0,
      "step": 47110
    },
    {
      "entropy": 5.494282627105713,
      "epoch": 3.9582860743541275,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035073007532104833,
      "loss": 4.972,
      "mean_token_accuracy": 0.19472778588533401,
      "num_tokens": 86936980.0,
      "step": 47115
    },
    {
      "entropy": 5.485511922836304,
      "epoch": 3.9587061541692923,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035070187531613563,
      "loss": 4.9804,
      "mean_token_accuracy": 0.20968035161495208,
      "num_tokens": 86946170.0,
      "step": 47120
    },
    {
      "entropy": 5.506824064254761,
      "epoch": 3.959126233984457,
      "grad_norm": 1.125,
      "learning_rate": 0.00035067367397020324,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21084468960762023,
      "num_tokens": 86955020.0,
      "step": 47125
    },
    {
      "entropy": 5.477758407592773,
      "epoch": 3.959546313799622,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003506454712837506,
      "loss": 4.9719,
      "mean_token_accuracy": 0.19588803201913835,
      "num_tokens": 86963825.0,
      "step": 47130
    },
    {
      "entropy": 5.5215785026550295,
      "epoch": 3.9599663936147866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035061726725727746,
      "loss": 5.0311,
      "mean_token_accuracy": 0.1985288977622986,
      "num_tokens": 86972915.0,
      "step": 47135
    },
    {
      "entropy": 5.535639429092408,
      "epoch": 3.960386473429952,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035058906189128333,
      "loss": 5.0694,
      "mean_token_accuracy": 0.20261967480182647,
      "num_tokens": 86982267.0,
      "step": 47140
    },
    {
      "entropy": 5.58118782043457,
      "epoch": 3.9608065532451167,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003505608551862679,
      "loss": 5.0855,
      "mean_token_accuracy": 0.20339512974023818,
      "num_tokens": 86992041.0,
      "step": 47145
    },
    {
      "entropy": 5.505764818191528,
      "epoch": 3.9612266330602814,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003505326471427308,
      "loss": 4.9495,
      "mean_token_accuracy": 0.2119356483221054,
      "num_tokens": 87001166.0,
      "step": 47150
    },
    {
      "entropy": 5.393270635604859,
      "epoch": 3.9616467128754462,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035050443776117184,
      "loss": 4.8897,
      "mean_token_accuracy": 0.20612401366233826,
      "num_tokens": 87010192.0,
      "step": 47155
    },
    {
      "entropy": 5.462163496017456,
      "epoch": 3.962066792690611,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003504762270420906,
      "loss": 4.9723,
      "mean_token_accuracy": 0.21356127560138702,
      "num_tokens": 87018960.0,
      "step": 47160
    },
    {
      "entropy": 5.460487318038941,
      "epoch": 3.9624868725057762,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035044801498598685,
      "loss": 4.9451,
      "mean_token_accuracy": 0.20755597054958344,
      "num_tokens": 87028000.0,
      "step": 47165
    },
    {
      "entropy": 5.484031677246094,
      "epoch": 3.962906952320941,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003504198015933604,
      "loss": 4.9665,
      "mean_token_accuracy": 0.19979339092969894,
      "num_tokens": 87036522.0,
      "step": 47170
    },
    {
      "entropy": 5.527369213104248,
      "epoch": 3.963327032136106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035039158686471093,
      "loss": 5.0309,
      "mean_token_accuracy": 0.19896012544631958,
      "num_tokens": 87046689.0,
      "step": 47175
    },
    {
      "entropy": 5.500444030761718,
      "epoch": 3.9637471119512706,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003503633708005383,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20801645815372466,
      "num_tokens": 87056702.0,
      "step": 47180
    },
    {
      "entropy": 5.4227118492126465,
      "epoch": 3.9641671917664354,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003503351534013425,
      "loss": 4.8323,
      "mean_token_accuracy": 0.21736528277397155,
      "num_tokens": 87065915.0,
      "step": 47185
    },
    {
      "entropy": 5.475740098953247,
      "epoch": 3.9645872715816006,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003503069346676231,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21513560116291047,
      "num_tokens": 87075137.0,
      "step": 47190
    },
    {
      "entropy": 5.383998918533325,
      "epoch": 3.9650073513967654,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035027871459988023,
      "loss": 4.8826,
      "mean_token_accuracy": 0.21496055871248246,
      "num_tokens": 87084382.0,
      "step": 47195
    },
    {
      "entropy": 5.392411375045777,
      "epoch": 3.96542743121193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035025049319861365,
      "loss": 4.9528,
      "mean_token_accuracy": 0.21278438717126846,
      "num_tokens": 87093957.0,
      "step": 47200
    },
    {
      "entropy": 5.419315671920776,
      "epoch": 3.965847511027095,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003502222704643233,
      "loss": 4.8918,
      "mean_token_accuracy": 0.21714086532592775,
      "num_tokens": 87102830.0,
      "step": 47205
    },
    {
      "entropy": 5.504873371124267,
      "epoch": 3.9662675908422598,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003501940463975092,
      "loss": 5.0203,
      "mean_token_accuracy": 0.20471025109291077,
      "num_tokens": 87112674.0,
      "step": 47210
    },
    {
      "entropy": 5.578301239013672,
      "epoch": 3.966687670657425,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035016582099867133,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19708752036094665,
      "num_tokens": 87122137.0,
      "step": 47215
    },
    {
      "entropy": 5.58116717338562,
      "epoch": 3.96710775047259,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003501375942683096,
      "loss": 5.0379,
      "mean_token_accuracy": 0.2021412506699562,
      "num_tokens": 87130993.0,
      "step": 47220
    },
    {
      "entropy": 5.474923181533813,
      "epoch": 3.9675278302877546,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003501093662069241,
      "loss": 4.9391,
      "mean_token_accuracy": 0.212189157307148,
      "num_tokens": 87139891.0,
      "step": 47225
    },
    {
      "entropy": 5.440056943893433,
      "epoch": 3.96794791010292,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035008113681501485,
      "loss": 4.9147,
      "mean_token_accuracy": 0.22054370045661925,
      "num_tokens": 87149496.0,
      "step": 47230
    },
    {
      "entropy": 5.548145771026611,
      "epoch": 3.9683679899180846,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035005290609308186,
      "loss": 5.121,
      "mean_token_accuracy": 0.19842641800642014,
      "num_tokens": 87159209.0,
      "step": 47235
    },
    {
      "entropy": 5.433603286743164,
      "epoch": 3.9687880697332494,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003500246740416254,
      "loss": 4.9683,
      "mean_token_accuracy": 0.21413608342409135,
      "num_tokens": 87168734.0,
      "step": 47240
    },
    {
      "entropy": 5.483098125457763,
      "epoch": 3.969208149548414,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034999644066114536,
      "loss": 4.9975,
      "mean_token_accuracy": 0.20812020152807237,
      "num_tokens": 87178172.0,
      "step": 47245
    },
    {
      "entropy": 5.560717821121216,
      "epoch": 3.969628229363579,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034996820595214204,
      "loss": 5.003,
      "mean_token_accuracy": 0.2081206902861595,
      "num_tokens": 87187739.0,
      "step": 47250
    },
    {
      "entropy": 5.4763984203338625,
      "epoch": 3.970048309178744,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003499399699151156,
      "loss": 4.9788,
      "mean_token_accuracy": 0.20720080882310868,
      "num_tokens": 87196734.0,
      "step": 47255
    },
    {
      "entropy": 5.512183380126953,
      "epoch": 3.970468388993909,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034991173255056606,
      "loss": 5.0223,
      "mean_token_accuracy": 0.20503581911325455,
      "num_tokens": 87206872.0,
      "step": 47260
    },
    {
      "entropy": 5.3979682445526125,
      "epoch": 3.9708884688090738,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003498834938589939,
      "loss": 4.8742,
      "mean_token_accuracy": 0.2165308564901352,
      "num_tokens": 87214995.0,
      "step": 47265
    },
    {
      "entropy": 5.444427728652954,
      "epoch": 3.9713085486242385,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034985525384089914,
      "loss": 4.9535,
      "mean_token_accuracy": 0.21267233192920684,
      "num_tokens": 87224178.0,
      "step": 47270
    },
    {
      "entropy": 5.546529102325439,
      "epoch": 3.9717286284394033,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003498270124967821,
      "loss": 5.0886,
      "mean_token_accuracy": 0.19620138555765151,
      "num_tokens": 87233839.0,
      "step": 47275
    },
    {
      "entropy": 5.479485511779785,
      "epoch": 3.9721487082545686,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034979876982714296,
      "loss": 4.9415,
      "mean_token_accuracy": 0.20793986916542054,
      "num_tokens": 87242191.0,
      "step": 47280
    },
    {
      "entropy": 5.659476232528687,
      "epoch": 3.9725687880697333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003497705258324823,
      "loss": 5.1873,
      "mean_token_accuracy": 0.19724384546279908,
      "num_tokens": 87251762.0,
      "step": 47285
    },
    {
      "entropy": 5.537154054641723,
      "epoch": 3.972988867884898,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003497422805133002,
      "loss": 5.0216,
      "mean_token_accuracy": 0.20371646285057068,
      "num_tokens": 87260489.0,
      "step": 47290
    },
    {
      "entropy": 5.5020777702331545,
      "epoch": 3.973408947700063,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003497140338700971,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20752036869525908,
      "num_tokens": 87270685.0,
      "step": 47295
    },
    {
      "entropy": 5.448262882232666,
      "epoch": 3.9738290275152277,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034968578590337333,
      "loss": 4.9645,
      "mean_token_accuracy": 0.21087597459554672,
      "num_tokens": 87279850.0,
      "step": 47300
    },
    {
      "entropy": 5.448714303970337,
      "epoch": 3.974249107330393,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003496575366136293,
      "loss": 4.9463,
      "mean_token_accuracy": 0.2115103378891945,
      "num_tokens": 87288644.0,
      "step": 47305
    },
    {
      "entropy": 5.590443181991577,
      "epoch": 3.9746691871455577,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003496292860013655,
      "loss": 5.1881,
      "mean_token_accuracy": 0.19496604800224304,
      "num_tokens": 87298919.0,
      "step": 47310
    },
    {
      "entropy": 5.575278425216675,
      "epoch": 3.9750892669607225,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034960103406708233,
      "loss": 5.0015,
      "mean_token_accuracy": 0.20084526836872102,
      "num_tokens": 87308130.0,
      "step": 47315
    },
    {
      "entropy": 5.463165616989135,
      "epoch": 3.9755093467758873,
      "grad_norm": 0.96875,
      "learning_rate": 0.00034957278081128025,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21261178106069564,
      "num_tokens": 87317801.0,
      "step": 47320
    },
    {
      "entropy": 5.518539142608643,
      "epoch": 3.975929426591052,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034954452623445973,
      "loss": 5.0403,
      "mean_token_accuracy": 0.20528682619333266,
      "num_tokens": 87327419.0,
      "step": 47325
    },
    {
      "entropy": 5.60525541305542,
      "epoch": 3.9763495064062173,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034951627033712136,
      "loss": 5.1134,
      "mean_token_accuracy": 0.19636772722005844,
      "num_tokens": 87338024.0,
      "step": 47330
    },
    {
      "entropy": 5.6316029071807865,
      "epoch": 3.976769586221382,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003494880131197656,
      "loss": 5.018,
      "mean_token_accuracy": 0.21109528541564943,
      "num_tokens": 87347760.0,
      "step": 47335
    },
    {
      "entropy": 5.479765701293945,
      "epoch": 3.977189666036547,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00034945975458289307,
      "loss": 4.945,
      "mean_token_accuracy": 0.21358390897512436,
      "num_tokens": 87357216.0,
      "step": 47340
    },
    {
      "entropy": 5.498654794692993,
      "epoch": 3.9776097458517117,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034943149472700425,
      "loss": 4.9878,
      "mean_token_accuracy": 0.21008117347955704,
      "num_tokens": 87367094.0,
      "step": 47345
    },
    {
      "entropy": 5.4985604763031,
      "epoch": 3.9780298256668765,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003494032335526,
      "loss": 4.9808,
      "mean_token_accuracy": 0.20565668046474456,
      "num_tokens": 87376019.0,
      "step": 47350
    },
    {
      "entropy": 5.450241899490356,
      "epoch": 3.9784499054820417,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003493749710601806,
      "loss": 5.0045,
      "mean_token_accuracy": 0.20573062747716903,
      "num_tokens": 87384804.0,
      "step": 47355
    },
    {
      "entropy": 5.470943212509155,
      "epoch": 3.9788699852972065,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000349346707250247,
      "loss": 4.9064,
      "mean_token_accuracy": 0.2124254137277603,
      "num_tokens": 87393275.0,
      "step": 47360
    },
    {
      "entropy": 5.497849750518799,
      "epoch": 3.9792900651123713,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034931844212329966,
      "loss": 5.0273,
      "mean_token_accuracy": 0.21374778747558593,
      "num_tokens": 87402590.0,
      "step": 47365
    },
    {
      "entropy": 5.4489397525787355,
      "epoch": 3.9797101449275365,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003492901756798394,
      "loss": 4.9308,
      "mean_token_accuracy": 0.20724691301584244,
      "num_tokens": 87411936.0,
      "step": 47370
    },
    {
      "entropy": 5.395819997787475,
      "epoch": 3.980130224742701,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000349261907920367,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22207052260637283,
      "num_tokens": 87421520.0,
      "step": 47375
    },
    {
      "entropy": 5.476603746414185,
      "epoch": 3.980550304557866,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003492336388453831,
      "loss": 5.0069,
      "mean_token_accuracy": 0.210677470266819,
      "num_tokens": 87430398.0,
      "step": 47380
    },
    {
      "entropy": 5.5033551216125485,
      "epoch": 3.980970384373031,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034920536845538854,
      "loss": 4.9618,
      "mean_token_accuracy": 0.204866261780262,
      "num_tokens": 87439398.0,
      "step": 47385
    },
    {
      "entropy": 5.4849522590637205,
      "epoch": 3.9813904641881956,
      "grad_norm": 1.125,
      "learning_rate": 0.00034917709675088405,
      "loss": 4.9465,
      "mean_token_accuracy": 0.2093954011797905,
      "num_tokens": 87448580.0,
      "step": 47390
    },
    {
      "entropy": 5.455413675308227,
      "epoch": 3.981810544003361,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003491488237323705,
      "loss": 4.9091,
      "mean_token_accuracy": 0.20950978398323059,
      "num_tokens": 87458168.0,
      "step": 47395
    },
    {
      "entropy": 5.459030723571777,
      "epoch": 3.9822306238185257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003491205494003487,
      "loss": 4.9845,
      "mean_token_accuracy": 0.21279743313789368,
      "num_tokens": 87467913.0,
      "step": 47400
    },
    {
      "entropy": 5.4211362361907955,
      "epoch": 3.9826507036336904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003490922737553196,
      "loss": 4.909,
      "mean_token_accuracy": 0.2119496375322342,
      "num_tokens": 87477447.0,
      "step": 47405
    },
    {
      "entropy": 5.405449295043946,
      "epoch": 3.9830707834488552,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000349063996797784,
      "loss": 4.85,
      "mean_token_accuracy": 0.2198975130915642,
      "num_tokens": 87486200.0,
      "step": 47410
    },
    {
      "entropy": 5.499718093872071,
      "epoch": 3.98349086326402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034903571852824285,
      "loss": 4.9924,
      "mean_token_accuracy": 0.21276859641075135,
      "num_tokens": 87495033.0,
      "step": 47415
    },
    {
      "entropy": 5.432205533981323,
      "epoch": 3.9839109430791853,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000349007438947197,
      "loss": 4.9091,
      "mean_token_accuracy": 0.21310048401355744,
      "num_tokens": 87504031.0,
      "step": 47420
    },
    {
      "entropy": 5.409788703918457,
      "epoch": 3.98433102289435,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003489791580551475,
      "loss": 4.9066,
      "mean_token_accuracy": 0.20666163116693498,
      "num_tokens": 87513073.0,
      "step": 47425
    },
    {
      "entropy": 5.518488359451294,
      "epoch": 3.984751102709515,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003489508758525953,
      "loss": 5.0856,
      "mean_token_accuracy": 0.20120918452739717,
      "num_tokens": 87522484.0,
      "step": 47430
    },
    {
      "entropy": 5.52045087814331,
      "epoch": 3.9851711825246796,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00034892259234004143,
      "loss": 4.9911,
      "mean_token_accuracy": 0.21441806405782698,
      "num_tokens": 87532187.0,
      "step": 47435
    },
    {
      "entropy": 5.483821630477905,
      "epoch": 3.9855912623398444,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003488943075179869,
      "loss": 4.9453,
      "mean_token_accuracy": 0.20755188465118407,
      "num_tokens": 87541264.0,
      "step": 47440
    },
    {
      "entropy": 5.471675300598145,
      "epoch": 3.9860113421550096,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034886602138693284,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20496366173028946,
      "num_tokens": 87550551.0,
      "step": 47445
    },
    {
      "entropy": 5.48110179901123,
      "epoch": 3.9864314219701744,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003488377339473801,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21194772124290467,
      "num_tokens": 87559844.0,
      "step": 47450
    },
    {
      "entropy": 5.462570095062256,
      "epoch": 3.986851501785339,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034880944519983,
      "loss": 4.9522,
      "mean_token_accuracy": 0.21105261594057084,
      "num_tokens": 87569096.0,
      "step": 47455
    },
    {
      "entropy": 5.501123332977295,
      "epoch": 3.987271581600504,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034878115514478357,
      "loss": 5.009,
      "mean_token_accuracy": 0.21101532876491547,
      "num_tokens": 87578000.0,
      "step": 47460
    },
    {
      "entropy": 5.56255989074707,
      "epoch": 3.9876916614156688,
      "grad_norm": 1.078125,
      "learning_rate": 0.000348752863782742,
      "loss": 5.094,
      "mean_token_accuracy": 0.1999652475118637,
      "num_tokens": 87587842.0,
      "step": 47465
    },
    {
      "entropy": 5.514011669158935,
      "epoch": 3.988111741230834,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034872457111420645,
      "loss": 4.9385,
      "mean_token_accuracy": 0.21229426562786102,
      "num_tokens": 87596636.0,
      "step": 47470
    },
    {
      "entropy": 5.519646644592285,
      "epoch": 3.988531821045999,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003486962771396781,
      "loss": 4.9452,
      "mean_token_accuracy": 0.19987015575170516,
      "num_tokens": 87605574.0,
      "step": 47475
    },
    {
      "entropy": 5.450894021987915,
      "epoch": 3.9889519008611636,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000348667981859658,
      "loss": 4.9449,
      "mean_token_accuracy": 0.20933422893285752,
      "num_tokens": 87614459.0,
      "step": 47480
    },
    {
      "entropy": 5.471692800521851,
      "epoch": 3.9893719806763284,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003486396852746476,
      "loss": 4.9567,
      "mean_token_accuracy": 0.21425307244062425,
      "num_tokens": 87623410.0,
      "step": 47485
    },
    {
      "entropy": 5.444355726242065,
      "epoch": 3.989792060491493,
      "grad_norm": 1.125,
      "learning_rate": 0.00034861138738514813,
      "loss": 4.8946,
      "mean_token_accuracy": 0.22203751653432846,
      "num_tokens": 87632150.0,
      "step": 47490
    },
    {
      "entropy": 5.479442977905274,
      "epoch": 3.9902121403066584,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034858308819166086,
      "loss": 4.9374,
      "mean_token_accuracy": 0.21466449797153472,
      "num_tokens": 87640169.0,
      "step": 47495
    },
    {
      "entropy": 5.457807493209839,
      "epoch": 3.990632220121823,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034855478769468703,
      "loss": 4.952,
      "mean_token_accuracy": 0.20829788446426392,
      "num_tokens": 87648934.0,
      "step": 47500
    },
    {
      "entropy": 5.3750263214111325,
      "epoch": 3.991052299936988,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034852648589472803,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2169223889708519,
      "num_tokens": 87658186.0,
      "step": 47505
    },
    {
      "entropy": 5.521238183975219,
      "epoch": 3.9914723797521527,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003484981827922852,
      "loss": 5.0573,
      "mean_token_accuracy": 0.19720627665519713,
      "num_tokens": 87667465.0,
      "step": 47510
    },
    {
      "entropy": 5.389655351638794,
      "epoch": 3.9918924595673175,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034846987838785986,
      "loss": 4.8174,
      "mean_token_accuracy": 0.21647294461727143,
      "num_tokens": 87676935.0,
      "step": 47515
    },
    {
      "entropy": 5.432281064987182,
      "epoch": 3.9923125393824828,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034844157268195346,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2153918907046318,
      "num_tokens": 87686779.0,
      "step": 47520
    },
    {
      "entropy": 5.428349781036377,
      "epoch": 3.9927326191976475,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003484132656750675,
      "loss": 4.9652,
      "mean_token_accuracy": 0.21430446654558183,
      "num_tokens": 87695761.0,
      "step": 47525
    },
    {
      "entropy": 5.415299034118652,
      "epoch": 3.9931526990128123,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034838495736770327,
      "loss": 4.8893,
      "mean_token_accuracy": 0.21577758342027664,
      "num_tokens": 87704717.0,
      "step": 47530
    },
    {
      "entropy": 5.4681642055511475,
      "epoch": 3.9935727788279776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003483566477603623,
      "loss": 4.9725,
      "mean_token_accuracy": 0.21223950237035752,
      "num_tokens": 87714561.0,
      "step": 47535
    },
    {
      "entropy": 5.541431283950805,
      "epoch": 3.9939928586431424,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034832833685354616,
      "loss": 5.0699,
      "mean_token_accuracy": 0.20451293736696244,
      "num_tokens": 87723816.0,
      "step": 47540
    },
    {
      "entropy": 5.500329160690308,
      "epoch": 3.994412938458307,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034830002464775623,
      "loss": 4.8958,
      "mean_token_accuracy": 0.21419288963079453,
      "num_tokens": 87732494.0,
      "step": 47545
    },
    {
      "entropy": 5.519484424591065,
      "epoch": 3.994833018273472,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003482717111434941,
      "loss": 5.0426,
      "mean_token_accuracy": 0.20278355181217195,
      "num_tokens": 87742030.0,
      "step": 47550
    },
    {
      "entropy": 5.493603229522705,
      "epoch": 3.9952530980886367,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034824339634126133,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20760940313339232,
      "num_tokens": 87750727.0,
      "step": 47555
    },
    {
      "entropy": 5.49902834892273,
      "epoch": 3.995673177903802,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003482150802415595,
      "loss": 5.0178,
      "mean_token_accuracy": 0.20991480797529222,
      "num_tokens": 87759940.0,
      "step": 47560
    },
    {
      "entropy": 5.489000558853149,
      "epoch": 3.9960932577189667,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034818676284489034,
      "loss": 4.9083,
      "mean_token_accuracy": 0.2077135756611824,
      "num_tokens": 87767968.0,
      "step": 47565
    },
    {
      "entropy": 5.411853885650634,
      "epoch": 3.9965133375341315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034815844415175525,
      "loss": 4.8757,
      "mean_token_accuracy": 0.21498088389635087,
      "num_tokens": 87776498.0,
      "step": 47570
    },
    {
      "entropy": 5.396696138381958,
      "epoch": 3.9969334173492963,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034813012416265606,
      "loss": 4.9572,
      "mean_token_accuracy": 0.20999955534934997,
      "num_tokens": 87786244.0,
      "step": 47575
    },
    {
      "entropy": 5.392178106307983,
      "epoch": 3.997353497164461,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034810180287809427,
      "loss": 4.8526,
      "mean_token_accuracy": 0.21473956257104873,
      "num_tokens": 87794988.0,
      "step": 47580
    },
    {
      "entropy": 5.3862457275390625,
      "epoch": 3.9977735769796263,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034807348029857176,
      "loss": 4.8342,
      "mean_token_accuracy": 0.21566221117973328,
      "num_tokens": 87803662.0,
      "step": 47585
    },
    {
      "entropy": 5.418942880630493,
      "epoch": 3.998193656794791,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034804515642459025,
      "loss": 4.9718,
      "mean_token_accuracy": 0.2075666978955269,
      "num_tokens": 87812651.0,
      "step": 47590
    },
    {
      "entropy": 5.474546432495117,
      "epoch": 3.998613736609956,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003480168312566512,
      "loss": 4.967,
      "mean_token_accuracy": 0.20529324412345887,
      "num_tokens": 87821882.0,
      "step": 47595
    },
    {
      "entropy": 5.468352031707764,
      "epoch": 3.9990338164251207,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034798850479525673,
      "loss": 4.9618,
      "mean_token_accuracy": 0.20732074826955796,
      "num_tokens": 87830822.0,
      "step": 47600
    },
    {
      "entropy": 5.541777801513672,
      "epoch": 3.9994538962402855,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003479601770409085,
      "loss": 5.0278,
      "mean_token_accuracy": 0.20811542570590974,
      "num_tokens": 87840649.0,
      "step": 47605
    },
    {
      "entropy": 5.547933530807495,
      "epoch": 3.9998739760554507,
      "grad_norm": 1.125,
      "learning_rate": 0.0003479318479941082,
      "loss": 5.0969,
      "mean_token_accuracy": 0.20077746510505676,
      "num_tokens": 87850174.0,
      "step": 47610
    },
    {
      "entropy": 5.516906049516466,
      "epoch": 4.000252047889099,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034790351765535786,
      "loss": 4.9369,
      "mean_token_accuracy": 0.20706819163428414,
      "num_tokens": 87857458.0,
      "step": 47615
    },
    {
      "entropy": 5.418432331085205,
      "epoch": 4.000672127704264,
      "grad_norm": 1.0,
      "learning_rate": 0.0003478751860251592,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21288297325372696,
      "num_tokens": 87866518.0,
      "step": 47620
    },
    {
      "entropy": 5.443150377273559,
      "epoch": 4.001092207519429,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034784685310401415,
      "loss": 4.8771,
      "mean_token_accuracy": 0.21760060042142867,
      "num_tokens": 87875482.0,
      "step": 47625
    },
    {
      "entropy": 5.490393257141113,
      "epoch": 4.001512287334593,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003478185188924246,
      "loss": 4.9306,
      "mean_token_accuracy": 0.21044571250677108,
      "num_tokens": 87885425.0,
      "step": 47630
    },
    {
      "entropy": 5.474499940872192,
      "epoch": 4.001932367149759,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034779018339089254,
      "loss": 4.9378,
      "mean_token_accuracy": 0.20897088944911957,
      "num_tokens": 87894603.0,
      "step": 47635
    },
    {
      "entropy": 5.44446873664856,
      "epoch": 4.002352446964923,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003477618465999198,
      "loss": 4.8453,
      "mean_token_accuracy": 0.21426943093538284,
      "num_tokens": 87903952.0,
      "step": 47640
    },
    {
      "entropy": 5.447433662414551,
      "epoch": 4.002772526780088,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003477335085200084,
      "loss": 4.8478,
      "mean_token_accuracy": 0.21339399367570877,
      "num_tokens": 87912353.0,
      "step": 47645
    },
    {
      "entropy": 5.322248220443726,
      "epoch": 4.0031926065952534,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003477051691516604,
      "loss": 4.7979,
      "mean_token_accuracy": 0.21964626610279084,
      "num_tokens": 87921665.0,
      "step": 47650
    },
    {
      "entropy": 5.404146766662597,
      "epoch": 4.003612686410418,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003476768284953777,
      "loss": 4.8366,
      "mean_token_accuracy": 0.22152095437049865,
      "num_tokens": 87930531.0,
      "step": 47655
    },
    {
      "entropy": 5.350139856338501,
      "epoch": 4.004032766225583,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034764848655166256,
      "loss": 4.7951,
      "mean_token_accuracy": 0.2168621301651001,
      "num_tokens": 87939782.0,
      "step": 47660
    },
    {
      "entropy": 5.51495246887207,
      "epoch": 4.004452846040747,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003476201433210167,
      "loss": 5.0269,
      "mean_token_accuracy": 0.20645253956317902,
      "num_tokens": 87949544.0,
      "step": 47665
    },
    {
      "entropy": 5.527499771118164,
      "epoch": 4.004872925855913,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003475917988039425,
      "loss": 4.9887,
      "mean_token_accuracy": 0.20449565649032592,
      "num_tokens": 87959052.0,
      "step": 47670
    },
    {
      "entropy": 5.522515821456909,
      "epoch": 4.005293005671078,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034756345300094194,
      "loss": 4.9419,
      "mean_token_accuracy": 0.20655843764543533,
      "num_tokens": 87969496.0,
      "step": 47675
    },
    {
      "entropy": 5.475372791290283,
      "epoch": 4.005713085486242,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003475351059125172,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2207314997911453,
      "num_tokens": 87978514.0,
      "step": 47680
    },
    {
      "entropy": 5.441000699996948,
      "epoch": 4.006133165301407,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034750675753917037,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2148973435163498,
      "num_tokens": 87987665.0,
      "step": 47685
    },
    {
      "entropy": 5.413665056228638,
      "epoch": 4.006553245116572,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034747840788140364,
      "loss": 4.834,
      "mean_token_accuracy": 0.21925269514322282,
      "num_tokens": 87995952.0,
      "step": 47690
    },
    {
      "entropy": 5.4817369937896725,
      "epoch": 4.006973324931737,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034745005693971934,
      "loss": 4.943,
      "mean_token_accuracy": 0.21170349419116974,
      "num_tokens": 88004927.0,
      "step": 47695
    },
    {
      "entropy": 5.37901725769043,
      "epoch": 4.007393404746902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003474217047146195,
      "loss": 4.8364,
      "mean_token_accuracy": 0.21948163360357284,
      "num_tokens": 88013715.0,
      "step": 47700
    },
    {
      "entropy": 5.413533020019531,
      "epoch": 4.0078134845620665,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003473933512066065,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21077515035867692,
      "num_tokens": 88021834.0,
      "step": 47705
    },
    {
      "entropy": 5.424646043777466,
      "epoch": 4.008233564377232,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034736499641618255,
      "loss": 4.9145,
      "mean_token_accuracy": 0.2079047903418541,
      "num_tokens": 88030508.0,
      "step": 47710
    },
    {
      "entropy": 5.4310389995574955,
      "epoch": 4.008653644192396,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034733664034384995,
      "loss": 4.8294,
      "mean_token_accuracy": 0.21560627371072769,
      "num_tokens": 88039488.0,
      "step": 47715
    },
    {
      "entropy": 5.474461555480957,
      "epoch": 4.009073724007561,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003473082829901109,
      "loss": 4.8698,
      "mean_token_accuracy": 0.21771395057439805,
      "num_tokens": 88048606.0,
      "step": 47720
    },
    {
      "entropy": 5.472630453109741,
      "epoch": 4.009493803822727,
      "grad_norm": 1.140625,
      "learning_rate": 0.000347279924355468,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22217725068330765,
      "num_tokens": 88057413.0,
      "step": 47725
    },
    {
      "entropy": 5.458142614364624,
      "epoch": 4.009913883637891,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003472515644404234,
      "loss": 4.9546,
      "mean_token_accuracy": 0.20703390687704087,
      "num_tokens": 88066457.0,
      "step": 47730
    },
    {
      "entropy": 5.389691686630249,
      "epoch": 4.010333963453056,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003472232032454795,
      "loss": 4.8696,
      "mean_token_accuracy": 0.21935302317142485,
      "num_tokens": 88075041.0,
      "step": 47735
    },
    {
      "entropy": 5.384860992431641,
      "epoch": 4.010754043268221,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034719484077113885,
      "loss": 4.8044,
      "mean_token_accuracy": 0.23132073730230332,
      "num_tokens": 88083635.0,
      "step": 47740
    },
    {
      "entropy": 5.500202703475952,
      "epoch": 4.011174123083386,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003471664770179037,
      "loss": 4.95,
      "mean_token_accuracy": 0.2068644419312477,
      "num_tokens": 88091984.0,
      "step": 47745
    },
    {
      "entropy": 5.466623973846436,
      "epoch": 4.011594202898551,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034713811198627657,
      "loss": 4.8225,
      "mean_token_accuracy": 0.2167079731822014,
      "num_tokens": 88101214.0,
      "step": 47750
    },
    {
      "entropy": 5.486351585388183,
      "epoch": 4.012014282713715,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034710974567675994,
      "loss": 4.8752,
      "mean_token_accuracy": 0.2140852317214012,
      "num_tokens": 88110258.0,
      "step": 47755
    },
    {
      "entropy": 5.384569072723389,
      "epoch": 4.0124343625288805,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034708137808985636,
      "loss": 4.8915,
      "mean_token_accuracy": 0.2172925055027008,
      "num_tokens": 88120096.0,
      "step": 47760
    },
    {
      "entropy": 5.3845014572143555,
      "epoch": 4.012854442344046,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034705300922606826,
      "loss": 4.8592,
      "mean_token_accuracy": 0.21799408048391342,
      "num_tokens": 88130251.0,
      "step": 47765
    },
    {
      "entropy": 5.444251585006714,
      "epoch": 4.01327452215921,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034702463908589816,
      "loss": 4.905,
      "mean_token_accuracy": 0.20742746740579604,
      "num_tokens": 88140008.0,
      "step": 47770
    },
    {
      "entropy": 5.530675554275513,
      "epoch": 4.013694601974375,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034699626766984866,
      "loss": 4.9385,
      "mean_token_accuracy": 0.20995734483003617,
      "num_tokens": 88149285.0,
      "step": 47775
    },
    {
      "entropy": 5.454541826248169,
      "epoch": 4.01411468178954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034696789497842237,
      "loss": 4.845,
      "mean_token_accuracy": 0.22042316496372222,
      "num_tokens": 88157384.0,
      "step": 47780
    },
    {
      "entropy": 5.40507869720459,
      "epoch": 4.014534761604705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034693952101212184,
      "loss": 4.8523,
      "mean_token_accuracy": 0.22278631925582887,
      "num_tokens": 88165904.0,
      "step": 47785
    },
    {
      "entropy": 5.459325647354126,
      "epoch": 4.01495484141987,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003469111457714498,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20882570296525954,
      "num_tokens": 88175302.0,
      "step": 47790
    },
    {
      "entropy": 5.452072381973267,
      "epoch": 4.0153749212350345,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034688276925690873,
      "loss": 4.8962,
      "mean_token_accuracy": 0.21092237085103988,
      "num_tokens": 88184379.0,
      "step": 47795
    },
    {
      "entropy": 5.431977844238281,
      "epoch": 4.0157950010502,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034685439146900154,
      "loss": 4.8804,
      "mean_token_accuracy": 0.21273224502801896,
      "num_tokens": 88193014.0,
      "step": 47800
    },
    {
      "entropy": 5.414346599578858,
      "epoch": 4.016215080865364,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034682601240823075,
      "loss": 4.8039,
      "mean_token_accuracy": 0.21710167676210404,
      "num_tokens": 88202743.0,
      "step": 47805
    },
    {
      "entropy": 5.4035200595855715,
      "epoch": 4.016635160680529,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003467976320750991,
      "loss": 4.8062,
      "mean_token_accuracy": 0.22459492236375808,
      "num_tokens": 88211926.0,
      "step": 47810
    },
    {
      "entropy": 5.500044918060302,
      "epoch": 4.0170552404956945,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034676925047010947,
      "loss": 4.9203,
      "mean_token_accuracy": 0.21914235055446624,
      "num_tokens": 88221114.0,
      "step": 47815
    },
    {
      "entropy": 5.433939456939697,
      "epoch": 4.017475320310859,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003467408675937645,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2129262775182724,
      "num_tokens": 88230546.0,
      "step": 47820
    },
    {
      "entropy": 5.551132249832153,
      "epoch": 4.017895400126024,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034671248344656696,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20316300690174102,
      "num_tokens": 88239945.0,
      "step": 47825
    },
    {
      "entropy": 5.393768739700318,
      "epoch": 4.018315479941188,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003466840980290197,
      "loss": 4.7746,
      "mean_token_accuracy": 0.22660475969314575,
      "num_tokens": 88249008.0,
      "step": 47830
    },
    {
      "entropy": 5.361328887939453,
      "epoch": 4.018735559756354,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003466557113416256,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2178609624505043,
      "num_tokens": 88257862.0,
      "step": 47835
    },
    {
      "entropy": 5.44624810218811,
      "epoch": 4.019155639571519,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034662732338488746,
      "loss": 4.8943,
      "mean_token_accuracy": 0.20815809518098832,
      "num_tokens": 88266361.0,
      "step": 47840
    },
    {
      "entropy": 5.429787254333496,
      "epoch": 4.019575719386683,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034659893415930817,
      "loss": 4.7987,
      "mean_token_accuracy": 0.21912445574998857,
      "num_tokens": 88275915.0,
      "step": 47845
    },
    {
      "entropy": 5.464273405075073,
      "epoch": 4.0199957992018485,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003465705436653905,
      "loss": 4.915,
      "mean_token_accuracy": 0.2050095096230507,
      "num_tokens": 88284987.0,
      "step": 47850
    },
    {
      "entropy": 5.449810457229614,
      "epoch": 4.020415879017013,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003465421519036377,
      "loss": 4.9425,
      "mean_token_accuracy": 0.2064710736274719,
      "num_tokens": 88294131.0,
      "step": 47855
    },
    {
      "entropy": 5.527467966079712,
      "epoch": 4.020835958832178,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034651375887455245,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21205778419971466,
      "num_tokens": 88303779.0,
      "step": 47860
    },
    {
      "entropy": 5.5257049083709715,
      "epoch": 4.021256038647343,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003464853645786377,
      "loss": 4.898,
      "mean_token_accuracy": 0.2150259256362915,
      "num_tokens": 88312928.0,
      "step": 47865
    },
    {
      "entropy": 5.415910148620606,
      "epoch": 4.021676118462508,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034645696901639656,
      "loss": 4.8624,
      "mean_token_accuracy": 0.21333756744861604,
      "num_tokens": 88321967.0,
      "step": 47870
    },
    {
      "entropy": 5.442799091339111,
      "epoch": 4.022096198277673,
      "grad_norm": 1.125,
      "learning_rate": 0.00034642857218833206,
      "loss": 4.919,
      "mean_token_accuracy": 0.21374125331640242,
      "num_tokens": 88331102.0,
      "step": 47875
    },
    {
      "entropy": 5.4338044166564945,
      "epoch": 4.022516278092838,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003464001740949472,
      "loss": 4.8973,
      "mean_token_accuracy": 0.2113591194152832,
      "num_tokens": 88340932.0,
      "step": 47880
    },
    {
      "entropy": 5.476074695587158,
      "epoch": 4.022936357908002,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000346371774736745,
      "loss": 4.8738,
      "mean_token_accuracy": 0.21915017813444138,
      "num_tokens": 88350467.0,
      "step": 47885
    },
    {
      "entropy": 5.539991807937622,
      "epoch": 4.023356437723168,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003463433741142285,
      "loss": 4.9649,
      "mean_token_accuracy": 0.20740985125303268,
      "num_tokens": 88359593.0,
      "step": 47890
    },
    {
      "entropy": 5.459808874130249,
      "epoch": 4.023776517538332,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034631497222790096,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21049996465444565,
      "num_tokens": 88368510.0,
      "step": 47895
    },
    {
      "entropy": 5.500935220718384,
      "epoch": 4.024196597353497,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003462865690782653,
      "loss": 5.0126,
      "mean_token_accuracy": 0.2041783571243286,
      "num_tokens": 88378121.0,
      "step": 47900
    },
    {
      "entropy": 5.480351448059082,
      "epoch": 4.0246166771686624,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034625816466582487,
      "loss": 4.872,
      "mean_token_accuracy": 0.22612225860357285,
      "num_tokens": 88386879.0,
      "step": 47905
    },
    {
      "entropy": 5.360937976837159,
      "epoch": 4.025036756983827,
      "grad_norm": 1.125,
      "learning_rate": 0.00034622975899108275,
      "loss": 4.7376,
      "mean_token_accuracy": 0.2208463579416275,
      "num_tokens": 88395489.0,
      "step": 47910
    },
    {
      "entropy": 5.377479028701782,
      "epoch": 4.025456836798992,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034620135205454207,
      "loss": 4.8688,
      "mean_token_accuracy": 0.21934324651956558,
      "num_tokens": 88404394.0,
      "step": 47915
    },
    {
      "entropy": 5.4739875316619875,
      "epoch": 4.025876916614156,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003461729438567061,
      "loss": 5.0063,
      "mean_token_accuracy": 0.2091548651456833,
      "num_tokens": 88414568.0,
      "step": 47920
    },
    {
      "entropy": 5.461609649658203,
      "epoch": 4.026296996429322,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034614453439807814,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2176066279411316,
      "num_tokens": 88423519.0,
      "step": 47925
    },
    {
      "entropy": 5.490313148498535,
      "epoch": 4.026717076244487,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034611612367916136,
      "loss": 4.9621,
      "mean_token_accuracy": 0.21074456125497817,
      "num_tokens": 88433609.0,
      "step": 47930
    },
    {
      "entropy": 5.476571989059448,
      "epoch": 4.027137156059651,
      "grad_norm": 1.046875,
      "learning_rate": 0.000346087711700459,
      "loss": 4.8841,
      "mean_token_accuracy": 0.2123868480324745,
      "num_tokens": 88442687.0,
      "step": 47935
    },
    {
      "entropy": 5.474448108673096,
      "epoch": 4.027557235874816,
      "grad_norm": 1.125,
      "learning_rate": 0.0003460592984624745,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21474373638629912,
      "num_tokens": 88451577.0,
      "step": 47940
    },
    {
      "entropy": 5.4042298793792725,
      "epoch": 4.027977315689981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034603088396571106,
      "loss": 4.8235,
      "mean_token_accuracy": 0.22276555597782136,
      "num_tokens": 88460341.0,
      "step": 47945
    },
    {
      "entropy": 5.383672094345092,
      "epoch": 4.028397395505146,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034600246821067215,
      "loss": 4.852,
      "mean_token_accuracy": 0.20994616001844407,
      "num_tokens": 88470145.0,
      "step": 47950
    },
    {
      "entropy": 5.523788595199585,
      "epoch": 4.028817475320311,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034597405119786103,
      "loss": 4.9385,
      "mean_token_accuracy": 0.22120608687400817,
      "num_tokens": 88479086.0,
      "step": 47955
    },
    {
      "entropy": 5.5008628368377686,
      "epoch": 4.0292375551354755,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003459456329277812,
      "loss": 4.9037,
      "mean_token_accuracy": 0.21052432358264922,
      "num_tokens": 88488377.0,
      "step": 47960
    },
    {
      "entropy": 5.335264539718628,
      "epoch": 4.029657634950641,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034591721340093603,
      "loss": 4.738,
      "mean_token_accuracy": 0.2348978564143181,
      "num_tokens": 88497610.0,
      "step": 47965
    },
    {
      "entropy": 5.447444486618042,
      "epoch": 4.030077714765805,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003458887926178288,
      "loss": 4.938,
      "mean_token_accuracy": 0.2101888343691826,
      "num_tokens": 88507513.0,
      "step": 47970
    },
    {
      "entropy": 5.4079100608825685,
      "epoch": 4.03049779458097,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034586037057896327,
      "loss": 4.893,
      "mean_token_accuracy": 0.21429445892572402,
      "num_tokens": 88516352.0,
      "step": 47975
    },
    {
      "entropy": 5.519663143157959,
      "epoch": 4.030917874396136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003458319472848427,
      "loss": 5.0143,
      "mean_token_accuracy": 0.20790700614452362,
      "num_tokens": 88525505.0,
      "step": 47980
    },
    {
      "entropy": 5.549775266647339,
      "epoch": 4.0313379542113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034580352273597066,
      "loss": 4.8847,
      "mean_token_accuracy": 0.21368385553359986,
      "num_tokens": 88534589.0,
      "step": 47985
    },
    {
      "entropy": 5.5229216575622555,
      "epoch": 4.031758034026465,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003457750969328506,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21311957240104676,
      "num_tokens": 88543720.0,
      "step": 47990
    },
    {
      "entropy": 5.388659811019897,
      "epoch": 4.0321781138416295,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034574666987598624,
      "loss": 4.8725,
      "mean_token_accuracy": 0.21599580198526383,
      "num_tokens": 88554020.0,
      "step": 47995
    },
    {
      "entropy": 5.419725751876831,
      "epoch": 4.032598193656795,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003457182415658809,
      "loss": 4.8355,
      "mean_token_accuracy": 0.21954182386398316,
      "num_tokens": 88563509.0,
      "step": 48000
    },
    {
      "epoch": 4.032598193656795,
      "eval_entropy": 5.220952764054586,
      "eval_loss": 5.05219841003418,
      "eval_mean_token_accuracy": 0.21357818736308545,
      "eval_num_tokens": 88563509.0,
      "eval_runtime": 27.4283,
      "eval_samples_per_second": 1362.317,
      "eval_steps_per_second": 170.299,
      "step": 48000
    },
    {
      "entropy": 5.430350017547608,
      "epoch": 4.03301827347196,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034568981200303853,
      "loss": 4.8445,
      "mean_token_accuracy": 0.21689768731594086,
      "num_tokens": 88572258.0,
      "step": 48005
    },
    {
      "entropy": 5.435139417648315,
      "epoch": 4.033438353287124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034566138118796247,
      "loss": 4.8747,
      "mean_token_accuracy": 0.21271519958972931,
      "num_tokens": 88581590.0,
      "step": 48010
    },
    {
      "entropy": 5.447424745559692,
      "epoch": 4.0338584331022895,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003456329491211564,
      "loss": 4.8236,
      "mean_token_accuracy": 0.21238983124494554,
      "num_tokens": 88591113.0,
      "step": 48015
    },
    {
      "entropy": 5.471697378158569,
      "epoch": 4.034278512917455,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000345604515803124,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2124950796365738,
      "num_tokens": 88600276.0,
      "step": 48020
    },
    {
      "entropy": 5.470139837265014,
      "epoch": 4.034698592732619,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000345576081234369,
      "loss": 4.935,
      "mean_token_accuracy": 0.20903220474720002,
      "num_tokens": 88609609.0,
      "step": 48025
    },
    {
      "entropy": 5.504128885269165,
      "epoch": 4.035118672547784,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003455476454153951,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20278454571962357,
      "num_tokens": 88618664.0,
      "step": 48030
    },
    {
      "entropy": 5.500490665435791,
      "epoch": 4.035538752362949,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003455192083467059,
      "loss": 4.8765,
      "mean_token_accuracy": 0.2128438726067543,
      "num_tokens": 88627679.0,
      "step": 48035
    },
    {
      "entropy": 5.414607524871826,
      "epoch": 4.035958832178114,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034549077002880526,
      "loss": 4.8438,
      "mean_token_accuracy": 0.21608188152313232,
      "num_tokens": 88637421.0,
      "step": 48040
    },
    {
      "entropy": 5.42626371383667,
      "epoch": 4.036378911993279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034546233046219693,
      "loss": 4.9333,
      "mean_token_accuracy": 0.20921924859285354,
      "num_tokens": 88646375.0,
      "step": 48045
    },
    {
      "entropy": 5.397241544723511,
      "epoch": 4.0367989918084435,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003454338896473847,
      "loss": 4.7977,
      "mean_token_accuracy": 0.22411223649978637,
      "num_tokens": 88654869.0,
      "step": 48050
    },
    {
      "entropy": 5.482892417907715,
      "epoch": 4.037219071623609,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034540544758487234,
      "loss": 4.9015,
      "mean_token_accuracy": 0.2083645686507225,
      "num_tokens": 88663172.0,
      "step": 48055
    },
    {
      "entropy": 5.494675731658935,
      "epoch": 4.037639151438773,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003453770042751638,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20841875523328782,
      "num_tokens": 88672581.0,
      "step": 48060
    },
    {
      "entropy": 5.435575771331787,
      "epoch": 4.038059231253938,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003453485597187629,
      "loss": 4.8213,
      "mean_token_accuracy": 0.2278233841061592,
      "num_tokens": 88681452.0,
      "step": 48065
    },
    {
      "entropy": 5.413199710845947,
      "epoch": 4.0384793110691035,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034532011391617335,
      "loss": 4.8161,
      "mean_token_accuracy": 0.21883077472448348,
      "num_tokens": 88690520.0,
      "step": 48070
    },
    {
      "entropy": 5.358054351806641,
      "epoch": 4.038899390884268,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034529166686789924,
      "loss": 4.7838,
      "mean_token_accuracy": 0.23018029928207398,
      "num_tokens": 88699192.0,
      "step": 48075
    },
    {
      "entropy": 5.458267259597778,
      "epoch": 4.039319470699433,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003452632185744445,
      "loss": 4.9767,
      "mean_token_accuracy": 0.20948563814163207,
      "num_tokens": 88708577.0,
      "step": 48080
    },
    {
      "entropy": 5.404564476013183,
      "epoch": 4.039739550514597,
      "grad_norm": 1.125,
      "learning_rate": 0.000345234769036313,
      "loss": 4.8659,
      "mean_token_accuracy": 0.21172239929437636,
      "num_tokens": 88716719.0,
      "step": 48085
    },
    {
      "entropy": 5.4572371482849125,
      "epoch": 4.040159630329763,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003452063182540088,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21773527562618256,
      "num_tokens": 88725654.0,
      "step": 48090
    },
    {
      "entropy": 5.435724258422852,
      "epoch": 4.040579710144928,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003451778662280357,
      "loss": 4.8973,
      "mean_token_accuracy": 0.2211460292339325,
      "num_tokens": 88735065.0,
      "step": 48095
    },
    {
      "entropy": 5.434030532836914,
      "epoch": 4.040999789960092,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003451494129588979,
      "loss": 4.8974,
      "mean_token_accuracy": 0.215413461625576,
      "num_tokens": 88744789.0,
      "step": 48100
    },
    {
      "entropy": 5.449179315567017,
      "epoch": 4.0414198697752575,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034512095844709933,
      "loss": 4.9103,
      "mean_token_accuracy": 0.21221819519996643,
      "num_tokens": 88754095.0,
      "step": 48105
    },
    {
      "entropy": 5.516205883026123,
      "epoch": 4.041839949590422,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003450925026931442,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21280751824378968,
      "num_tokens": 88762275.0,
      "step": 48110
    },
    {
      "entropy": 5.462885236740112,
      "epoch": 4.042260029405587,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003450640456975364,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21465744823217392,
      "num_tokens": 88771781.0,
      "step": 48115
    },
    {
      "entropy": 5.51840181350708,
      "epoch": 4.042680109220752,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034503558746078005,
      "loss": 4.9583,
      "mean_token_accuracy": 0.20140764564275743,
      "num_tokens": 88780529.0,
      "step": 48120
    },
    {
      "entropy": 5.467638301849365,
      "epoch": 4.043100189035917,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003450071279833794,
      "loss": 4.9534,
      "mean_token_accuracy": 0.20787497162818908,
      "num_tokens": 88789949.0,
      "step": 48125
    },
    {
      "entropy": 5.434797191619873,
      "epoch": 4.043520268851082,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034497866726583847,
      "loss": 4.9321,
      "mean_token_accuracy": 0.2110128805041313,
      "num_tokens": 88799673.0,
      "step": 48130
    },
    {
      "entropy": 5.433090496063232,
      "epoch": 4.043940348666246,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003449502053086615,
      "loss": 4.843,
      "mean_token_accuracy": 0.21590443849563598,
      "num_tokens": 88808970.0,
      "step": 48135
    },
    {
      "entropy": 5.464733409881592,
      "epoch": 4.044360428481411,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034492174211235273,
      "loss": 4.8697,
      "mean_token_accuracy": 0.22042596638202666,
      "num_tokens": 88818368.0,
      "step": 48140
    },
    {
      "entropy": 5.54519100189209,
      "epoch": 4.044780508296577,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034489327767741625,
      "loss": 5.044,
      "mean_token_accuracy": 0.20597912967205048,
      "num_tokens": 88828290.0,
      "step": 48145
    },
    {
      "entropy": 5.416010046005249,
      "epoch": 4.045200588111741,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034486481200435634,
      "loss": 4.8463,
      "mean_token_accuracy": 0.21515184193849562,
      "num_tokens": 88837561.0,
      "step": 48150
    },
    {
      "entropy": 5.5015833377838135,
      "epoch": 4.045620667926906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034483634509367727,
      "loss": 4.9909,
      "mean_token_accuracy": 0.20409805476665496,
      "num_tokens": 88846694.0,
      "step": 48155
    },
    {
      "entropy": 5.503232383728028,
      "epoch": 4.046040747742071,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003448078769458833,
      "loss": 4.9591,
      "mean_token_accuracy": 0.20989683866500855,
      "num_tokens": 88856497.0,
      "step": 48160
    },
    {
      "entropy": 5.433514213562011,
      "epoch": 4.046460827557236,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003447794075614788,
      "loss": 4.8207,
      "mean_token_accuracy": 0.21725666373968125,
      "num_tokens": 88865635.0,
      "step": 48165
    },
    {
      "entropy": 5.41500563621521,
      "epoch": 4.046880907372401,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003447509369409679,
      "loss": 4.7828,
      "mean_token_accuracy": 0.22344522476196288,
      "num_tokens": 88874859.0,
      "step": 48170
    },
    {
      "entropy": 5.452467823028565,
      "epoch": 4.047300987187565,
      "grad_norm": 1.125,
      "learning_rate": 0.0003447224650848552,
      "loss": 4.9572,
      "mean_token_accuracy": 0.20779687762260438,
      "num_tokens": 88884266.0,
      "step": 48175
    },
    {
      "entropy": 5.510038757324219,
      "epoch": 4.047721067002731,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034469399199364486,
      "loss": 4.9505,
      "mean_token_accuracy": 0.20517540574073792,
      "num_tokens": 88893733.0,
      "step": 48180
    },
    {
      "entropy": 5.490356397628784,
      "epoch": 4.048141146817896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003446655176678414,
      "loss": 4.852,
      "mean_token_accuracy": 0.21078796237707137,
      "num_tokens": 88902175.0,
      "step": 48185
    },
    {
      "entropy": 5.397002840042115,
      "epoch": 4.04856122663306,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003446370421079491,
      "loss": 4.8616,
      "mean_token_accuracy": 0.2197696104645729,
      "num_tokens": 88911562.0,
      "step": 48190
    },
    {
      "entropy": 5.433954572677612,
      "epoch": 4.048981306448225,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034460856531447247,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2187170699238777,
      "num_tokens": 88920315.0,
      "step": 48195
    },
    {
      "entropy": 5.459287548065186,
      "epoch": 4.04940138626339,
      "grad_norm": 1.03125,
      "learning_rate": 0.000344580087287916,
      "loss": 4.87,
      "mean_token_accuracy": 0.21925814896821977,
      "num_tokens": 88929960.0,
      "step": 48200
    },
    {
      "entropy": 5.370856380462646,
      "epoch": 4.049821466078555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034455160802878397,
      "loss": 4.832,
      "mean_token_accuracy": 0.22227868139743806,
      "num_tokens": 88938558.0,
      "step": 48205
    },
    {
      "entropy": 5.5110414028167725,
      "epoch": 4.05024154589372,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034452312753758114,
      "loss": 5.0215,
      "mean_token_accuracy": 0.2055961787700653,
      "num_tokens": 88947981.0,
      "step": 48210
    },
    {
      "entropy": 5.532423305511474,
      "epoch": 4.0506616257088846,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003444946458148119,
      "loss": 5.0493,
      "mean_token_accuracy": 0.2099823072552681,
      "num_tokens": 88957695.0,
      "step": 48215
    },
    {
      "entropy": 5.55178747177124,
      "epoch": 4.05108170552405,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034446616286098076,
      "loss": 5.001,
      "mean_token_accuracy": 0.21182742714881897,
      "num_tokens": 88967456.0,
      "step": 48220
    },
    {
      "entropy": 5.465855312347412,
      "epoch": 4.051501785339214,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003444376786765923,
      "loss": 4.7844,
      "mean_token_accuracy": 0.22468207627534867,
      "num_tokens": 88975236.0,
      "step": 48225
    },
    {
      "entropy": 5.426042890548706,
      "epoch": 4.051921865154379,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00034440919326215113,
      "loss": 4.8298,
      "mean_token_accuracy": 0.21790796369314194,
      "num_tokens": 88985045.0,
      "step": 48230
    },
    {
      "entropy": 5.4769388198852536,
      "epoch": 4.052341944969545,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003443807066181619,
      "loss": 5.0693,
      "mean_token_accuracy": 0.20116005837917328,
      "num_tokens": 88994645.0,
      "step": 48235
    },
    {
      "entropy": 5.472827911376953,
      "epoch": 4.052762024784709,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034435221874512913,
      "loss": 4.9274,
      "mean_token_accuracy": 0.21641045361757277,
      "num_tokens": 89003162.0,
      "step": 48240
    },
    {
      "entropy": 5.507418298721314,
      "epoch": 4.053182104599874,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003443237296435575,
      "loss": 4.8389,
      "mean_token_accuracy": 0.21550143957138063,
      "num_tokens": 89011722.0,
      "step": 48245
    },
    {
      "entropy": 5.509763765335083,
      "epoch": 4.0536021844150385,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034429523931395174,
      "loss": 4.9699,
      "mean_token_accuracy": 0.21203378587961197,
      "num_tokens": 89020640.0,
      "step": 48250
    },
    {
      "entropy": 5.3888352394104,
      "epoch": 4.054022264230204,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003442667477568165,
      "loss": 4.7582,
      "mean_token_accuracy": 0.23313769400119783,
      "num_tokens": 89029083.0,
      "step": 48255
    },
    {
      "entropy": 5.444520807266235,
      "epoch": 4.054442344045369,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034423825497265645,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21402686834335327,
      "num_tokens": 89038752.0,
      "step": 48260
    },
    {
      "entropy": 5.5690977573394775,
      "epoch": 4.054862423860533,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003442097609619764,
      "loss": 4.9901,
      "mean_token_accuracy": 0.2089531421661377,
      "num_tokens": 89048334.0,
      "step": 48265
    },
    {
      "entropy": 5.528855228424073,
      "epoch": 4.0552825036756985,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000344181265725281,
      "loss": 4.9112,
      "mean_token_accuracy": 0.2144190028309822,
      "num_tokens": 89057472.0,
      "step": 48270
    },
    {
      "entropy": 5.461596441268921,
      "epoch": 4.055702583490863,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034415276926307514,
      "loss": 4.8418,
      "mean_token_accuracy": 0.21545660942792894,
      "num_tokens": 89065913.0,
      "step": 48275
    },
    {
      "entropy": 5.447983169555664,
      "epoch": 4.056122663306028,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034412427157586365,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21452681422233583,
      "num_tokens": 89075035.0,
      "step": 48280
    },
    {
      "entropy": 5.4054240703582765,
      "epoch": 4.056542743121193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003440957726641512,
      "loss": 4.8467,
      "mean_token_accuracy": 0.21049136966466903,
      "num_tokens": 89085473.0,
      "step": 48285
    },
    {
      "entropy": 5.44423279762268,
      "epoch": 4.056962822936358,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003440672725284428,
      "loss": 4.8469,
      "mean_token_accuracy": 0.2183714896440506,
      "num_tokens": 89094749.0,
      "step": 48290
    },
    {
      "entropy": 5.489782047271729,
      "epoch": 4.057382902751523,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003440387711692431,
      "loss": 4.9753,
      "mean_token_accuracy": 0.20265518724918366,
      "num_tokens": 89104817.0,
      "step": 48295
    },
    {
      "entropy": 5.526382493972778,
      "epoch": 4.057802982566687,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034401026858705717,
      "loss": 4.8786,
      "mean_token_accuracy": 0.2185013473033905,
      "num_tokens": 89113738.0,
      "step": 48300
    },
    {
      "entropy": 5.512725114822388,
      "epoch": 4.0582230623818525,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034398176478238996,
      "loss": 4.945,
      "mean_token_accuracy": 0.21446824669837952,
      "num_tokens": 89122863.0,
      "step": 48305
    },
    {
      "entropy": 5.433584976196289,
      "epoch": 4.058643142197018,
      "grad_norm": 1.125,
      "learning_rate": 0.0003439532597557462,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20689885914325715,
      "num_tokens": 89131260.0,
      "step": 48310
    },
    {
      "entropy": 5.499776363372803,
      "epoch": 4.059063222012182,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000343924753507631,
      "loss": 4.9221,
      "mean_token_accuracy": 0.20973650366067886,
      "num_tokens": 89140561.0,
      "step": 48315
    },
    {
      "entropy": 5.4799206256866455,
      "epoch": 4.059483301827347,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003438962460385493,
      "loss": 4.922,
      "mean_token_accuracy": 0.21267178952693938,
      "num_tokens": 89149176.0,
      "step": 48320
    },
    {
      "entropy": 5.463288640975952,
      "epoch": 4.059903381642512,
      "grad_norm": 1.09375,
      "learning_rate": 0.000343867737349006,
      "loss": 4.8766,
      "mean_token_accuracy": 0.21649943590164183,
      "num_tokens": 89158846.0,
      "step": 48325
    },
    {
      "entropy": 5.426419353485107,
      "epoch": 4.060323461457677,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003438392274395063,
      "loss": 4.8338,
      "mean_token_accuracy": 0.22248025983572006,
      "num_tokens": 89167508.0,
      "step": 48330
    },
    {
      "entropy": 5.395200872421265,
      "epoch": 4.060743541272842,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034381071631055504,
      "loss": 4.8823,
      "mean_token_accuracy": 0.21020621061325073,
      "num_tokens": 89176296.0,
      "step": 48335
    },
    {
      "entropy": 5.451747417449951,
      "epoch": 4.061163621088006,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034378220396265736,
      "loss": 4.8381,
      "mean_token_accuracy": 0.21732264012098312,
      "num_tokens": 89184255.0,
      "step": 48340
    },
    {
      "entropy": 5.447785711288452,
      "epoch": 4.061583700903172,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034375369039631843,
      "loss": 4.8489,
      "mean_token_accuracy": 0.21652388274669648,
      "num_tokens": 89193537.0,
      "step": 48345
    },
    {
      "entropy": 5.397338628768921,
      "epoch": 4.062003780718337,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034372517561204327,
      "loss": 4.8734,
      "mean_token_accuracy": 0.2221662938594818,
      "num_tokens": 89202603.0,
      "step": 48350
    },
    {
      "entropy": 5.431818437576294,
      "epoch": 4.062423860533501,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000343696659610337,
      "loss": 4.8746,
      "mean_token_accuracy": 0.21519548892974855,
      "num_tokens": 89211325.0,
      "step": 48355
    },
    {
      "entropy": 5.463487958908081,
      "epoch": 4.0628439403486665,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034366814239170475,
      "loss": 4.8176,
      "mean_token_accuracy": 0.2212815076112747,
      "num_tokens": 89220712.0,
      "step": 48360
    },
    {
      "entropy": 5.474435472488404,
      "epoch": 4.063264020163831,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003436396239566517,
      "loss": 4.9307,
      "mean_token_accuracy": 0.2129898563027382,
      "num_tokens": 89229737.0,
      "step": 48365
    },
    {
      "entropy": 5.470993852615356,
      "epoch": 4.063684099978996,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034361110430568314,
      "loss": 4.9057,
      "mean_token_accuracy": 0.211501345038414,
      "num_tokens": 89239066.0,
      "step": 48370
    },
    {
      "entropy": 5.5056853771209715,
      "epoch": 4.064104179794161,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034358258343930413,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21174760162830353,
      "num_tokens": 89248352.0,
      "step": 48375
    },
    {
      "entropy": 5.404784393310547,
      "epoch": 4.064524259609326,
      "grad_norm": 1.125,
      "learning_rate": 0.00034355406135802,
      "loss": 4.7782,
      "mean_token_accuracy": 0.22139686495065689,
      "num_tokens": 89257557.0,
      "step": 48380
    },
    {
      "entropy": 5.442447853088379,
      "epoch": 4.064944339424491,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000343525538062336,
      "loss": 4.9197,
      "mean_token_accuracy": 0.21127384454011916,
      "num_tokens": 89267552.0,
      "step": 48385
    },
    {
      "entropy": 5.42244815826416,
      "epoch": 4.065364419239655,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003434970135527573,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2160183921456337,
      "num_tokens": 89276440.0,
      "step": 48390
    },
    {
      "entropy": 5.420693635940552,
      "epoch": 4.06578449905482,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034346848782978935,
      "loss": 4.8431,
      "mean_token_accuracy": 0.22381447851657868,
      "num_tokens": 89285399.0,
      "step": 48395
    },
    {
      "entropy": 5.382793235778808,
      "epoch": 4.066204578869986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034343996089393735,
      "loss": 4.8018,
      "mean_token_accuracy": 0.22666371315717698,
      "num_tokens": 89294102.0,
      "step": 48400
    },
    {
      "entropy": 5.4453919410705565,
      "epoch": 4.06662465868515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003434114327457067,
      "loss": 4.9476,
      "mean_token_accuracy": 0.21110558211803437,
      "num_tokens": 89303097.0,
      "step": 48405
    },
    {
      "entropy": 5.426329755783081,
      "epoch": 4.067044738500315,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003433829033856027,
      "loss": 4.8989,
      "mean_token_accuracy": 0.2203555464744568,
      "num_tokens": 89311888.0,
      "step": 48410
    },
    {
      "entropy": 5.435410737991333,
      "epoch": 4.06746481831548,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034335437281413084,
      "loss": 4.836,
      "mean_token_accuracy": 0.22096314579248427,
      "num_tokens": 89320683.0,
      "step": 48415
    },
    {
      "entropy": 5.557487154006958,
      "epoch": 4.067884898130645,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034332584103179635,
      "loss": 4.9746,
      "mean_token_accuracy": 0.21050649732351304,
      "num_tokens": 89329829.0,
      "step": 48420
    },
    {
      "entropy": 5.531410837173462,
      "epoch": 4.06830497794581,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003432973080391049,
      "loss": 4.9739,
      "mean_token_accuracy": 0.20442449450492858,
      "num_tokens": 89339465.0,
      "step": 48425
    },
    {
      "entropy": 5.4511326313018795,
      "epoch": 4.068725057760974,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003432687738365618,
      "loss": 4.872,
      "mean_token_accuracy": 0.22298807948827742,
      "num_tokens": 89348333.0,
      "step": 48430
    },
    {
      "entropy": 5.436066055297852,
      "epoch": 4.06914513757614,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034324023842467245,
      "loss": 4.868,
      "mean_token_accuracy": 0.21325802356004714,
      "num_tokens": 89357562.0,
      "step": 48435
    },
    {
      "entropy": 5.414322996139527,
      "epoch": 4.069565217391304,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003432117018039425,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21587299853563308,
      "num_tokens": 89366903.0,
      "step": 48440
    },
    {
      "entropy": 5.500495862960816,
      "epoch": 4.069985297206469,
      "grad_norm": 1.125,
      "learning_rate": 0.0003431831639748773,
      "loss": 4.864,
      "mean_token_accuracy": 0.21432138085365296,
      "num_tokens": 89375297.0,
      "step": 48445
    },
    {
      "entropy": 5.452077913284302,
      "epoch": 4.070405377021634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003431546249379825,
      "loss": 4.8877,
      "mean_token_accuracy": 0.21704449653625488,
      "num_tokens": 89384024.0,
      "step": 48450
    },
    {
      "entropy": 5.471312046051025,
      "epoch": 4.070825456836799,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034312608469376363,
      "loss": 4.9119,
      "mean_token_accuracy": 0.2171371266245842,
      "num_tokens": 89393249.0,
      "step": 48455
    },
    {
      "entropy": 5.421911907196045,
      "epoch": 4.071245536651964,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034309754324272625,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2130526691675186,
      "num_tokens": 89402297.0,
      "step": 48460
    },
    {
      "entropy": 5.3936474323272705,
      "epoch": 4.071665616467128,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034306900058537587,
      "loss": 4.8274,
      "mean_token_accuracy": 0.22413912266492844,
      "num_tokens": 89411549.0,
      "step": 48465
    },
    {
      "entropy": 5.4386743068695065,
      "epoch": 4.0720856962822936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003430404567222183,
      "loss": 4.928,
      "mean_token_accuracy": 0.21304743736982346,
      "num_tokens": 89420810.0,
      "step": 48470
    },
    {
      "entropy": 5.415219974517822,
      "epoch": 4.072505776097459,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000343011911653759,
      "loss": 4.9201,
      "mean_token_accuracy": 0.20689937323331833,
      "num_tokens": 89430491.0,
      "step": 48475
    },
    {
      "entropy": 5.568768548965454,
      "epoch": 4.072925855912623,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003429833653805037,
      "loss": 5.0176,
      "mean_token_accuracy": 0.20574821531772614,
      "num_tokens": 89440460.0,
      "step": 48480
    },
    {
      "entropy": 5.409222221374511,
      "epoch": 4.073345935727788,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003429548179029581,
      "loss": 4.8197,
      "mean_token_accuracy": 0.22076423913240434,
      "num_tokens": 89450429.0,
      "step": 48485
    },
    {
      "entropy": 5.525858545303345,
      "epoch": 4.073766015542954,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034292626922162797,
      "loss": 4.9884,
      "mean_token_accuracy": 0.21008821278810502,
      "num_tokens": 89459297.0,
      "step": 48490
    },
    {
      "entropy": 5.495998239517212,
      "epoch": 4.074186095358118,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003428977193370189,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21790860295295716,
      "num_tokens": 89468201.0,
      "step": 48495
    },
    {
      "entropy": 5.4467450141906735,
      "epoch": 4.074606175173283,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003428691682496367,
      "loss": 4.8047,
      "mean_token_accuracy": 0.22758691757917404,
      "num_tokens": 89476917.0,
      "step": 48500
    },
    {
      "entropy": 5.439471244812012,
      "epoch": 4.0750262549884475,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034284061595998713,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2163897693157196,
      "num_tokens": 89485662.0,
      "step": 48505
    },
    {
      "entropy": 5.3269209384918215,
      "epoch": 4.075446334803613,
      "grad_norm": 1.109375,
      "learning_rate": 0.000342812062468576,
      "loss": 4.799,
      "mean_token_accuracy": 0.22770470529794692,
      "num_tokens": 89494685.0,
      "step": 48510
    },
    {
      "entropy": 5.461977624893189,
      "epoch": 4.075866414618778,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034278350777590904,
      "loss": 4.9365,
      "mean_token_accuracy": 0.21587733179330826,
      "num_tokens": 89504219.0,
      "step": 48515
    },
    {
      "entropy": 5.401722526550293,
      "epoch": 4.076286494433942,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003427549518824922,
      "loss": 4.789,
      "mean_token_accuracy": 0.21943629384040833,
      "num_tokens": 89513372.0,
      "step": 48520
    },
    {
      "entropy": 5.443851661682129,
      "epoch": 4.0767065742491075,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034272639478883125,
      "loss": 4.8734,
      "mean_token_accuracy": 0.21788883209228516,
      "num_tokens": 89522812.0,
      "step": 48525
    },
    {
      "entropy": 5.428087282180786,
      "epoch": 4.077126654064272,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003426978364954321,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21538830399513245,
      "num_tokens": 89531865.0,
      "step": 48530
    },
    {
      "entropy": 5.454250288009644,
      "epoch": 4.077546733879437,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034266927700280067,
      "loss": 4.9391,
      "mean_token_accuracy": 0.2166384071111679,
      "num_tokens": 89541796.0,
      "step": 48535
    },
    {
      "entropy": 5.475990962982178,
      "epoch": 4.077966813694602,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034264071631144285,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2146402820944786,
      "num_tokens": 89550638.0,
      "step": 48540
    },
    {
      "entropy": 5.5517518520355225,
      "epoch": 4.078386893509767,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034261215442186446,
      "loss": 5.0162,
      "mean_token_accuracy": 0.20507666915655137,
      "num_tokens": 89559617.0,
      "step": 48545
    },
    {
      "entropy": 5.423297357559204,
      "epoch": 4.078806973324932,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003425835913345716,
      "loss": 4.9465,
      "mean_token_accuracy": 0.2115691363811493,
      "num_tokens": 89569260.0,
      "step": 48550
    },
    {
      "entropy": 5.455110931396485,
      "epoch": 4.079227053140096,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003425550270500703,
      "loss": 4.8987,
      "mean_token_accuracy": 0.20814870595932006,
      "num_tokens": 89578496.0,
      "step": 48555
    },
    {
      "entropy": 5.436783170700073,
      "epoch": 4.0796471329552615,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034252646156886647,
      "loss": 4.8572,
      "mean_token_accuracy": 0.21880287379026414,
      "num_tokens": 89587815.0,
      "step": 48560
    },
    {
      "entropy": 5.366738796234131,
      "epoch": 4.080067212770427,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003424978948914661,
      "loss": 4.7918,
      "mean_token_accuracy": 0.2119365304708481,
      "num_tokens": 89597525.0,
      "step": 48565
    },
    {
      "entropy": 5.38158073425293,
      "epoch": 4.080487292585591,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003424693270183753,
      "loss": 4.8413,
      "mean_token_accuracy": 0.21960222572088242,
      "num_tokens": 89606517.0,
      "step": 48570
    },
    {
      "entropy": 5.513168478012085,
      "epoch": 4.080907372400756,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003424407579501001,
      "loss": 4.9895,
      "mean_token_accuracy": 0.20127397924661636,
      "num_tokens": 89615936.0,
      "step": 48575
    },
    {
      "entropy": 5.512947273254395,
      "epoch": 4.081327452215921,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003424121876871466,
      "loss": 4.9666,
      "mean_token_accuracy": 0.20742573887109755,
      "num_tokens": 89625624.0,
      "step": 48580
    },
    {
      "entropy": 5.510631370544433,
      "epoch": 4.081747532031086,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003423836162300209,
      "loss": 4.8884,
      "mean_token_accuracy": 0.21244272589683533,
      "num_tokens": 89635203.0,
      "step": 48585
    },
    {
      "entropy": 5.480962038040161,
      "epoch": 4.082167611846251,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034235504357922917,
      "loss": 4.9026,
      "mean_token_accuracy": 0.21242786049842835,
      "num_tokens": 89643768.0,
      "step": 48590
    },
    {
      "entropy": 5.579091215133667,
      "epoch": 4.082587691661415,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003423264697352775,
      "loss": 5.096,
      "mean_token_accuracy": 0.20199289470911025,
      "num_tokens": 89653040.0,
      "step": 48595
    },
    {
      "entropy": 5.476264238357544,
      "epoch": 4.083007771476581,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034229789469867206,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21969449520111084,
      "num_tokens": 89663586.0,
      "step": 48600
    },
    {
      "entropy": 5.4424957752227785,
      "epoch": 4.083427851291745,
      "grad_norm": 1.125,
      "learning_rate": 0.00034226931846991914,
      "loss": 4.8236,
      "mean_token_accuracy": 0.22091757953166963,
      "num_tokens": 89672288.0,
      "step": 48605
    },
    {
      "entropy": 5.507123184204102,
      "epoch": 4.08384793110691,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003422407410495248,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21002403497695923,
      "num_tokens": 89681901.0,
      "step": 48610
    },
    {
      "entropy": 5.4709429264068605,
      "epoch": 4.0842680109220755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003422121624379954,
      "loss": 4.8975,
      "mean_token_accuracy": 0.21563825756311417,
      "num_tokens": 89690727.0,
      "step": 48615
    },
    {
      "entropy": 5.441308784484863,
      "epoch": 4.08468809073724,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034218358263583717,
      "loss": 4.9182,
      "mean_token_accuracy": 0.21754315495491028,
      "num_tokens": 89699856.0,
      "step": 48620
    },
    {
      "entropy": 5.4871378421783445,
      "epoch": 4.085108170552405,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034215500164355635,
      "loss": 4.928,
      "mean_token_accuracy": 0.21235115677118302,
      "num_tokens": 89708473.0,
      "step": 48625
    },
    {
      "entropy": 5.452647352218628,
      "epoch": 4.08552825036757,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034212641946165925,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21276025474071503,
      "num_tokens": 89719284.0,
      "step": 48630
    },
    {
      "entropy": 5.542075061798096,
      "epoch": 4.085948330182735,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003420978360906522,
      "loss": 5.0475,
      "mean_token_accuracy": 0.2006456509232521,
      "num_tokens": 89729590.0,
      "step": 48635
    },
    {
      "entropy": 5.4017884731292725,
      "epoch": 4.0863684099979,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003420692515310415,
      "loss": 4.8301,
      "mean_token_accuracy": 0.21422287374734877,
      "num_tokens": 89739230.0,
      "step": 48640
    },
    {
      "entropy": 5.396897602081299,
      "epoch": 4.086788489813064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003420406657833336,
      "loss": 4.8346,
      "mean_token_accuracy": 0.22365574538707733,
      "num_tokens": 89748552.0,
      "step": 48645
    },
    {
      "entropy": 5.479071521759034,
      "epoch": 4.087208569628229,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003420120788480348,
      "loss": 4.9592,
      "mean_token_accuracy": 0.21172400414943696,
      "num_tokens": 89758460.0,
      "step": 48650
    },
    {
      "entropy": 5.530460786819458,
      "epoch": 4.087628649443395,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034198349072565154,
      "loss": 4.9569,
      "mean_token_accuracy": 0.20725245922803878,
      "num_tokens": 89767983.0,
      "step": 48655
    },
    {
      "entropy": 5.513722610473633,
      "epoch": 4.088048729258559,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003419549014166902,
      "loss": 4.919,
      "mean_token_accuracy": 0.21492732763290406,
      "num_tokens": 89777795.0,
      "step": 48660
    },
    {
      "entropy": 5.468631172180176,
      "epoch": 4.088468809073724,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034192631092165734,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21693365424871444,
      "num_tokens": 89787231.0,
      "step": 48665
    },
    {
      "entropy": 5.4315698623657225,
      "epoch": 4.088888888888889,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034189771924105926,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22428373843431473,
      "num_tokens": 89796149.0,
      "step": 48670
    },
    {
      "entropy": 5.4216203689575195,
      "epoch": 4.089308968704054,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034186912637540256,
      "loss": 4.8764,
      "mean_token_accuracy": 0.2153610110282898,
      "num_tokens": 89805014.0,
      "step": 48675
    },
    {
      "entropy": 5.4506227493286135,
      "epoch": 4.089729048519219,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003418405323251937,
      "loss": 4.9136,
      "mean_token_accuracy": 0.21406846344470978,
      "num_tokens": 89813271.0,
      "step": 48680
    },
    {
      "entropy": 5.517696380615234,
      "epoch": 4.090149128334383,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003418119370909392,
      "loss": 4.9688,
      "mean_token_accuracy": 0.2091924950480461,
      "num_tokens": 89822808.0,
      "step": 48685
    },
    {
      "entropy": 5.433244228363037,
      "epoch": 4.090569208149549,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003417833406731457,
      "loss": 4.8253,
      "mean_token_accuracy": 0.21525892168283461,
      "num_tokens": 89831205.0,
      "step": 48690
    },
    {
      "entropy": 5.417109155654908,
      "epoch": 4.090989287964713,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003417547430723197,
      "loss": 4.9509,
      "mean_token_accuracy": 0.20913898199796677,
      "num_tokens": 89840127.0,
      "step": 48695
    },
    {
      "entropy": 5.4528604507446286,
      "epoch": 4.091409367779878,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003417261442889678,
      "loss": 4.9344,
      "mean_token_accuracy": 0.20753520876169204,
      "num_tokens": 89849488.0,
      "step": 48700
    },
    {
      "entropy": 5.576860570907593,
      "epoch": 4.091829447595043,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034169754432359663,
      "loss": 5.0194,
      "mean_token_accuracy": 0.20976399034261703,
      "num_tokens": 89859307.0,
      "step": 48705
    },
    {
      "entropy": 5.4907971858978275,
      "epoch": 4.092249527410208,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003416689431767128,
      "loss": 4.944,
      "mean_token_accuracy": 0.20890357792377473,
      "num_tokens": 89868234.0,
      "step": 48710
    },
    {
      "entropy": 5.474100160598755,
      "epoch": 4.092669607225373,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003416403408488229,
      "loss": 4.8294,
      "mean_token_accuracy": 0.2236727237701416,
      "num_tokens": 89876887.0,
      "step": 48715
    },
    {
      "entropy": 5.437413263320923,
      "epoch": 4.093089687040537,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003416117373404338,
      "loss": 4.8948,
      "mean_token_accuracy": 0.21501969844102858,
      "num_tokens": 89885882.0,
      "step": 48720
    },
    {
      "entropy": 5.571390390396118,
      "epoch": 4.093509766855703,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034158313265205195,
      "loss": 5.0635,
      "mean_token_accuracy": 0.19949886202812195,
      "num_tokens": 89895706.0,
      "step": 48725
    },
    {
      "entropy": 5.454870510101318,
      "epoch": 4.093929846670868,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034155452678418424,
      "loss": 4.902,
      "mean_token_accuracy": 0.212904854118824,
      "num_tokens": 89905190.0,
      "step": 48730
    },
    {
      "entropy": 5.416832733154297,
      "epoch": 4.094349926486032,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034152591973733736,
      "loss": 4.784,
      "mean_token_accuracy": 0.21774980872869493,
      "num_tokens": 89913644.0,
      "step": 48735
    },
    {
      "entropy": 5.4783073425292965,
      "epoch": 4.094770006301197,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034149731151201807,
      "loss": 4.9321,
      "mean_token_accuracy": 0.21130862981081008,
      "num_tokens": 89923836.0,
      "step": 48740
    },
    {
      "entropy": 5.455379581451416,
      "epoch": 4.095190086116362,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003414687021087331,
      "loss": 4.8919,
      "mean_token_accuracy": 0.21533976048231124,
      "num_tokens": 89932830.0,
      "step": 48745
    },
    {
      "entropy": 5.429896354675293,
      "epoch": 4.095610165931527,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003414400915279893,
      "loss": 4.8114,
      "mean_token_accuracy": 0.2125555008649826,
      "num_tokens": 89941956.0,
      "step": 48750
    },
    {
      "entropy": 5.51005539894104,
      "epoch": 4.096030245746692,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034141147977029355,
      "loss": 4.9704,
      "mean_token_accuracy": 0.21121810823678971,
      "num_tokens": 89950833.0,
      "step": 48755
    },
    {
      "entropy": 5.438475751876831,
      "epoch": 4.0964503255618565,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034138286683615265,
      "loss": 4.7999,
      "mean_token_accuracy": 0.22513141334056855,
      "num_tokens": 89961090.0,
      "step": 48760
    },
    {
      "entropy": 5.4021697521209715,
      "epoch": 4.096870405377022,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003413542527260734,
      "loss": 4.9248,
      "mean_token_accuracy": 0.20407486110925674,
      "num_tokens": 89970516.0,
      "step": 48765
    },
    {
      "entropy": 5.379996109008789,
      "epoch": 4.097290485192186,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003413256374405627,
      "loss": 4.8505,
      "mean_token_accuracy": 0.21746205687522888,
      "num_tokens": 89980690.0,
      "step": 48770
    },
    {
      "entropy": 5.426535034179688,
      "epoch": 4.097710565007351,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003412970209801276,
      "loss": 4.8349,
      "mean_token_accuracy": 0.2165640890598297,
      "num_tokens": 89989760.0,
      "step": 48775
    },
    {
      "entropy": 5.473084545135498,
      "epoch": 4.0981306448225165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003412684033452748,
      "loss": 4.8787,
      "mean_token_accuracy": 0.21694184541702272,
      "num_tokens": 89999036.0,
      "step": 48780
    },
    {
      "entropy": 5.448706150054932,
      "epoch": 4.098550724637681,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034123978453651146,
      "loss": 4.8786,
      "mean_token_accuracy": 0.21578812897205352,
      "num_tokens": 90009001.0,
      "step": 48785
    },
    {
      "entropy": 5.543638086318969,
      "epoch": 4.098970804452846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003412111645543444,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2089127406477928,
      "num_tokens": 90018014.0,
      "step": 48790
    },
    {
      "entropy": 5.449006652832031,
      "epoch": 4.099390884268011,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034118254339928064,
      "loss": 4.9228,
      "mean_token_accuracy": 0.20724666267633438,
      "num_tokens": 90027754.0,
      "step": 48795
    },
    {
      "entropy": 5.4431041240692135,
      "epoch": 4.099810964083176,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003411539210718271,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21155078560113907,
      "num_tokens": 90036694.0,
      "step": 48800
    },
    {
      "entropy": 5.452693128585816,
      "epoch": 4.100231043898341,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003411252975724911,
      "loss": 4.8748,
      "mean_token_accuracy": 0.21079428791999816,
      "num_tokens": 90045823.0,
      "step": 48805
    },
    {
      "entropy": 5.547165060043335,
      "epoch": 4.100651123713505,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034109667290177937,
      "loss": 4.9717,
      "mean_token_accuracy": 0.20989043712615968,
      "num_tokens": 90055868.0,
      "step": 48810
    },
    {
      "entropy": 5.4203200340271,
      "epoch": 4.1010712035286705,
      "grad_norm": 1.125,
      "learning_rate": 0.0003410680470601991,
      "loss": 4.8343,
      "mean_token_accuracy": 0.21145274490118027,
      "num_tokens": 90065481.0,
      "step": 48815
    },
    {
      "entropy": 5.568504428863525,
      "epoch": 4.101491283343836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003410394200482574,
      "loss": 5.068,
      "mean_token_accuracy": 0.19694897085428237,
      "num_tokens": 90075472.0,
      "step": 48820
    },
    {
      "entropy": 5.405906772613525,
      "epoch": 4.101911363159,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003410107918664615,
      "loss": 4.873,
      "mean_token_accuracy": 0.21246306896209716,
      "num_tokens": 90085134.0,
      "step": 48825
    },
    {
      "entropy": 5.455071496963501,
      "epoch": 4.102331442974165,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003409821625153184,
      "loss": 4.9124,
      "mean_token_accuracy": 0.21734305173158647,
      "num_tokens": 90094807.0,
      "step": 48830
    },
    {
      "entropy": 5.472237586975098,
      "epoch": 4.10275152278933,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003409535319953351,
      "loss": 4.9334,
      "mean_token_accuracy": 0.216220161318779,
      "num_tokens": 90103710.0,
      "step": 48835
    },
    {
      "entropy": 5.454630184173584,
      "epoch": 4.103171602604495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034092490030701914,
      "loss": 4.8535,
      "mean_token_accuracy": 0.21983163356781005,
      "num_tokens": 90112548.0,
      "step": 48840
    },
    {
      "entropy": 5.371830940246582,
      "epoch": 4.10359168241966,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034089626745087735,
      "loss": 4.829,
      "mean_token_accuracy": 0.22186780273914336,
      "num_tokens": 90122082.0,
      "step": 48845
    },
    {
      "entropy": 5.431590175628662,
      "epoch": 4.1040117622348244,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003408676334274173,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21375294923782348,
      "num_tokens": 90131389.0,
      "step": 48850
    },
    {
      "entropy": 5.407285499572754,
      "epoch": 4.10443184204999,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034083899823714593,
      "loss": 4.8518,
      "mean_token_accuracy": 0.21992966085672377,
      "num_tokens": 90140748.0,
      "step": 48855
    },
    {
      "entropy": 5.496236658096313,
      "epoch": 4.104851921865154,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003408103618805706,
      "loss": 4.8878,
      "mean_token_accuracy": 0.21463771015405655,
      "num_tokens": 90150641.0,
      "step": 48860
    },
    {
      "entropy": 5.402340936660766,
      "epoch": 4.105272001680319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003407817243581986,
      "loss": 4.8542,
      "mean_token_accuracy": 0.211268113553524,
      "num_tokens": 90159042.0,
      "step": 48865
    },
    {
      "entropy": 5.427292728424073,
      "epoch": 4.1056920814954845,
      "grad_norm": 1.125,
      "learning_rate": 0.0003407530856705372,
      "loss": 4.8959,
      "mean_token_accuracy": 0.21322524845600127,
      "num_tokens": 90167057.0,
      "step": 48870
    },
    {
      "entropy": 5.4029113292694095,
      "epoch": 4.106112161310649,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003407244458180938,
      "loss": 4.8833,
      "mean_token_accuracy": 0.21388753801584243,
      "num_tokens": 90175093.0,
      "step": 48875
    },
    {
      "entropy": 5.422864007949829,
      "epoch": 4.106532241125814,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003406958048013757,
      "loss": 4.8919,
      "mean_token_accuracy": 0.21046964526176454,
      "num_tokens": 90183646.0,
      "step": 48880
    },
    {
      "entropy": 5.443767309188843,
      "epoch": 4.106952320940978,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003406671626208901,
      "loss": 4.9244,
      "mean_token_accuracy": 0.2213581919670105,
      "num_tokens": 90192616.0,
      "step": 48885
    },
    {
      "entropy": 5.504984712600708,
      "epoch": 4.107372400756144,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034063851927714465,
      "loss": 4.9817,
      "mean_token_accuracy": 0.2025129020214081,
      "num_tokens": 90202815.0,
      "step": 48890
    },
    {
      "entropy": 5.48927674293518,
      "epoch": 4.107792480571309,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003406098747706467,
      "loss": 4.885,
      "mean_token_accuracy": 0.21291518211364746,
      "num_tokens": 90212407.0,
      "step": 48895
    },
    {
      "entropy": 5.426323318481446,
      "epoch": 4.108212560386473,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034058122910190353,
      "loss": 4.9912,
      "mean_token_accuracy": 0.20698896199464797,
      "num_tokens": 90221703.0,
      "step": 48900
    },
    {
      "entropy": 5.502981472015381,
      "epoch": 4.108632640201638,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003405525822714226,
      "loss": 4.9558,
      "mean_token_accuracy": 0.2108398661017418,
      "num_tokens": 90230871.0,
      "step": 48905
    },
    {
      "entropy": 5.443475818634033,
      "epoch": 4.109052720016803,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003405239342797115,
      "loss": 4.8181,
      "mean_token_accuracy": 0.21839816272258758,
      "num_tokens": 90240062.0,
      "step": 48910
    },
    {
      "entropy": 5.488246440887451,
      "epoch": 4.109472799831968,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003404952851272776,
      "loss": 5.0025,
      "mean_token_accuracy": 0.21373023539781572,
      "num_tokens": 90248789.0,
      "step": 48915
    },
    {
      "entropy": 5.513105964660644,
      "epoch": 4.109892879647133,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003404666348146285,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2094008207321167,
      "num_tokens": 90258062.0,
      "step": 48920
    },
    {
      "entropy": 5.469409418106079,
      "epoch": 4.110312959462298,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034043798334227167,
      "loss": 4.8509,
      "mean_token_accuracy": 0.215934856235981,
      "num_tokens": 90267444.0,
      "step": 48925
    },
    {
      "entropy": 5.403511381149292,
      "epoch": 4.110733039277463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034040933071071463,
      "loss": 4.8352,
      "mean_token_accuracy": 0.21810789555311202,
      "num_tokens": 90275880.0,
      "step": 48930
    },
    {
      "entropy": 5.443936443328857,
      "epoch": 4.111153119092627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034038067692046493,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21616844683885575,
      "num_tokens": 90285698.0,
      "step": 48935
    },
    {
      "entropy": 5.4655698299407955,
      "epoch": 4.111573198907792,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003403520219720303,
      "loss": 4.8984,
      "mean_token_accuracy": 0.20692733079195022,
      "num_tokens": 90295635.0,
      "step": 48940
    },
    {
      "entropy": 5.430548143386841,
      "epoch": 4.111993278722958,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003403233658659182,
      "loss": 4.9614,
      "mean_token_accuracy": 0.207371024787426,
      "num_tokens": 90304734.0,
      "step": 48945
    },
    {
      "entropy": 5.438444137573242,
      "epoch": 4.112413358538122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003402947086026363,
      "loss": 4.9141,
      "mean_token_accuracy": 0.2102106988430023,
      "num_tokens": 90313511.0,
      "step": 48950
    },
    {
      "entropy": 5.468373346328735,
      "epoch": 4.112833438353287,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034026605018269224,
      "loss": 4.8879,
      "mean_token_accuracy": 0.21068360656499863,
      "num_tokens": 90323144.0,
      "step": 48955
    },
    {
      "entropy": 5.4760987758636475,
      "epoch": 4.113253518168452,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003402373906065938,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2176375225186348,
      "num_tokens": 90332045.0,
      "step": 48960
    },
    {
      "entropy": 5.410698270797729,
      "epoch": 4.113673597983617,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034020872987484854,
      "loss": 4.9132,
      "mean_token_accuracy": 0.21610239893198013,
      "num_tokens": 90340780.0,
      "step": 48965
    },
    {
      "entropy": 5.447531604766846,
      "epoch": 4.114093677798782,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034018006798796415,
      "loss": 4.8472,
      "mean_token_accuracy": 0.22495196908712387,
      "num_tokens": 90350659.0,
      "step": 48970
    },
    {
      "entropy": 5.435651874542236,
      "epoch": 4.114513757613946,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034015140494644846,
      "loss": 4.8305,
      "mean_token_accuracy": 0.2194320172071457,
      "num_tokens": 90359777.0,
      "step": 48975
    },
    {
      "entropy": 5.525208616256714,
      "epoch": 4.114933837429112,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003401227407508092,
      "loss": 4.991,
      "mean_token_accuracy": 0.21361469477415085,
      "num_tokens": 90369454.0,
      "step": 48980
    },
    {
      "entropy": 5.5002035140991214,
      "epoch": 4.115353917244277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003400940754015541,
      "loss": 4.9169,
      "mean_token_accuracy": 0.21388550400733947,
      "num_tokens": 90379349.0,
      "step": 48985
    },
    {
      "entropy": 5.4763072490692135,
      "epoch": 4.115773997059441,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000340065408899191,
      "loss": 4.9318,
      "mean_token_accuracy": 0.2090523585677147,
      "num_tokens": 90387510.0,
      "step": 48990
    },
    {
      "entropy": 5.473828601837158,
      "epoch": 4.116194076874606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034003674124422767,
      "loss": 4.8676,
      "mean_token_accuracy": 0.2195633813738823,
      "num_tokens": 90398550.0,
      "step": 48995
    },
    {
      "entropy": 5.434631729125977,
      "epoch": 4.116614156689771,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003400080724371719,
      "loss": 4.8621,
      "mean_token_accuracy": 0.21521117985248567,
      "num_tokens": 90407404.0,
      "step": 49000
    },
    {
      "entropy": 5.442934036254883,
      "epoch": 4.117034236504936,
      "grad_norm": 1.125,
      "learning_rate": 0.0003399794024785317,
      "loss": 4.9077,
      "mean_token_accuracy": 0.20963890105485916,
      "num_tokens": 90416918.0,
      "step": 49005
    },
    {
      "entropy": 5.4704619407653805,
      "epoch": 4.117454316320101,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003399507313688148,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21464958637952805,
      "num_tokens": 90426902.0,
      "step": 49010
    },
    {
      "entropy": 5.409541988372803,
      "epoch": 4.1178743961352655,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033992205910852904,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22227884083986282,
      "num_tokens": 90436461.0,
      "step": 49015
    },
    {
      "entropy": 5.439622068405152,
      "epoch": 4.118294475950431,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033989338569818256,
      "loss": 4.8952,
      "mean_token_accuracy": 0.2145725518465042,
      "num_tokens": 90446398.0,
      "step": 49020
    },
    {
      "entropy": 5.551218938827515,
      "epoch": 4.118714555765595,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003398647111382831,
      "loss": 4.9716,
      "mean_token_accuracy": 0.21190235316753386,
      "num_tokens": 90456390.0,
      "step": 49025
    },
    {
      "entropy": 5.416483402252197,
      "epoch": 4.11913463558076,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003398360354293387,
      "loss": 4.8915,
      "mean_token_accuracy": 0.21469003856182098,
      "num_tokens": 90465932.0,
      "step": 49030
    },
    {
      "entropy": 5.442652654647827,
      "epoch": 4.1195547153959255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033980735857185734,
      "loss": 4.9466,
      "mean_token_accuracy": 0.2111184850335121,
      "num_tokens": 90474899.0,
      "step": 49035
    },
    {
      "entropy": 5.504509496688843,
      "epoch": 4.11997479521109,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000339778680566347,
      "loss": 4.9459,
      "mean_token_accuracy": 0.2154335454106331,
      "num_tokens": 90483163.0,
      "step": 49040
    },
    {
      "entropy": 5.508759784698486,
      "epoch": 4.120394875026255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003397500014133157,
      "loss": 4.8924,
      "mean_token_accuracy": 0.2160376563668251,
      "num_tokens": 90492599.0,
      "step": 49045
    },
    {
      "entropy": 5.528632259368896,
      "epoch": 4.1208149548414195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003397213211132714,
      "loss": 4.9407,
      "mean_token_accuracy": 0.20881386399269103,
      "num_tokens": 90501229.0,
      "step": 49050
    },
    {
      "entropy": 5.527171802520752,
      "epoch": 4.121235034656585,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003396926396667223,
      "loss": 4.9851,
      "mean_token_accuracy": 0.21299864053726197,
      "num_tokens": 90510964.0,
      "step": 49055
    },
    {
      "entropy": 5.460836172103882,
      "epoch": 4.12165511447175,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033966395707417634,
      "loss": 4.9198,
      "mean_token_accuracy": 0.2126084089279175,
      "num_tokens": 90520456.0,
      "step": 49060
    },
    {
      "entropy": 5.472267770767212,
      "epoch": 4.122075194286914,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033963527333614167,
      "loss": 4.9391,
      "mean_token_accuracy": 0.21028509587049485,
      "num_tokens": 90530180.0,
      "step": 49065
    },
    {
      "entropy": 5.444977951049805,
      "epoch": 4.1224952741020795,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003396065884531265,
      "loss": 4.8577,
      "mean_token_accuracy": 0.21967940777540207,
      "num_tokens": 90539718.0,
      "step": 49070
    },
    {
      "entropy": 5.411546230316162,
      "epoch": 4.122915353917244,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003395779024256388,
      "loss": 4.8793,
      "mean_token_accuracy": 0.21321499943733216,
      "num_tokens": 90548919.0,
      "step": 49075
    },
    {
      "entropy": 5.445837450027466,
      "epoch": 4.123335433732409,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003395492152541869,
      "loss": 4.9562,
      "mean_token_accuracy": 0.20321591794490815,
      "num_tokens": 90558082.0,
      "step": 49080
    },
    {
      "entropy": 5.496758890151978,
      "epoch": 4.123755513547574,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003395205269392788,
      "loss": 4.9691,
      "mean_token_accuracy": 0.2178713247179985,
      "num_tokens": 90567437.0,
      "step": 49085
    },
    {
      "entropy": 5.405319881439209,
      "epoch": 4.124175593362739,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033949183748142284,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21429071575403214,
      "num_tokens": 90576932.0,
      "step": 49090
    },
    {
      "entropy": 5.4657481670379635,
      "epoch": 4.124595673177904,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033946314688112713,
      "loss": 4.9698,
      "mean_token_accuracy": 0.20792529881000518,
      "num_tokens": 90586011.0,
      "step": 49095
    },
    {
      "entropy": 5.489633798599243,
      "epoch": 4.125015752993069,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003394344551389,
      "loss": 4.9224,
      "mean_token_accuracy": 0.21362358033657075,
      "num_tokens": 90594499.0,
      "step": 49100
    },
    {
      "entropy": 5.433336400985718,
      "epoch": 4.1254358328082334,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033940576225524973,
      "loss": 4.9081,
      "mean_token_accuracy": 0.2104132130742073,
      "num_tokens": 90603877.0,
      "step": 49105
    },
    {
      "entropy": 5.460953617095948,
      "epoch": 4.125855912623399,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003393770682306844,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21144997030496598,
      "num_tokens": 90612912.0,
      "step": 49110
    },
    {
      "entropy": 5.472721767425537,
      "epoch": 4.126275992438563,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003393483730657126,
      "loss": 4.9002,
      "mean_token_accuracy": 0.21153903305530547,
      "num_tokens": 90622261.0,
      "step": 49115
    },
    {
      "entropy": 5.362814283370971,
      "epoch": 4.126696072253728,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033931967676084244,
      "loss": 4.855,
      "mean_token_accuracy": 0.2230096861720085,
      "num_tokens": 90632122.0,
      "step": 49120
    },
    {
      "entropy": 5.46785979270935,
      "epoch": 4.1271161520688935,
      "grad_norm": 1.125,
      "learning_rate": 0.0003392909793165824,
      "loss": 4.9118,
      "mean_token_accuracy": 0.21660617887973785,
      "num_tokens": 90640797.0,
      "step": 49125
    },
    {
      "entropy": 5.3359026432037355,
      "epoch": 4.127536231884058,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003392622807334407,
      "loss": 4.7807,
      "mean_token_accuracy": 0.22221786826848983,
      "num_tokens": 90649671.0,
      "step": 49130
    },
    {
      "entropy": 5.3691479682922365,
      "epoch": 4.127956311699223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033923358101192573,
      "loss": 4.8369,
      "mean_token_accuracy": 0.23003667443990708,
      "num_tokens": 90658882.0,
      "step": 49135
    },
    {
      "entropy": 5.521657514572143,
      "epoch": 4.128376391514387,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000339204880152546,
      "loss": 4.9161,
      "mean_token_accuracy": 0.21343975514173508,
      "num_tokens": 90667435.0,
      "step": 49140
    },
    {
      "entropy": 5.43339352607727,
      "epoch": 4.128796471329553,
      "grad_norm": 1.125,
      "learning_rate": 0.00033917617815580994,
      "loss": 4.8845,
      "mean_token_accuracy": 0.20700767934322356,
      "num_tokens": 90677385.0,
      "step": 49145
    },
    {
      "entropy": 5.497691440582275,
      "epoch": 4.129216551144718,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003391474750222258,
      "loss": 5.0065,
      "mean_token_accuracy": 0.19987151622772217,
      "num_tokens": 90686567.0,
      "step": 49150
    },
    {
      "entropy": 5.478266191482544,
      "epoch": 4.129636630959882,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003391187707523023,
      "loss": 4.8509,
      "mean_token_accuracy": 0.2145782083272934,
      "num_tokens": 90695588.0,
      "step": 49155
    },
    {
      "entropy": 5.4349112033844,
      "epoch": 4.130056710775047,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00033909006534654773,
      "loss": 4.8926,
      "mean_token_accuracy": 0.2117029011249542,
      "num_tokens": 90705305.0,
      "step": 49160
    },
    {
      "entropy": 5.432046365737915,
      "epoch": 4.130476790590212,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033906135880547067,
      "loss": 4.871,
      "mean_token_accuracy": 0.21811338216066362,
      "num_tokens": 90714942.0,
      "step": 49165
    },
    {
      "entropy": 5.463308668136596,
      "epoch": 4.130896870405377,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033903265112957963,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20889516919851303,
      "num_tokens": 90724792.0,
      "step": 49170
    },
    {
      "entropy": 5.416538572311401,
      "epoch": 4.131316950220542,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033900394231938316,
      "loss": 4.8513,
      "mean_token_accuracy": 0.22062987387180327,
      "num_tokens": 90734399.0,
      "step": 49175
    },
    {
      "entropy": 5.455066871643067,
      "epoch": 4.131737030035707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003389752323753898,
      "loss": 4.86,
      "mean_token_accuracy": 0.21671660840511323,
      "num_tokens": 90743455.0,
      "step": 49180
    },
    {
      "entropy": 5.546169567108154,
      "epoch": 4.132157109850872,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003389465212981081,
      "loss": 5.0738,
      "mean_token_accuracy": 0.20256628841161728,
      "num_tokens": 90753402.0,
      "step": 49185
    },
    {
      "entropy": 5.46407995223999,
      "epoch": 4.132577189666036,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033891780908804677,
      "loss": 4.9066,
      "mean_token_accuracy": 0.21003707200288774,
      "num_tokens": 90762066.0,
      "step": 49190
    },
    {
      "entropy": 5.471004676818848,
      "epoch": 4.132997269481201,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003388890957457143,
      "loss": 4.9059,
      "mean_token_accuracy": 0.21461438983678818,
      "num_tokens": 90770255.0,
      "step": 49195
    },
    {
      "entropy": 5.421842288970947,
      "epoch": 4.133417349296367,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033886038127161944,
      "loss": 4.8197,
      "mean_token_accuracy": 0.2217722937464714,
      "num_tokens": 90779166.0,
      "step": 49200
    },
    {
      "entropy": 5.444605112075806,
      "epoch": 4.133837429111531,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033883166566627085,
      "loss": 4.8877,
      "mean_token_accuracy": 0.21234841644763947,
      "num_tokens": 90788616.0,
      "step": 49205
    },
    {
      "entropy": 5.480521488189697,
      "epoch": 4.134257508926696,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033880294893017703,
      "loss": 4.9615,
      "mean_token_accuracy": 0.20958988815546037,
      "num_tokens": 90798738.0,
      "step": 49210
    },
    {
      "entropy": 5.453914022445678,
      "epoch": 4.1346775887418605,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003387742310638469,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2118554964661598,
      "num_tokens": 90808579.0,
      "step": 49215
    },
    {
      "entropy": 5.521174478530884,
      "epoch": 4.135097668557026,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033874551206778916,
      "loss": 4.888,
      "mean_token_accuracy": 0.21386638432741165,
      "num_tokens": 90818197.0,
      "step": 49220
    },
    {
      "entropy": 5.413176870346069,
      "epoch": 4.135517748372191,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003387167919425124,
      "loss": 4.83,
      "mean_token_accuracy": 0.21724434942007065,
      "num_tokens": 90828141.0,
      "step": 49225
    },
    {
      "entropy": 5.44381275177002,
      "epoch": 4.135937828187355,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033868807068852553,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21569108366966247,
      "num_tokens": 90837519.0,
      "step": 49230
    },
    {
      "entropy": 5.427679252624512,
      "epoch": 4.136357908002521,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033865934830633733,
      "loss": 4.8923,
      "mean_token_accuracy": 0.2201737031340599,
      "num_tokens": 90846361.0,
      "step": 49235
    },
    {
      "entropy": 5.4542454242706295,
      "epoch": 4.136777987817686,
      "grad_norm": 1.125,
      "learning_rate": 0.00033863062479645646,
      "loss": 4.8967,
      "mean_token_accuracy": 0.22194981724023818,
      "num_tokens": 90855719.0,
      "step": 49240
    },
    {
      "entropy": 5.452984380722046,
      "epoch": 4.13719806763285,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003386019001593919,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2267346426844597,
      "num_tokens": 90865834.0,
      "step": 49245
    },
    {
      "entropy": 5.544527053833008,
      "epoch": 4.137618147448015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033857317439565243,
      "loss": 5.0614,
      "mean_token_accuracy": 0.19822860658168792,
      "num_tokens": 90875938.0,
      "step": 49250
    },
    {
      "entropy": 5.533943796157837,
      "epoch": 4.13803822726318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033854444750574684,
      "loss": 4.9203,
      "mean_token_accuracy": 0.20860413908958436,
      "num_tokens": 90885157.0,
      "step": 49255
    },
    {
      "entropy": 5.505201435089111,
      "epoch": 4.138458307078345,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003385157194901842,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21573213934898378,
      "num_tokens": 90893955.0,
      "step": 49260
    },
    {
      "entropy": 5.416156864166259,
      "epoch": 4.13887838689351,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003384869903494732,
      "loss": 4.9693,
      "mean_token_accuracy": 0.2134597197175026,
      "num_tokens": 90903194.0,
      "step": 49265
    },
    {
      "entropy": 5.46958327293396,
      "epoch": 4.1392984667086745,
      "grad_norm": 1.125,
      "learning_rate": 0.0003384582600841229,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21749838441610336,
      "num_tokens": 90912293.0,
      "step": 49270
    },
    {
      "entropy": 5.426399517059326,
      "epoch": 4.13971854652384,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033842952869464216,
      "loss": 4.7663,
      "mean_token_accuracy": 0.22693876177072525,
      "num_tokens": 90921500.0,
      "step": 49275
    },
    {
      "entropy": 5.483104181289673,
      "epoch": 4.140138626339004,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033840079618154006,
      "loss": 4.9857,
      "mean_token_accuracy": 0.2127663254737854,
      "num_tokens": 90930557.0,
      "step": 49280
    },
    {
      "entropy": 5.441448974609375,
      "epoch": 4.140558706154169,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033837206254532545,
      "loss": 4.8971,
      "mean_token_accuracy": 0.21301811933517456,
      "num_tokens": 90939711.0,
      "step": 49285
    },
    {
      "entropy": 5.4494068145751955,
      "epoch": 4.1409787859693346,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003383433277865073,
      "loss": 4.8733,
      "mean_token_accuracy": 0.21268670558929442,
      "num_tokens": 90949029.0,
      "step": 49290
    },
    {
      "entropy": 5.375485134124756,
      "epoch": 4.141398865784499,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003383145919055949,
      "loss": 4.767,
      "mean_token_accuracy": 0.21445381194353103,
      "num_tokens": 90957219.0,
      "step": 49295
    },
    {
      "entropy": 5.4837809085845945,
      "epoch": 4.141818945599664,
      "grad_norm": 1.25,
      "learning_rate": 0.00033828585490309695,
      "loss": 4.9271,
      "mean_token_accuracy": 0.2124520167708397,
      "num_tokens": 90965921.0,
      "step": 49300
    },
    {
      "entropy": 5.47648606300354,
      "epoch": 4.1422390254148285,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003382571167795227,
      "loss": 4.9325,
      "mean_token_accuracy": 0.22052749395370483,
      "num_tokens": 90974253.0,
      "step": 49305
    },
    {
      "entropy": 5.466106796264649,
      "epoch": 4.142659105229994,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033822837753538125,
      "loss": 4.9672,
      "mean_token_accuracy": 0.2117884263396263,
      "num_tokens": 90983609.0,
      "step": 49310
    },
    {
      "entropy": 5.517012119293213,
      "epoch": 4.143079185045159,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003381996371711816,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21524217277765273,
      "num_tokens": 90992691.0,
      "step": 49315
    },
    {
      "entropy": 5.453322649002075,
      "epoch": 4.143499264860323,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033817089568743296,
      "loss": 4.8726,
      "mean_token_accuracy": 0.21135242879390717,
      "num_tokens": 91001986.0,
      "step": 49320
    },
    {
      "entropy": 5.409732389450073,
      "epoch": 4.1439193446754885,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033814215308464445,
      "loss": 4.8142,
      "mean_token_accuracy": 0.21880849003791808,
      "num_tokens": 91011557.0,
      "step": 49325
    },
    {
      "entropy": 5.434754943847656,
      "epoch": 4.144339424490653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003381134093633251,
      "loss": 4.8536,
      "mean_token_accuracy": 0.22304107248783112,
      "num_tokens": 91020010.0,
      "step": 49330
    },
    {
      "entropy": 5.425201559066773,
      "epoch": 4.144759504305818,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033808466452398434,
      "loss": 4.9252,
      "mean_token_accuracy": 0.20765574872493744,
      "num_tokens": 91030021.0,
      "step": 49335
    },
    {
      "entropy": 5.4477475643157955,
      "epoch": 4.145179584120983,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003380559185671312,
      "loss": 4.8448,
      "mean_token_accuracy": 0.2202816292643547,
      "num_tokens": 91038185.0,
      "step": 49340
    },
    {
      "entropy": 5.40934042930603,
      "epoch": 4.145599663936148,
      "grad_norm": 1.125,
      "learning_rate": 0.0003380271714932749,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22015077620744705,
      "num_tokens": 91047828.0,
      "step": 49345
    },
    {
      "entropy": 5.461309099197388,
      "epoch": 4.146019743751313,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033799842330292467,
      "loss": 4.8557,
      "mean_token_accuracy": 0.21529140323400497,
      "num_tokens": 91056961.0,
      "step": 49350
    },
    {
      "entropy": 5.439701652526855,
      "epoch": 4.146439823566477,
      "grad_norm": 1.125,
      "learning_rate": 0.0003379696739965898,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2154422238469124,
      "num_tokens": 91066463.0,
      "step": 49355
    },
    {
      "entropy": 5.356677770614624,
      "epoch": 4.1468599033816425,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003379409235747797,
      "loss": 4.7577,
      "mean_token_accuracy": 0.2294149413704872,
      "num_tokens": 91074691.0,
      "step": 49360
    },
    {
      "entropy": 5.382204914093018,
      "epoch": 4.147279983196808,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033791217203800347,
      "loss": 4.8097,
      "mean_token_accuracy": 0.2203153908252716,
      "num_tokens": 91084667.0,
      "step": 49365
    },
    {
      "entropy": 5.430002880096436,
      "epoch": 4.147700063011972,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003378834193867705,
      "loss": 4.8237,
      "mean_token_accuracy": 0.21387601792812347,
      "num_tokens": 91093429.0,
      "step": 49370
    },
    {
      "entropy": 5.494110918045044,
      "epoch": 4.148120142827137,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003378546656215902,
      "loss": 4.9784,
      "mean_token_accuracy": 0.21592427641153336,
      "num_tokens": 91102406.0,
      "step": 49375
    },
    {
      "entropy": 5.424199485778809,
      "epoch": 4.148540222642302,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033782591074297175,
      "loss": 4.8629,
      "mean_token_accuracy": 0.21563737988471984,
      "num_tokens": 91112154.0,
      "step": 49380
    },
    {
      "entropy": 5.43016881942749,
      "epoch": 4.148960302457467,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003377971547514247,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21010666340589523,
      "num_tokens": 91122309.0,
      "step": 49385
    },
    {
      "entropy": 5.436651086807251,
      "epoch": 4.149380382272632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033776839764745843,
      "loss": 4.9048,
      "mean_token_accuracy": 0.21043651849031447,
      "num_tokens": 91130573.0,
      "step": 49390
    },
    {
      "entropy": 5.536124038696289,
      "epoch": 4.149800462087796,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033773963943158227,
      "loss": 4.9819,
      "mean_token_accuracy": 0.21568571031093597,
      "num_tokens": 91140621.0,
      "step": 49395
    },
    {
      "entropy": 5.403270483016968,
      "epoch": 4.150220541902962,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033771088010430575,
      "loss": 4.848,
      "mean_token_accuracy": 0.22340443581342698,
      "num_tokens": 91148952.0,
      "step": 49400
    },
    {
      "entropy": 5.3841996669769285,
      "epoch": 4.150640621718127,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033768211966613823,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21399164199829102,
      "num_tokens": 91157479.0,
      "step": 49405
    },
    {
      "entropy": 5.452832889556885,
      "epoch": 4.151060701533291,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003376533581175893,
      "loss": 4.8882,
      "mean_token_accuracy": 0.21453724503517152,
      "num_tokens": 91166035.0,
      "step": 49410
    },
    {
      "entropy": 5.536070203781128,
      "epoch": 4.151480781348456,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003376245954591683,
      "loss": 5.011,
      "mean_token_accuracy": 0.20416174381971358,
      "num_tokens": 91175948.0,
      "step": 49415
    },
    {
      "entropy": 5.469054222106934,
      "epoch": 4.151900861163621,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003375958316913849,
      "loss": 4.8879,
      "mean_token_accuracy": 0.21952632069587708,
      "num_tokens": 91185651.0,
      "step": 49420
    },
    {
      "entropy": 5.434456920623779,
      "epoch": 4.152320940978786,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033756706681474855,
      "loss": 4.9223,
      "mean_token_accuracy": 0.21161147356033325,
      "num_tokens": 91195035.0,
      "step": 49425
    },
    {
      "entropy": 5.413233709335327,
      "epoch": 4.152741020793951,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033753830082976883,
      "loss": 4.9492,
      "mean_token_accuracy": 0.21140142232179643,
      "num_tokens": 91205009.0,
      "step": 49430
    },
    {
      "entropy": 5.487736749649048,
      "epoch": 4.153161100609116,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033750953373695524,
      "loss": 4.9959,
      "mean_token_accuracy": 0.2087916612625122,
      "num_tokens": 91214303.0,
      "step": 49435
    },
    {
      "entropy": 5.439702796936035,
      "epoch": 4.153581180424281,
      "grad_norm": 1.125,
      "learning_rate": 0.0003374807655368176,
      "loss": 4.832,
      "mean_token_accuracy": 0.217040354013443,
      "num_tokens": 91223461.0,
      "step": 49440
    },
    {
      "entropy": 5.352570009231568,
      "epoch": 4.154001260239445,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003374519962298653,
      "loss": 4.7651,
      "mean_token_accuracy": 0.22528192400932312,
      "num_tokens": 91231082.0,
      "step": 49445
    },
    {
      "entropy": 5.454906034469604,
      "epoch": 4.15442134005461,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033742322581660803,
      "loss": 4.986,
      "mean_token_accuracy": 0.20954966247081758,
      "num_tokens": 91240781.0,
      "step": 49450
    },
    {
      "entropy": 5.404587936401367,
      "epoch": 4.154841419869776,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003373944542975555,
      "loss": 4.8449,
      "mean_token_accuracy": 0.2161679282784462,
      "num_tokens": 91250027.0,
      "step": 49455
    },
    {
      "entropy": 5.480359125137329,
      "epoch": 4.15526149968494,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003373656816732173,
      "loss": 4.911,
      "mean_token_accuracy": 0.21056126356124877,
      "num_tokens": 91260255.0,
      "step": 49460
    },
    {
      "entropy": 5.532597637176513,
      "epoch": 4.155681579500105,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003373369079441032,
      "loss": 4.9432,
      "mean_token_accuracy": 0.2134013995528221,
      "num_tokens": 91269714.0,
      "step": 49465
    },
    {
      "entropy": 5.488373136520385,
      "epoch": 4.1561016593152695,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003373081331107229,
      "loss": 4.937,
      "mean_token_accuracy": 0.21360894441604614,
      "num_tokens": 91279137.0,
      "step": 49470
    },
    {
      "entropy": 5.410505533218384,
      "epoch": 4.156521739130435,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033727935717358596,
      "loss": 4.8535,
      "mean_token_accuracy": 0.21539751142263414,
      "num_tokens": 91286901.0,
      "step": 49475
    },
    {
      "entropy": 5.477430534362793,
      "epoch": 4.1569418189456,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003372505801332024,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21053725033998488,
      "num_tokens": 91296541.0,
      "step": 49480
    },
    {
      "entropy": 5.428315258026123,
      "epoch": 4.157361898760764,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033722180199008184,
      "loss": 4.8761,
      "mean_token_accuracy": 0.21350030601024628,
      "num_tokens": 91305598.0,
      "step": 49485
    },
    {
      "entropy": 5.419946479797363,
      "epoch": 4.15778197857593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033719302274473406,
      "loss": 4.8724,
      "mean_token_accuracy": 0.21522714346647262,
      "num_tokens": 91314442.0,
      "step": 49490
    },
    {
      "entropy": 5.462004470825195,
      "epoch": 4.158202058391094,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000337164242397669,
      "loss": 4.9184,
      "mean_token_accuracy": 0.20921044796705246,
      "num_tokens": 91323943.0,
      "step": 49495
    },
    {
      "entropy": 5.455473518371582,
      "epoch": 4.158622138206259,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003371354609493963,
      "loss": 4.8757,
      "mean_token_accuracy": 0.21090578585863112,
      "num_tokens": 91333051.0,
      "step": 49500
    },
    {
      "entropy": 5.463384580612183,
      "epoch": 4.159042218021424,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003371066784004259,
      "loss": 4.876,
      "mean_token_accuracy": 0.21844634860754014,
      "num_tokens": 91342066.0,
      "step": 49505
    },
    {
      "entropy": 5.457704830169678,
      "epoch": 4.159462297836589,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003370778947512678,
      "loss": 4.8819,
      "mean_token_accuracy": 0.2148846134543419,
      "num_tokens": 91350910.0,
      "step": 49510
    },
    {
      "entropy": 5.505492305755615,
      "epoch": 4.159882377651754,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033704911000243166,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20416591614484786,
      "num_tokens": 91360622.0,
      "step": 49515
    },
    {
      "entropy": 5.450218963623047,
      "epoch": 4.160302457466918,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033702032415442747,
      "loss": 4.9866,
      "mean_token_accuracy": 0.2099937379360199,
      "num_tokens": 91370063.0,
      "step": 49520
    },
    {
      "entropy": 5.541170740127564,
      "epoch": 4.1607225372820835,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003369915372077652,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21249401271343232,
      "num_tokens": 91380345.0,
      "step": 49525
    },
    {
      "entropy": 5.497491407394409,
      "epoch": 4.161142617097249,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003369627491629548,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21131690144538878,
      "num_tokens": 91390561.0,
      "step": 49530
    },
    {
      "entropy": 5.501194334030151,
      "epoch": 4.161562696912413,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003369339600205062,
      "loss": 4.9641,
      "mean_token_accuracy": 0.20918680876493453,
      "num_tokens": 91398616.0,
      "step": 49535
    },
    {
      "entropy": 5.482608509063721,
      "epoch": 4.161982776727578,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003369051697809294,
      "loss": 4.9642,
      "mean_token_accuracy": 0.2079118713736534,
      "num_tokens": 91408182.0,
      "step": 49540
    },
    {
      "entropy": 5.421412181854248,
      "epoch": 4.162402856542743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003368763784447343,
      "loss": 4.8421,
      "mean_token_accuracy": 0.21774207651615143,
      "num_tokens": 91416639.0,
      "step": 49545
    },
    {
      "entropy": 5.496817684173584,
      "epoch": 4.162822936357908,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033684758601243116,
      "loss": 4.9286,
      "mean_token_accuracy": 0.20626967251300812,
      "num_tokens": 91425433.0,
      "step": 49550
    },
    {
      "entropy": 5.476628398895263,
      "epoch": 4.163243016173073,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003368187924845297,
      "loss": 4.8931,
      "mean_token_accuracy": 0.21958423256874085,
      "num_tokens": 91435134.0,
      "step": 49555
    },
    {
      "entropy": 5.422406911849976,
      "epoch": 4.1636630959882375,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003367899978615403,
      "loss": 4.8521,
      "mean_token_accuracy": 0.21227235496044158,
      "num_tokens": 91444652.0,
      "step": 49560
    },
    {
      "entropy": 5.473106813430786,
      "epoch": 4.164083175803403,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003367612021439728,
      "loss": 4.9026,
      "mean_token_accuracy": 0.21487175673246384,
      "num_tokens": 91453243.0,
      "step": 49565
    },
    {
      "entropy": 5.50023775100708,
      "epoch": 4.164503255618568,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003367324053323376,
      "loss": 4.9766,
      "mean_token_accuracy": 0.1966303199529648,
      "num_tokens": 91462448.0,
      "step": 49570
    },
    {
      "entropy": 5.460331392288208,
      "epoch": 4.164923335433732,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033670360742714444,
      "loss": 4.9627,
      "mean_token_accuracy": 0.20664464831352233,
      "num_tokens": 91471314.0,
      "step": 49575
    },
    {
      "entropy": 5.56047945022583,
      "epoch": 4.1653434152488975,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033667480842890376,
      "loss": 4.9823,
      "mean_token_accuracy": 0.2136394813656807,
      "num_tokens": 91481032.0,
      "step": 49580
    },
    {
      "entropy": 5.422235441207886,
      "epoch": 4.165763495064062,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003366460083381255,
      "loss": 4.7537,
      "mean_token_accuracy": 0.22108454406261444,
      "num_tokens": 91489383.0,
      "step": 49585
    },
    {
      "entropy": 5.426509714126587,
      "epoch": 4.166183574879227,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00033661720715532,
      "loss": 4.8862,
      "mean_token_accuracy": 0.2145221412181854,
      "num_tokens": 91498715.0,
      "step": 49590
    },
    {
      "entropy": 5.361617708206177,
      "epoch": 4.166603654694392,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033658840488099746,
      "loss": 4.8984,
      "mean_token_accuracy": 0.2185390040278435,
      "num_tokens": 91508012.0,
      "step": 49595
    },
    {
      "entropy": 5.513964700698852,
      "epoch": 4.167023734509557,
      "grad_norm": 1.015625,
      "learning_rate": 0.000336559601515668,
      "loss": 4.991,
      "mean_token_accuracy": 0.20532617419958116,
      "num_tokens": 91518333.0,
      "step": 49600
    },
    {
      "entropy": 5.480838298797607,
      "epoch": 4.167443814324722,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003365307970598419,
      "loss": 4.8191,
      "mean_token_accuracy": 0.22356561124324797,
      "num_tokens": 91527043.0,
      "step": 49605
    },
    {
      "entropy": 5.531879901885986,
      "epoch": 4.167863894139886,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033650199151402936,
      "loss": 5.0359,
      "mean_token_accuracy": 0.20228552222251892,
      "num_tokens": 91536218.0,
      "step": 49610
    },
    {
      "entropy": 5.4345550537109375,
      "epoch": 4.1682839739550515,
      "grad_norm": 1.546875,
      "learning_rate": 0.0003364731848787408,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22103408575057984,
      "num_tokens": 91545197.0,
      "step": 49615
    },
    {
      "entropy": 5.464449501037597,
      "epoch": 4.168704053770217,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003364443771544863,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2269335001707077,
      "num_tokens": 91554518.0,
      "step": 49620
    },
    {
      "entropy": 5.364137077331543,
      "epoch": 4.169124133585381,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003364155683417763,
      "loss": 4.8054,
      "mean_token_accuracy": 0.22151668518781661,
      "num_tokens": 91564061.0,
      "step": 49625
    },
    {
      "entropy": 5.490591669082642,
      "epoch": 4.169544213400546,
      "grad_norm": 1.125,
      "learning_rate": 0.0003363867584411212,
      "loss": 4.9053,
      "mean_token_accuracy": 0.21892084330320358,
      "num_tokens": 91572353.0,
      "step": 49630
    },
    {
      "entropy": 5.421031332015991,
      "epoch": 4.169964293215711,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003363579474530313,
      "loss": 4.8615,
      "mean_token_accuracy": 0.21949559450149536,
      "num_tokens": 91580315.0,
      "step": 49635
    },
    {
      "entropy": 5.437713146209717,
      "epoch": 4.170384373030876,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033632913537801697,
      "loss": 4.9084,
      "mean_token_accuracy": 0.216676265001297,
      "num_tokens": 91588553.0,
      "step": 49640
    },
    {
      "entropy": 5.526314640045166,
      "epoch": 4.170804452846041,
      "grad_norm": 1.125,
      "learning_rate": 0.0003363003222165885,
      "loss": 5.0397,
      "mean_token_accuracy": 0.20481038838624954,
      "num_tokens": 91597293.0,
      "step": 49645
    },
    {
      "entropy": 5.510669088363647,
      "epoch": 4.171224532661205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033627150796925647,
      "loss": 5.0277,
      "mean_token_accuracy": 0.2088194340467453,
      "num_tokens": 91607332.0,
      "step": 49650
    },
    {
      "entropy": 5.472129249572754,
      "epoch": 4.171644612476371,
      "grad_norm": 1.125,
      "learning_rate": 0.00033624269263653124,
      "loss": 4.8872,
      "mean_token_accuracy": 0.21342216283082963,
      "num_tokens": 91616635.0,
      "step": 49655
    },
    {
      "entropy": 5.462571668624878,
      "epoch": 4.172064692291535,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003362138762189232,
      "loss": 4.8676,
      "mean_token_accuracy": 0.21873790621757508,
      "num_tokens": 91626429.0,
      "step": 49660
    },
    {
      "entropy": 5.509075260162353,
      "epoch": 4.1724847721067,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003361850587169429,
      "loss": 4.9368,
      "mean_token_accuracy": 0.2134292960166931,
      "num_tokens": 91635509.0,
      "step": 49665
    },
    {
      "entropy": 5.548685836791992,
      "epoch": 4.172904851921865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033615624013110077,
      "loss": 4.9546,
      "mean_token_accuracy": 0.2085220992565155,
      "num_tokens": 91645094.0,
      "step": 49670
    },
    {
      "entropy": 5.447911167144776,
      "epoch": 4.17332493173703,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033612742046190724,
      "loss": 4.9367,
      "mean_token_accuracy": 0.21517597138881683,
      "num_tokens": 91653491.0,
      "step": 49675
    },
    {
      "entropy": 5.489999294281006,
      "epoch": 4.173745011552195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003360985997098731,
      "loss": 4.9578,
      "mean_token_accuracy": 0.21367536187171937,
      "num_tokens": 91662582.0,
      "step": 49680
    },
    {
      "entropy": 5.527881383895874,
      "epoch": 4.17416509136736,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003360697778755086,
      "loss": 4.9746,
      "mean_token_accuracy": 0.2084466502070427,
      "num_tokens": 91672057.0,
      "step": 49685
    },
    {
      "entropy": 5.56822829246521,
      "epoch": 4.174585171182525,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003360409549593244,
      "loss": 5.0423,
      "mean_token_accuracy": 0.20536321997642518,
      "num_tokens": 91682177.0,
      "step": 49690
    },
    {
      "entropy": 5.525495719909668,
      "epoch": 4.17500525099769,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003360121309618312,
      "loss": 4.9913,
      "mean_token_accuracy": 0.20799137353897096,
      "num_tokens": 91691560.0,
      "step": 49695
    },
    {
      "entropy": 5.48063473701477,
      "epoch": 4.175425330812854,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003359833058835395,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20916716307401656,
      "num_tokens": 91701586.0,
      "step": 49700
    },
    {
      "entropy": 5.478703784942627,
      "epoch": 4.175845410628019,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033595447972496,
      "loss": 4.8789,
      "mean_token_accuracy": 0.21381037980318068,
      "num_tokens": 91711643.0,
      "step": 49705
    },
    {
      "entropy": 5.49973316192627,
      "epoch": 4.176265490443185,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003359256524866032,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20658591389656067,
      "num_tokens": 91721741.0,
      "step": 49710
    },
    {
      "entropy": 5.476660537719726,
      "epoch": 4.176685570258349,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003358968241689799,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21045770198106767,
      "num_tokens": 91731278.0,
      "step": 49715
    },
    {
      "entropy": 5.460616731643677,
      "epoch": 4.177105650073514,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033586799477260067,
      "loss": 4.8791,
      "mean_token_accuracy": 0.21344579011201859,
      "num_tokens": 91740803.0,
      "step": 49720
    },
    {
      "entropy": 5.44691309928894,
      "epoch": 4.1775257298886785,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033583916429797626,
      "loss": 4.8965,
      "mean_token_accuracy": 0.21465131342411042,
      "num_tokens": 91749832.0,
      "step": 49725
    },
    {
      "entropy": 5.422392320632935,
      "epoch": 4.177945809703844,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003358103327456173,
      "loss": 4.989,
      "mean_token_accuracy": 0.20218094438314438,
      "num_tokens": 91760560.0,
      "step": 49730
    },
    {
      "entropy": 5.480791330337524,
      "epoch": 4.178365889519009,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003357815001160348,
      "loss": 4.874,
      "mean_token_accuracy": 0.21676035225391388,
      "num_tokens": 91770598.0,
      "step": 49735
    },
    {
      "entropy": 5.493861103057862,
      "epoch": 4.178785969334173,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003357526664097392,
      "loss": 4.8922,
      "mean_token_accuracy": 0.21944651305675505,
      "num_tokens": 91779236.0,
      "step": 49740
    },
    {
      "entropy": 5.462776041030883,
      "epoch": 4.179206049149339,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003357238316272414,
      "loss": 4.9044,
      "mean_token_accuracy": 0.21384128481149672,
      "num_tokens": 91788990.0,
      "step": 49745
    },
    {
      "entropy": 5.404764413833618,
      "epoch": 4.179626128964503,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033569499576905216,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2221020504832268,
      "num_tokens": 91798568.0,
      "step": 49750
    },
    {
      "entropy": 5.515078115463257,
      "epoch": 4.180046208779668,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003356661588356824,
      "loss": 5.0344,
      "mean_token_accuracy": 0.20907863974571228,
      "num_tokens": 91807789.0,
      "step": 49755
    },
    {
      "entropy": 5.485972356796265,
      "epoch": 4.180466288594833,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003356373208276428,
      "loss": 4.9515,
      "mean_token_accuracy": 0.21093525290489196,
      "num_tokens": 91817349.0,
      "step": 49760
    },
    {
      "entropy": 5.540488004684448,
      "epoch": 4.180886368409998,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033560848174544436,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20880354791879654,
      "num_tokens": 91826710.0,
      "step": 49765
    },
    {
      "entropy": 5.472825002670288,
      "epoch": 4.181306448225163,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003355796415895979,
      "loss": 4.9145,
      "mean_token_accuracy": 0.2050171509385109,
      "num_tokens": 91837304.0,
      "step": 49770
    },
    {
      "entropy": 5.464612245559692,
      "epoch": 4.181726528040327,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003355508003606142,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22109843492507936,
      "num_tokens": 91845597.0,
      "step": 49775
    },
    {
      "entropy": 5.553024244308472,
      "epoch": 4.1821466078554925,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003355219580590043,
      "loss": 5.155,
      "mean_token_accuracy": 0.19952321499586106,
      "num_tokens": 91855853.0,
      "step": 49780
    },
    {
      "entropy": 5.412912464141845,
      "epoch": 4.182566687670658,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000335493114685279,
      "loss": 4.8489,
      "mean_token_accuracy": 0.21512220650911332,
      "num_tokens": 91865008.0,
      "step": 49785
    },
    {
      "entropy": 5.470025968551636,
      "epoch": 4.182986767485822,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003354642702399494,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21468976736068726,
      "num_tokens": 91873975.0,
      "step": 49790
    },
    {
      "entropy": 5.376262760162353,
      "epoch": 4.183406847300987,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033543542472352634,
      "loss": 4.8002,
      "mean_token_accuracy": 0.2177663505077362,
      "num_tokens": 91883291.0,
      "step": 49795
    },
    {
      "entropy": 5.519950008392334,
      "epoch": 4.183826927116152,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003354065781365209,
      "loss": 4.9924,
      "mean_token_accuracy": 0.2104595795273781,
      "num_tokens": 91892961.0,
      "step": 49800
    },
    {
      "entropy": 5.390466976165771,
      "epoch": 4.184247006931317,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033537773047944395,
      "loss": 4.793,
      "mean_token_accuracy": 0.22909185588359832,
      "num_tokens": 91902117.0,
      "step": 49805
    },
    {
      "entropy": 5.469184112548828,
      "epoch": 4.184667086746482,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033534888175280664,
      "loss": 4.9803,
      "mean_token_accuracy": 0.2112918019294739,
      "num_tokens": 91912211.0,
      "step": 49810
    },
    {
      "entropy": 5.4005053520202635,
      "epoch": 4.1850871665616465,
      "grad_norm": 1.125,
      "learning_rate": 0.00033532003195711996,
      "loss": 4.829,
      "mean_token_accuracy": 0.22399494647979737,
      "num_tokens": 91919912.0,
      "step": 49815
    },
    {
      "entropy": 5.495564746856689,
      "epoch": 4.185507246376812,
      "grad_norm": 1.0625,
      "learning_rate": 0.000335291181092895,
      "loss": 4.9725,
      "mean_token_accuracy": 0.20648096948862077,
      "num_tokens": 91929587.0,
      "step": 49820
    },
    {
      "entropy": 5.406581926345825,
      "epoch": 4.185927326191976,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003352623291606429,
      "loss": 4.908,
      "mean_token_accuracy": 0.2155707448720932,
      "num_tokens": 91938557.0,
      "step": 49825
    },
    {
      "entropy": 5.39763765335083,
      "epoch": 4.186347406007141,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003352334761608745,
      "loss": 4.8344,
      "mean_token_accuracy": 0.22457422018051149,
      "num_tokens": 91948016.0,
      "step": 49830
    },
    {
      "entropy": 5.3528049945831295,
      "epoch": 4.1867674858223065,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003352046220941012,
      "loss": 4.7775,
      "mean_token_accuracy": 0.2191427007317543,
      "num_tokens": 91956592.0,
      "step": 49835
    },
    {
      "entropy": 5.441923427581787,
      "epoch": 4.187187565637471,
      "grad_norm": 1.125,
      "learning_rate": 0.000335175766960834,
      "loss": 4.913,
      "mean_token_accuracy": 0.21116355806589127,
      "num_tokens": 91966376.0,
      "step": 49840
    },
    {
      "entropy": 5.546475648880005,
      "epoch": 4.187607645452636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033514691076158397,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20298859775066375,
      "num_tokens": 91974933.0,
      "step": 49845
    },
    {
      "entropy": 5.499896240234375,
      "epoch": 4.188027725267801,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003351180534968625,
      "loss": 4.9743,
      "mean_token_accuracy": 0.2172553598880768,
      "num_tokens": 91984217.0,
      "step": 49850
    },
    {
      "entropy": 5.436986637115479,
      "epoch": 4.188447805082966,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003350891951671806,
      "loss": 4.8864,
      "mean_token_accuracy": 0.21641076803207399,
      "num_tokens": 91993590.0,
      "step": 49855
    },
    {
      "entropy": 5.42027702331543,
      "epoch": 4.188867884898131,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003350603357730497,
      "loss": 4.8634,
      "mean_token_accuracy": 0.21774102598428727,
      "num_tokens": 92003251.0,
      "step": 49860
    },
    {
      "entropy": 5.503121709823608,
      "epoch": 4.189287964713295,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033503147531498075,
      "loss": 4.9692,
      "mean_token_accuracy": 0.21144904494285582,
      "num_tokens": 92012047.0,
      "step": 49865
    },
    {
      "entropy": 5.4687819480896,
      "epoch": 4.1897080445284605,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033500261379348505,
      "loss": 4.9676,
      "mean_token_accuracy": 0.20656338930130005,
      "num_tokens": 92021186.0,
      "step": 49870
    },
    {
      "entropy": 5.480315208435059,
      "epoch": 4.190128124343626,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003349737512090741,
      "loss": 4.9697,
      "mean_token_accuracy": 0.21200181245803834,
      "num_tokens": 92030419.0,
      "step": 49875
    },
    {
      "entropy": 5.418538045883179,
      "epoch": 4.19054820415879,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033494488756225893,
      "loss": 4.8431,
      "mean_token_accuracy": 0.21015167683362962,
      "num_tokens": 92039532.0,
      "step": 49880
    },
    {
      "entropy": 5.48201003074646,
      "epoch": 4.190968283973955,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033491602285355096,
      "loss": 4.9595,
      "mean_token_accuracy": 0.20846581608057022,
      "num_tokens": 92049699.0,
      "step": 49885
    },
    {
      "entropy": 5.507870626449585,
      "epoch": 4.19138836378912,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003348871570834615,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21864189207553864,
      "num_tokens": 92059140.0,
      "step": 49890
    },
    {
      "entropy": 5.451932430267334,
      "epoch": 4.191808443604285,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033485829025250194,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21541476994752884,
      "num_tokens": 92068655.0,
      "step": 49895
    },
    {
      "entropy": 5.46851863861084,
      "epoch": 4.19222852341945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033482942236118355,
      "loss": 4.8965,
      "mean_token_accuracy": 0.21351695507764817,
      "num_tokens": 92077545.0,
      "step": 49900
    },
    {
      "entropy": 5.45010666847229,
      "epoch": 4.192648603234614,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033480055341001774,
      "loss": 4.8767,
      "mean_token_accuracy": 0.21564369797706603,
      "num_tokens": 92085600.0,
      "step": 49905
    },
    {
      "entropy": 5.496879386901855,
      "epoch": 4.19306868304978,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033477168339951594,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2114274248480797,
      "num_tokens": 92094822.0,
      "step": 49910
    },
    {
      "entropy": 5.463861465454102,
      "epoch": 4.193488762864944,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003347428123301895,
      "loss": 4.9788,
      "mean_token_accuracy": 0.20882842391729356,
      "num_tokens": 92104677.0,
      "step": 49915
    },
    {
      "entropy": 5.4615387439727785,
      "epoch": 4.193908842680109,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033471394020255,
      "loss": 4.9511,
      "mean_token_accuracy": 0.20945189595222474,
      "num_tokens": 92113221.0,
      "step": 49920
    },
    {
      "entropy": 5.450220966339112,
      "epoch": 4.1943289224952744,
      "grad_norm": 1.125,
      "learning_rate": 0.0003346850670171087,
      "loss": 4.784,
      "mean_token_accuracy": 0.2198694333434105,
      "num_tokens": 92122538.0,
      "step": 49925
    },
    {
      "entropy": 5.5216898918151855,
      "epoch": 4.194749002310439,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003346561927743772,
      "loss": 4.9898,
      "mean_token_accuracy": 0.2165497988462448,
      "num_tokens": 92130538.0,
      "step": 49930
    },
    {
      "entropy": 5.456250953674316,
      "epoch": 4.195169082125604,
      "grad_norm": 1.25,
      "learning_rate": 0.00033462731747486697,
      "loss": 4.9034,
      "mean_token_accuracy": 0.21872368305921555,
      "num_tokens": 92138970.0,
      "step": 49935
    },
    {
      "entropy": 5.498387956619263,
      "epoch": 4.195589161940768,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033459844111908947,
      "loss": 5.0821,
      "mean_token_accuracy": 0.19949263334274292,
      "num_tokens": 92148224.0,
      "step": 49940
    },
    {
      "entropy": 5.3781873226165775,
      "epoch": 4.196009241755934,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033456956370755644,
      "loss": 4.8043,
      "mean_token_accuracy": 0.23162881582975386,
      "num_tokens": 92156861.0,
      "step": 49945
    },
    {
      "entropy": 5.526919412612915,
      "epoch": 4.196429321571099,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003345406852407791,
      "loss": 4.9878,
      "mean_token_accuracy": 0.21109385639429093,
      "num_tokens": 92166911.0,
      "step": 49950
    },
    {
      "entropy": 5.478198909759522,
      "epoch": 4.196849401386263,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033451180571926925,
      "loss": 4.9238,
      "mean_token_accuracy": 0.22050390988588334,
      "num_tokens": 92176454.0,
      "step": 49955
    },
    {
      "entropy": 5.339148664474488,
      "epoch": 4.197269481201428,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033448292514353837,
      "loss": 4.7331,
      "mean_token_accuracy": 0.2292677268385887,
      "num_tokens": 92185477.0,
      "step": 49960
    },
    {
      "entropy": 5.4934654712677,
      "epoch": 4.197689561016593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033445404351409807,
      "loss": 4.9655,
      "mean_token_accuracy": 0.2072744071483612,
      "num_tokens": 92194624.0,
      "step": 49965
    },
    {
      "entropy": 5.4793168067932125,
      "epoch": 4.198109640831758,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003344251608314601,
      "loss": 4.975,
      "mean_token_accuracy": 0.21467892825603485,
      "num_tokens": 92204176.0,
      "step": 49970
    },
    {
      "entropy": 5.4271337509155275,
      "epoch": 4.198529720646923,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003343962770961359,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2209216207265854,
      "num_tokens": 92213356.0,
      "step": 49975
    },
    {
      "entropy": 5.512272596359253,
      "epoch": 4.1989498004620875,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003343673923086373,
      "loss": 4.9796,
      "mean_token_accuracy": 0.20914561450481414,
      "num_tokens": 92222249.0,
      "step": 49980
    },
    {
      "entropy": 5.483822917938232,
      "epoch": 4.199369880277253,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003343385064694759,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21273811161518097,
      "num_tokens": 92231531.0,
      "step": 49985
    },
    {
      "entropy": 5.458845090866089,
      "epoch": 4.199789960092417,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003343096195791634,
      "loss": 4.8574,
      "mean_token_accuracy": 0.21882094740867614,
      "num_tokens": 92240480.0,
      "step": 49990
    },
    {
      "entropy": 5.407445478439331,
      "epoch": 4.200210039907582,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033428073163821155,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2250191643834114,
      "num_tokens": 92249788.0,
      "step": 49995
    },
    {
      "entropy": 5.37434196472168,
      "epoch": 4.200630119722748,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033425184264713207,
      "loss": 4.7684,
      "mean_token_accuracy": 0.2228720963001251,
      "num_tokens": 92258545.0,
      "step": 50000
    },
    {
      "entropy": 5.329419326782227,
      "epoch": 4.201050199537912,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033422295260643667,
      "loss": 4.7891,
      "mean_token_accuracy": 0.2262770414352417,
      "num_tokens": 92267952.0,
      "step": 50005
    },
    {
      "entropy": 5.448588275909424,
      "epoch": 4.201470279353077,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033419406151663717,
      "loss": 4.9022,
      "mean_token_accuracy": 0.2051183119416237,
      "num_tokens": 92276569.0,
      "step": 50010
    },
    {
      "entropy": 5.508931112289429,
      "epoch": 4.201890359168242,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033416516937824535,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20988464057445527,
      "num_tokens": 92285678.0,
      "step": 50015
    },
    {
      "entropy": 5.401798486709595,
      "epoch": 4.202310438983407,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003341362761917731,
      "loss": 4.8359,
      "mean_token_accuracy": 0.21321238279342652,
      "num_tokens": 92295288.0,
      "step": 50020
    },
    {
      "entropy": 5.421548748016358,
      "epoch": 4.202730518798572,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003341073819577321,
      "loss": 4.9102,
      "mean_token_accuracy": 0.21827584356069565,
      "num_tokens": 92304604.0,
      "step": 50025
    },
    {
      "entropy": 5.387268733978272,
      "epoch": 4.203150598613736,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033407848667663424,
      "loss": 4.7678,
      "mean_token_accuracy": 0.22487749308347701,
      "num_tokens": 92313688.0,
      "step": 50030
    },
    {
      "entropy": 5.493468618392944,
      "epoch": 4.2035706784289015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003340495903489915,
      "loss": 4.914,
      "mean_token_accuracy": 0.2208656743168831,
      "num_tokens": 92322580.0,
      "step": 50035
    },
    {
      "entropy": 5.394868803024292,
      "epoch": 4.203990758244067,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003340206929753156,
      "loss": 4.8771,
      "mean_token_accuracy": 0.2148826777935028,
      "num_tokens": 92332669.0,
      "step": 50040
    },
    {
      "entropy": 5.436852502822876,
      "epoch": 4.204410838059231,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003339917945561185,
      "loss": 4.9342,
      "mean_token_accuracy": 0.21326981037855147,
      "num_tokens": 92341763.0,
      "step": 50045
    },
    {
      "entropy": 5.4621302604675295,
      "epoch": 4.204830917874396,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033396289509191214,
      "loss": 4.8632,
      "mean_token_accuracy": 0.2165800228714943,
      "num_tokens": 92351229.0,
      "step": 50050
    },
    {
      "entropy": 5.462176179885864,
      "epoch": 4.205250997689561,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033393399458320845,
      "loss": 4.9337,
      "mean_token_accuracy": 0.20921170711517334,
      "num_tokens": 92360642.0,
      "step": 50055
    },
    {
      "entropy": 5.448937797546387,
      "epoch": 4.205671077504726,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003339050930305194,
      "loss": 4.9533,
      "mean_token_accuracy": 0.21095844805240632,
      "num_tokens": 92369207.0,
      "step": 50060
    },
    {
      "entropy": 5.510989570617676,
      "epoch": 4.206091157319891,
      "grad_norm": 1.09375,
      "learning_rate": 0.000333876190434357,
      "loss": 5.0338,
      "mean_token_accuracy": 0.20605260133743286,
      "num_tokens": 92378753.0,
      "step": 50065
    },
    {
      "entropy": 5.595637035369873,
      "epoch": 4.2065112371350555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003338472867952332,
      "loss": 4.9851,
      "mean_token_accuracy": 0.21294120848178863,
      "num_tokens": 92387485.0,
      "step": 50070
    },
    {
      "entropy": 5.452888345718383,
      "epoch": 4.206931316950221,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033381838211365994,
      "loss": 4.8771,
      "mean_token_accuracy": 0.21293977349996568,
      "num_tokens": 92396349.0,
      "step": 50075
    },
    {
      "entropy": 5.432862329483032,
      "epoch": 4.207351396765385,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033378947639014947,
      "loss": 4.9055,
      "mean_token_accuracy": 0.21301922500133513,
      "num_tokens": 92405302.0,
      "step": 50080
    },
    {
      "entropy": 5.375572538375854,
      "epoch": 4.20777147658055,
      "grad_norm": 1.125,
      "learning_rate": 0.00033376056962521354,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22066505998373032,
      "num_tokens": 92413324.0,
      "step": 50085
    },
    {
      "entropy": 5.433729457855224,
      "epoch": 4.2081915563957155,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003337316618193645,
      "loss": 4.7964,
      "mean_token_accuracy": 0.22458938360214234,
      "num_tokens": 92422746.0,
      "step": 50090
    },
    {
      "entropy": 5.505091047286987,
      "epoch": 4.20861163621088,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033370275297311426,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20979628562927247,
      "num_tokens": 92432981.0,
      "step": 50095
    },
    {
      "entropy": 5.402790212631226,
      "epoch": 4.209031716026045,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033367384308697495,
      "loss": 4.8894,
      "mean_token_accuracy": 0.21657396852970123,
      "num_tokens": 92441478.0,
      "step": 50100
    },
    {
      "entropy": 5.403192090988159,
      "epoch": 4.209451795841209,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003336449321614588,
      "loss": 4.8694,
      "mean_token_accuracy": 0.21045950055122375,
      "num_tokens": 92450312.0,
      "step": 50105
    },
    {
      "entropy": 5.490628576278686,
      "epoch": 4.209871875656375,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003336160201970777,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21072694510221482,
      "num_tokens": 92459632.0,
      "step": 50110
    },
    {
      "entropy": 5.5312305927276615,
      "epoch": 4.21029195547154,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033358710719434416,
      "loss": 4.951,
      "mean_token_accuracy": 0.21453560143709183,
      "num_tokens": 92468462.0,
      "step": 50115
    },
    {
      "entropy": 5.535597085952759,
      "epoch": 4.210712035286704,
      "grad_norm": 1.125,
      "learning_rate": 0.0003335581931537702,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21609316468238832,
      "num_tokens": 92477934.0,
      "step": 50120
    },
    {
      "entropy": 5.446789312362671,
      "epoch": 4.2111321151018695,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003335292780758679,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21131249219179155,
      "num_tokens": 92487226.0,
      "step": 50125
    },
    {
      "entropy": 5.426482677459717,
      "epoch": 4.211552194917034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003335003619611496,
      "loss": 4.9045,
      "mean_token_accuracy": 0.21506811082363128,
      "num_tokens": 92496837.0,
      "step": 50130
    },
    {
      "entropy": 5.421823406219483,
      "epoch": 4.211972274732199,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033347144481012754,
      "loss": 4.7947,
      "mean_token_accuracy": 0.2300365999341011,
      "num_tokens": 92505232.0,
      "step": 50135
    },
    {
      "entropy": 5.469023036956787,
      "epoch": 4.212392354547364,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033344252662331395,
      "loss": 5.0179,
      "mean_token_accuracy": 0.20659370273351668,
      "num_tokens": 92515286.0,
      "step": 50140
    },
    {
      "entropy": 5.417231225967408,
      "epoch": 4.212812434362529,
      "grad_norm": 1.125,
      "learning_rate": 0.0003334136074012211,
      "loss": 4.8526,
      "mean_token_accuracy": 0.2202725887298584,
      "num_tokens": 92523883.0,
      "step": 50145
    },
    {
      "entropy": 5.483329820632934,
      "epoch": 4.213232514177694,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003333846871443612,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21361942440271378,
      "num_tokens": 92533622.0,
      "step": 50150
    },
    {
      "entropy": 5.532946586608887,
      "epoch": 4.213652593992858,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003333557658532467,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20320040434598924,
      "num_tokens": 92542955.0,
      "step": 50155
    },
    {
      "entropy": 5.439728689193726,
      "epoch": 4.214072673808023,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003333268435283898,
      "loss": 4.8564,
      "mean_token_accuracy": 0.20902353078126906,
      "num_tokens": 92551098.0,
      "step": 50160
    },
    {
      "entropy": 5.45811243057251,
      "epoch": 4.214492753623189,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033329792017030294,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21704686433076859,
      "num_tokens": 92560179.0,
      "step": 50165
    },
    {
      "entropy": 5.469087409973144,
      "epoch": 4.214912833438353,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00033326899577949836,
      "loss": 4.9345,
      "mean_token_accuracy": 0.20867166817188262,
      "num_tokens": 92570136.0,
      "step": 50170
    },
    {
      "entropy": 5.511742305755615,
      "epoch": 4.215332913253518,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033324007035648865,
      "loss": 4.985,
      "mean_token_accuracy": 0.21211338192224502,
      "num_tokens": 92579874.0,
      "step": 50175
    },
    {
      "entropy": 5.543657207489014,
      "epoch": 4.2157529930686835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033321114390178603,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21360254138708115,
      "num_tokens": 92588677.0,
      "step": 50180
    },
    {
      "entropy": 5.416254138946533,
      "epoch": 4.216173072883848,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033318221641590295,
      "loss": 4.806,
      "mean_token_accuracy": 0.22413853406906128,
      "num_tokens": 92598142.0,
      "step": 50185
    },
    {
      "entropy": 5.401333808898926,
      "epoch": 4.216593152699013,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033315328789935197,
      "loss": 4.8197,
      "mean_token_accuracy": 0.22048381865024566,
      "num_tokens": 92607216.0,
      "step": 50190
    },
    {
      "entropy": 5.428829622268677,
      "epoch": 4.217013232514177,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003331243583526453,
      "loss": 4.914,
      "mean_token_accuracy": 0.2059685617685318,
      "num_tokens": 92616636.0,
      "step": 50195
    },
    {
      "entropy": 5.4539055824279785,
      "epoch": 4.217433312329343,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033309542777629566,
      "loss": 4.9394,
      "mean_token_accuracy": 0.2121203914284706,
      "num_tokens": 92626272.0,
      "step": 50200
    },
    {
      "entropy": 5.474274921417236,
      "epoch": 4.217853392144508,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033306649617081534,
      "loss": 4.9119,
      "mean_token_accuracy": 0.2128757119178772,
      "num_tokens": 92635932.0,
      "step": 50205
    },
    {
      "entropy": 5.46093521118164,
      "epoch": 4.218273471959672,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033303756353671697,
      "loss": 4.8679,
      "mean_token_accuracy": 0.21361829936504365,
      "num_tokens": 92644618.0,
      "step": 50210
    },
    {
      "entropy": 5.428878355026245,
      "epoch": 4.218693551774837,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003330086298745131,
      "loss": 4.9106,
      "mean_token_accuracy": 0.22154903113842012,
      "num_tokens": 92653958.0,
      "step": 50215
    },
    {
      "entropy": 5.410597848892212,
      "epoch": 4.219113631590002,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003329796951847162,
      "loss": 4.9119,
      "mean_token_accuracy": 0.2052294820547104,
      "num_tokens": 92662972.0,
      "step": 50220
    },
    {
      "entropy": 5.426933336257934,
      "epoch": 4.219533711405167,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003329507594678388,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21325749158859253,
      "num_tokens": 92671238.0,
      "step": 50225
    },
    {
      "entropy": 5.5087676525115965,
      "epoch": 4.219953791220332,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033292182272439357,
      "loss": 4.9457,
      "mean_token_accuracy": 0.21846396327018738,
      "num_tokens": 92679664.0,
      "step": 50230
    },
    {
      "entropy": 5.424381875991822,
      "epoch": 4.2203738710354965,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033289288495489305,
      "loss": 4.8297,
      "mean_token_accuracy": 0.22369007617235184,
      "num_tokens": 92689289.0,
      "step": 50235
    },
    {
      "entropy": 5.452687311172485,
      "epoch": 4.220793950850662,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003328639461598499,
      "loss": 4.898,
      "mean_token_accuracy": 0.21660930663347244,
      "num_tokens": 92698021.0,
      "step": 50240
    },
    {
      "entropy": 5.396657514572143,
      "epoch": 4.221214030665826,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033283500633977676,
      "loss": 4.7631,
      "mean_token_accuracy": 0.2233582019805908,
      "num_tokens": 92706590.0,
      "step": 50245
    },
    {
      "entropy": 5.414116430282593,
      "epoch": 4.221634110480991,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033280606549518626,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22005607485771178,
      "num_tokens": 92714565.0,
      "step": 50250
    },
    {
      "entropy": 5.4467394828796385,
      "epoch": 4.222054190296157,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003327771236265911,
      "loss": 4.9884,
      "mean_token_accuracy": 0.21054133921861648,
      "num_tokens": 92725133.0,
      "step": 50255
    },
    {
      "entropy": 5.548566246032715,
      "epoch": 4.222474270111321,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003327481807345039,
      "loss": 4.9938,
      "mean_token_accuracy": 0.20350890308618547,
      "num_tokens": 92735114.0,
      "step": 50260
    },
    {
      "entropy": 5.52703447341919,
      "epoch": 4.222894349926486,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033271923681943736,
      "loss": 4.9,
      "mean_token_accuracy": 0.21640807539224624,
      "num_tokens": 92744020.0,
      "step": 50265
    },
    {
      "entropy": 5.435284280776978,
      "epoch": 4.2233144297416505,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003326902918819044,
      "loss": 4.9051,
      "mean_token_accuracy": 0.21550808995962142,
      "num_tokens": 92753549.0,
      "step": 50270
    },
    {
      "entropy": 5.435986328125,
      "epoch": 4.223734509556816,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003326613459224176,
      "loss": 4.9637,
      "mean_token_accuracy": 0.20675351619720458,
      "num_tokens": 92763102.0,
      "step": 50275
    },
    {
      "entropy": 5.522110557556152,
      "epoch": 4.224154589371981,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003326323989414897,
      "loss": 5.0474,
      "mean_token_accuracy": 0.20717355459928513,
      "num_tokens": 92773026.0,
      "step": 50280
    },
    {
      "entropy": 5.511414813995361,
      "epoch": 4.224574669187145,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003326034509396335,
      "loss": 4.8987,
      "mean_token_accuracy": 0.21552482396364211,
      "num_tokens": 92782630.0,
      "step": 50285
    },
    {
      "entropy": 5.646433544158936,
      "epoch": 4.2249947490023105,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033257450191736187,
      "loss": 5.0919,
      "mean_token_accuracy": 0.19809215962886811,
      "num_tokens": 92792092.0,
      "step": 50290
    },
    {
      "entropy": 5.526135587692261,
      "epoch": 4.225414828817476,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033254555187518754,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20546719580888748,
      "num_tokens": 92801438.0,
      "step": 50295
    },
    {
      "entropy": 5.455972623825073,
      "epoch": 4.22583490863264,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033251660081362345,
      "loss": 4.7996,
      "mean_token_accuracy": 0.22212506979703903,
      "num_tokens": 92809663.0,
      "step": 50300
    },
    {
      "entropy": 5.391632175445556,
      "epoch": 4.226254988447805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033248764873318226,
      "loss": 4.9042,
      "mean_token_accuracy": 0.20518580824136734,
      "num_tokens": 92818484.0,
      "step": 50305
    },
    {
      "entropy": 5.41616039276123,
      "epoch": 4.22667506826297,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033245869563437714,
      "loss": 4.927,
      "mean_token_accuracy": 0.2184198170900345,
      "num_tokens": 92828185.0,
      "step": 50310
    },
    {
      "entropy": 5.521256828308106,
      "epoch": 4.227095148078135,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003324297415177208,
      "loss": 4.952,
      "mean_token_accuracy": 0.20865603387355805,
      "num_tokens": 92836695.0,
      "step": 50315
    },
    {
      "entropy": 5.538404417037964,
      "epoch": 4.2275152278933,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033240078638372604,
      "loss": 4.9034,
      "mean_token_accuracy": 0.21253711283206939,
      "num_tokens": 92846173.0,
      "step": 50320
    },
    {
      "entropy": 5.4661236763000485,
      "epoch": 4.2279353077084645,
      "grad_norm": 1.1875,
      "learning_rate": 0.000332371830232906,
      "loss": 4.9306,
      "mean_token_accuracy": 0.2151191607117653,
      "num_tokens": 92856177.0,
      "step": 50325
    },
    {
      "entropy": 5.358396911621094,
      "epoch": 4.22835538752363,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033234287306577355,
      "loss": 4.8531,
      "mean_token_accuracy": 0.21884407997131347,
      "num_tokens": 92864492.0,
      "step": 50330
    },
    {
      "entropy": 5.50024676322937,
      "epoch": 4.228775467338794,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033231391488284156,
      "loss": 4.9462,
      "mean_token_accuracy": 0.2089312642812729,
      "num_tokens": 92873770.0,
      "step": 50335
    },
    {
      "entropy": 5.538315296173096,
      "epoch": 4.229195547153959,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033228495568462315,
      "loss": 4.9312,
      "mean_token_accuracy": 0.21088510155677795,
      "num_tokens": 92882962.0,
      "step": 50340
    },
    {
      "entropy": 5.456344318389893,
      "epoch": 4.2296156269691245,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033225599547163123,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21411036252975463,
      "num_tokens": 92891523.0,
      "step": 50345
    },
    {
      "entropy": 5.436241245269775,
      "epoch": 4.230035706784289,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033222703424437875,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21256220936775208,
      "num_tokens": 92901679.0,
      "step": 50350
    },
    {
      "entropy": 5.456876277923584,
      "epoch": 4.230455786599454,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003321980720033789,
      "loss": 4.8639,
      "mean_token_accuracy": 0.22523596733808518,
      "num_tokens": 92910266.0,
      "step": 50355
    },
    {
      "entropy": 5.431916046142578,
      "epoch": 4.230875866414618,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033216910874914467,
      "loss": 4.805,
      "mean_token_accuracy": 0.22457967102527618,
      "num_tokens": 92919152.0,
      "step": 50360
    },
    {
      "entropy": 5.452326488494873,
      "epoch": 4.231295946229784,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003321401444821891,
      "loss": 4.8991,
      "mean_token_accuracy": 0.2117028594017029,
      "num_tokens": 92928580.0,
      "step": 50365
    },
    {
      "entropy": 5.391513967514038,
      "epoch": 4.231716026044949,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033211117920302525,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22246543318033218,
      "num_tokens": 92938831.0,
      "step": 50370
    },
    {
      "entropy": 5.484472274780273,
      "epoch": 4.232136105860113,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003320822129121663,
      "loss": 4.9709,
      "mean_token_accuracy": 0.2143175184726715,
      "num_tokens": 92948091.0,
      "step": 50375
    },
    {
      "entropy": 5.483952045440674,
      "epoch": 4.2325561856752785,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033205324561012535,
      "loss": 4.9912,
      "mean_token_accuracy": 0.21219434291124345,
      "num_tokens": 92957577.0,
      "step": 50380
    },
    {
      "entropy": 5.540793800354004,
      "epoch": 4.232976265490443,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003320242772974156,
      "loss": 5.0413,
      "mean_token_accuracy": 0.20936320424079896,
      "num_tokens": 92967421.0,
      "step": 50385
    },
    {
      "entropy": 5.500285816192627,
      "epoch": 4.233396345305608,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033199530797455,
      "loss": 4.9495,
      "mean_token_accuracy": 0.20925973802804948,
      "num_tokens": 92976471.0,
      "step": 50390
    },
    {
      "entropy": 5.497266054153442,
      "epoch": 4.233816425120773,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033196633764204193,
      "loss": 4.9622,
      "mean_token_accuracy": 0.2176534876227379,
      "num_tokens": 92985496.0,
      "step": 50395
    },
    {
      "entropy": 5.591092824935913,
      "epoch": 4.234236504935938,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033193736630040454,
      "loss": 5.0405,
      "mean_token_accuracy": 0.20834615379571914,
      "num_tokens": 92995110.0,
      "step": 50400
    },
    {
      "entropy": 5.456932783126831,
      "epoch": 4.234656584751103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033190839395015105,
      "loss": 4.8837,
      "mean_token_accuracy": 0.21644229292869568,
      "num_tokens": 93004138.0,
      "step": 50405
    },
    {
      "entropy": 5.443277835845947,
      "epoch": 4.235076664566267,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003318794205917946,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2165342539548874,
      "num_tokens": 93013099.0,
      "step": 50410
    },
    {
      "entropy": 5.43425178527832,
      "epoch": 4.235496744381432,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033185044622584855,
      "loss": 4.834,
      "mean_token_accuracy": 0.22116823643445968,
      "num_tokens": 93021521.0,
      "step": 50415
    },
    {
      "entropy": 5.3369334697723385,
      "epoch": 4.235916824196598,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003318214708528261,
      "loss": 4.8241,
      "mean_token_accuracy": 0.2208465650677681,
      "num_tokens": 93030809.0,
      "step": 50420
    },
    {
      "entropy": 5.480668592453003,
      "epoch": 4.236336904011762,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033179249447324054,
      "loss": 4.946,
      "mean_token_accuracy": 0.2147635206580162,
      "num_tokens": 93040022.0,
      "step": 50425
    },
    {
      "entropy": 5.482736730575562,
      "epoch": 4.236756983826927,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003317635170876052,
      "loss": 4.8689,
      "mean_token_accuracy": 0.21443869471549987,
      "num_tokens": 93048864.0,
      "step": 50430
    },
    {
      "entropy": 5.3663629531860355,
      "epoch": 4.237177063642092,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033173453869643336,
      "loss": 4.8435,
      "mean_token_accuracy": 0.21496578752994538,
      "num_tokens": 93057355.0,
      "step": 50435
    },
    {
      "entropy": 5.566607522964477,
      "epoch": 4.237597143457257,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033170555930023846,
      "loss": 5.0331,
      "mean_token_accuracy": 0.20641722530126572,
      "num_tokens": 93067553.0,
      "step": 50440
    },
    {
      "entropy": 5.453421497344971,
      "epoch": 4.238017223272422,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033167657889953366,
      "loss": 4.8611,
      "mean_token_accuracy": 0.21488398015499116,
      "num_tokens": 93076496.0,
      "step": 50445
    },
    {
      "entropy": 5.480789613723755,
      "epoch": 4.238437303087586,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033164759749483255,
      "loss": 4.8404,
      "mean_token_accuracy": 0.21802469342947006,
      "num_tokens": 93086348.0,
      "step": 50450
    },
    {
      "entropy": 5.465948009490967,
      "epoch": 4.238857382902752,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033161861508664833,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2114957943558693,
      "num_tokens": 93095578.0,
      "step": 50455
    },
    {
      "entropy": 5.4034229755401615,
      "epoch": 4.239277462717917,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033158963167549454,
      "loss": 4.832,
      "mean_token_accuracy": 0.21842210441827775,
      "num_tokens": 93104419.0,
      "step": 50460
    },
    {
      "entropy": 5.355076551437378,
      "epoch": 4.239697542533081,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033156064726188456,
      "loss": 4.8049,
      "mean_token_accuracy": 0.21505447775125502,
      "num_tokens": 93113644.0,
      "step": 50465
    },
    {
      "entropy": 5.483653497695923,
      "epoch": 4.240117622348246,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033153166184633184,
      "loss": 5.0049,
      "mean_token_accuracy": 0.2080223947763443,
      "num_tokens": 93122114.0,
      "step": 50470
    },
    {
      "entropy": 5.397109794616699,
      "epoch": 4.240537702163411,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003315026754293499,
      "loss": 4.914,
      "mean_token_accuracy": 0.21164339184761047,
      "num_tokens": 93131842.0,
      "step": 50475
    },
    {
      "entropy": 5.519757223129273,
      "epoch": 4.240957781978576,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033147368801145205,
      "loss": 4.8896,
      "mean_token_accuracy": 0.22093234807252884,
      "num_tokens": 93140370.0,
      "step": 50480
    },
    {
      "entropy": 5.471499109268189,
      "epoch": 4.241377861793741,
      "grad_norm": 1.109375,
      "learning_rate": 0.000331444699593152,
      "loss": 4.9348,
      "mean_token_accuracy": 0.2101935938000679,
      "num_tokens": 93148824.0,
      "step": 50485
    },
    {
      "entropy": 5.360157442092896,
      "epoch": 4.2417979416089056,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003314157101749631,
      "loss": 4.7874,
      "mean_token_accuracy": 0.22322819828987123,
      "num_tokens": 93157373.0,
      "step": 50490
    },
    {
      "entropy": 5.451534461975098,
      "epoch": 4.242218021424071,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033138671975739895,
      "loss": 4.8669,
      "mean_token_accuracy": 0.21692661046981812,
      "num_tokens": 93165380.0,
      "step": 50495
    },
    {
      "entropy": 5.443373537063598,
      "epoch": 4.242638101239235,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003313577283409731,
      "loss": 4.916,
      "mean_token_accuracy": 0.22056781649589538,
      "num_tokens": 93174255.0,
      "step": 50500
    },
    {
      "entropy": 5.384482526779175,
      "epoch": 4.2430581810544,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033132873592619903,
      "loss": 4.8437,
      "mean_token_accuracy": 0.22044906467199327,
      "num_tokens": 93183551.0,
      "step": 50505
    },
    {
      "entropy": 5.459675264358521,
      "epoch": 4.243478260869566,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003312997425135905,
      "loss": 4.9311,
      "mean_token_accuracy": 0.218624147772789,
      "num_tokens": 93192680.0,
      "step": 50510
    },
    {
      "entropy": 5.489514112472534,
      "epoch": 4.24389834068473,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033127074810366104,
      "loss": 5.018,
      "mean_token_accuracy": 0.20854494869709014,
      "num_tokens": 93203775.0,
      "step": 50515
    },
    {
      "entropy": 5.459809064865112,
      "epoch": 4.244318420499895,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033124175269692423,
      "loss": 4.8808,
      "mean_token_accuracy": 0.220803926885128,
      "num_tokens": 93212211.0,
      "step": 50520
    },
    {
      "entropy": 5.452726030349732,
      "epoch": 4.2447385003150595,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033121275629389367,
      "loss": 4.9757,
      "mean_token_accuracy": 0.20654300600290298,
      "num_tokens": 93221695.0,
      "step": 50525
    },
    {
      "entropy": 5.409875059127808,
      "epoch": 4.245158580130225,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003311837588950832,
      "loss": 4.8535,
      "mean_token_accuracy": 0.21266258358955384,
      "num_tokens": 93230373.0,
      "step": 50530
    },
    {
      "entropy": 5.466826057434082,
      "epoch": 4.24557865994539,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003311547605010063,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21321145296096802,
      "num_tokens": 93239454.0,
      "step": 50535
    },
    {
      "entropy": 5.435885000228882,
      "epoch": 4.245998739760554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033112576111217674,
      "loss": 4.8739,
      "mean_token_accuracy": 0.21669241040945053,
      "num_tokens": 93249058.0,
      "step": 50540
    },
    {
      "entropy": 5.423387336730957,
      "epoch": 4.2464188195757195,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003310967607291082,
      "loss": 4.9112,
      "mean_token_accuracy": 0.2093704342842102,
      "num_tokens": 93258341.0,
      "step": 50545
    },
    {
      "entropy": 5.509046840667724,
      "epoch": 4.246838899390884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003310677593523145,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21511084586381912,
      "num_tokens": 93266783.0,
      "step": 50550
    },
    {
      "entropy": 5.402621555328369,
      "epoch": 4.247258979206049,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003310387569823092,
      "loss": 4.8009,
      "mean_token_accuracy": 0.21851725429296492,
      "num_tokens": 93275880.0,
      "step": 50555
    },
    {
      "entropy": 5.368157577514649,
      "epoch": 4.247679059021214,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003310097536196063,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21020583659410477,
      "num_tokens": 93285433.0,
      "step": 50560
    },
    {
      "entropy": 5.540182304382324,
      "epoch": 4.248099138836379,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00033098074926471937,
      "loss": 4.9615,
      "mean_token_accuracy": 0.22528525292873383,
      "num_tokens": 93295701.0,
      "step": 50565
    },
    {
      "entropy": 5.489024257659912,
      "epoch": 4.248519218651544,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033095174391816227,
      "loss": 4.9204,
      "mean_token_accuracy": 0.21607749164104462,
      "num_tokens": 93304297.0,
      "step": 50570
    },
    {
      "entropy": 5.45756983757019,
      "epoch": 4.248939298466708,
      "grad_norm": 1.125,
      "learning_rate": 0.0003309227375804489,
      "loss": 4.992,
      "mean_token_accuracy": 0.2058843642473221,
      "num_tokens": 93313960.0,
      "step": 50575
    },
    {
      "entropy": 5.443807983398438,
      "epoch": 4.2493593782818735,
      "grad_norm": 1.0625,
      "learning_rate": 0.000330893730252093,
      "loss": 4.8702,
      "mean_token_accuracy": 0.2211660251021385,
      "num_tokens": 93324159.0,
      "step": 50580
    },
    {
      "entropy": 5.492820596694946,
      "epoch": 4.249779458097039,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00033086472193360843,
      "loss": 4.9775,
      "mean_token_accuracy": 0.2017222061753273,
      "num_tokens": 93333986.0,
      "step": 50585
    },
    {
      "entropy": 5.41809778213501,
      "epoch": 4.250199537912203,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003308357126255092,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2087764859199524,
      "num_tokens": 93342728.0,
      "step": 50590
    },
    {
      "entropy": 5.431238603591919,
      "epoch": 4.250619617727368,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033080670232830896,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21965670436620713,
      "num_tokens": 93351913.0,
      "step": 50595
    },
    {
      "entropy": 5.474802732467651,
      "epoch": 4.251039697542533,
      "grad_norm": 1.125,
      "learning_rate": 0.0003307776910425217,
      "loss": 4.9327,
      "mean_token_accuracy": 0.2154631197452545,
      "num_tokens": 93361476.0,
      "step": 50600
    },
    {
      "entropy": 5.469869661331177,
      "epoch": 4.251459777357698,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033074867876866146,
      "loss": 4.9686,
      "mean_token_accuracy": 0.21073969453573227,
      "num_tokens": 93370784.0,
      "step": 50605
    },
    {
      "entropy": 5.469280147552491,
      "epoch": 4.251879857172863,
      "grad_norm": 1.140625,
      "learning_rate": 0.000330719665507242,
      "loss": 4.9003,
      "mean_token_accuracy": 0.21518454253673552,
      "num_tokens": 93379455.0,
      "step": 50610
    },
    {
      "entropy": 5.4896454334259035,
      "epoch": 4.252299936988027,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033069065125877745,
      "loss": 4.9416,
      "mean_token_accuracy": 0.20274905264377593,
      "num_tokens": 93388440.0,
      "step": 50615
    },
    {
      "entropy": 5.484405660629273,
      "epoch": 4.252720016803193,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003306616360237816,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21030529886484145,
      "num_tokens": 93398238.0,
      "step": 50620
    },
    {
      "entropy": 5.4986354351043705,
      "epoch": 4.253140096618358,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033063261980276855,
      "loss": 4.9239,
      "mean_token_accuracy": 0.22156186848878862,
      "num_tokens": 93406909.0,
      "step": 50625
    },
    {
      "entropy": 5.488940811157226,
      "epoch": 4.253560176433522,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003306036025962523,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22112748622894288,
      "num_tokens": 93415945.0,
      "step": 50630
    },
    {
      "entropy": 5.473673105239868,
      "epoch": 4.2539802562486875,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033057458440474685,
      "loss": 4.9687,
      "mean_token_accuracy": 0.20353299379348755,
      "num_tokens": 93424802.0,
      "step": 50635
    },
    {
      "entropy": 5.475709772109985,
      "epoch": 4.254400336063852,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003305455652287662,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21776902228593825,
      "num_tokens": 93433301.0,
      "step": 50640
    },
    {
      "entropy": 5.436055660247803,
      "epoch": 4.254820415879017,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033051654506882454,
      "loss": 4.9368,
      "mean_token_accuracy": 0.21210787296295167,
      "num_tokens": 93442878.0,
      "step": 50645
    },
    {
      "entropy": 5.440820264816284,
      "epoch": 4.255240495694182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003304875239254359,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21551933735609055,
      "num_tokens": 93453060.0,
      "step": 50650
    },
    {
      "entropy": 5.426444005966187,
      "epoch": 4.255660575509347,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003304585017991143,
      "loss": 4.8853,
      "mean_token_accuracy": 0.23109302967786788,
      "num_tokens": 93463719.0,
      "step": 50655
    },
    {
      "entropy": 5.5532975673675535,
      "epoch": 4.256080655324512,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003304294786903739,
      "loss": 4.9937,
      "mean_token_accuracy": 0.21347564160823823,
      "num_tokens": 93473784.0,
      "step": 50660
    },
    {
      "entropy": 5.42771282196045,
      "epoch": 4.256500735139676,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003304004545997288,
      "loss": 4.8717,
      "mean_token_accuracy": 0.21859369575977325,
      "num_tokens": 93482433.0,
      "step": 50665
    },
    {
      "entropy": 5.5180277824401855,
      "epoch": 4.256920814954841,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033037142952769324,
      "loss": 4.9564,
      "mean_token_accuracy": 0.21442057490348815,
      "num_tokens": 93491031.0,
      "step": 50670
    },
    {
      "entropy": 5.414604759216308,
      "epoch": 4.257340894770007,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003303424034747812,
      "loss": 4.8943,
      "mean_token_accuracy": 0.21597530394792558,
      "num_tokens": 93499710.0,
      "step": 50675
    },
    {
      "entropy": 5.377393198013306,
      "epoch": 4.257760974585171,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003303133764415071,
      "loss": 4.8251,
      "mean_token_accuracy": 0.22257234752178193,
      "num_tokens": 93509572.0,
      "step": 50680
    },
    {
      "entropy": 5.4186560153961185,
      "epoch": 4.258181054400336,
      "grad_norm": 1.203125,
      "learning_rate": 0.000330284348428385,
      "loss": 4.8942,
      "mean_token_accuracy": 0.2114245757460594,
      "num_tokens": 93518615.0,
      "step": 50685
    },
    {
      "entropy": 5.5014917850494385,
      "epoch": 4.258601134215501,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033025531943592904,
      "loss": 4.9078,
      "mean_token_accuracy": 0.2146948903799057,
      "num_tokens": 93529076.0,
      "step": 50690
    },
    {
      "entropy": 5.502608871459961,
      "epoch": 4.259021214030666,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033022628946465367,
      "loss": 4.9306,
      "mean_token_accuracy": 0.20285771489143373,
      "num_tokens": 93538606.0,
      "step": 50695
    },
    {
      "entropy": 5.428446626663208,
      "epoch": 4.259441293845831,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033019725851507297,
      "loss": 4.9425,
      "mean_token_accuracy": 0.21338364183902742,
      "num_tokens": 93547984.0,
      "step": 50700
    },
    {
      "entropy": 5.4718259334564205,
      "epoch": 4.259861373660995,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003301682265877012,
      "loss": 4.9907,
      "mean_token_accuracy": 0.21119515150785445,
      "num_tokens": 93557720.0,
      "step": 50705
    },
    {
      "entropy": 5.492273283004761,
      "epoch": 4.260281453476161,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003301391936830528,
      "loss": 4.9428,
      "mean_token_accuracy": 0.20979673713445662,
      "num_tokens": 93567596.0,
      "step": 50710
    },
    {
      "entropy": 5.507753181457519,
      "epoch": 4.260701533291325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033011015980164197,
      "loss": 4.8455,
      "mean_token_accuracy": 0.21805894523859023,
      "num_tokens": 93576762.0,
      "step": 50715
    },
    {
      "entropy": 5.472816801071167,
      "epoch": 4.26112161310649,
      "grad_norm": 1.171875,
      "learning_rate": 0.000330081124943983,
      "loss": 4.9039,
      "mean_token_accuracy": 0.2146998390555382,
      "num_tokens": 93585311.0,
      "step": 50720
    },
    {
      "entropy": 5.510363054275513,
      "epoch": 4.261541692921655,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003300520891105902,
      "loss": 4.9748,
      "mean_token_accuracy": 0.2067165866494179,
      "num_tokens": 93594252.0,
      "step": 50725
    },
    {
      "entropy": 5.397850227355957,
      "epoch": 4.26196177273682,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003300230523019781,
      "loss": 4.8566,
      "mean_token_accuracy": 0.21794022619724274,
      "num_tokens": 93604206.0,
      "step": 50730
    },
    {
      "entropy": 5.425153064727783,
      "epoch": 4.262381852551985,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003299940145186609,
      "loss": 4.905,
      "mean_token_accuracy": 0.21701283901929855,
      "num_tokens": 93613436.0,
      "step": 50735
    },
    {
      "entropy": 5.381421613693237,
      "epoch": 4.26280193236715,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032996497576115306,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21460035890340806,
      "num_tokens": 93622942.0,
      "step": 50740
    },
    {
      "entropy": 5.511899948120117,
      "epoch": 4.263222012182315,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032993593602996903,
      "loss": 4.9373,
      "mean_token_accuracy": 0.20797878056764602,
      "num_tokens": 93633363.0,
      "step": 50745
    },
    {
      "entropy": 5.453534460067749,
      "epoch": 4.26364209199748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032990689532562314,
      "loss": 4.8798,
      "mean_token_accuracy": 0.2170957386493683,
      "num_tokens": 93641395.0,
      "step": 50750
    },
    {
      "entropy": 5.482496976852417,
      "epoch": 4.264062171812644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003298778536486298,
      "loss": 4.9462,
      "mean_token_accuracy": 0.21342443972826003,
      "num_tokens": 93650727.0,
      "step": 50755
    },
    {
      "entropy": 5.409433078765869,
      "epoch": 4.264482251627809,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032984881099950365,
      "loss": 4.8346,
      "mean_token_accuracy": 0.22229166179895402,
      "num_tokens": 93659362.0,
      "step": 50760
    },
    {
      "entropy": 5.523710155487061,
      "epoch": 4.264902331442974,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032981976737875903,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21016030311584472,
      "num_tokens": 93669315.0,
      "step": 50765
    },
    {
      "entropy": 5.452968311309815,
      "epoch": 4.265322411258139,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032979072278691045,
      "loss": 4.9871,
      "mean_token_accuracy": 0.2066267177462578,
      "num_tokens": 93678101.0,
      "step": 50770
    },
    {
      "entropy": 5.435059452056885,
      "epoch": 4.265742491073304,
      "grad_norm": 1.125,
      "learning_rate": 0.0003297616772244724,
      "loss": 4.9013,
      "mean_token_accuracy": 0.21186215728521346,
      "num_tokens": 93687597.0,
      "step": 50775
    },
    {
      "entropy": 5.4119298458099365,
      "epoch": 4.2661625708884685,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003297326306919594,
      "loss": 4.8283,
      "mean_token_accuracy": 0.2222735732793808,
      "num_tokens": 93697055.0,
      "step": 50780
    },
    {
      "entropy": 5.485584878921509,
      "epoch": 4.266582650703634,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032970358318988607,
      "loss": 4.8283,
      "mean_token_accuracy": 0.21930935680866243,
      "num_tokens": 93706754.0,
      "step": 50785
    },
    {
      "entropy": 5.396703672409058,
      "epoch": 4.267002730518799,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032967453471876695,
      "loss": 4.8176,
      "mean_token_accuracy": 0.22168561369180678,
      "num_tokens": 93715243.0,
      "step": 50790
    },
    {
      "entropy": 5.471732902526855,
      "epoch": 4.267422810333963,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032964548527911663,
      "loss": 4.9447,
      "mean_token_accuracy": 0.2115967556834221,
      "num_tokens": 93725395.0,
      "step": 50795
    },
    {
      "entropy": 5.433556699752808,
      "epoch": 4.2678428901491285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032961643487144964,
      "loss": 4.8558,
      "mean_token_accuracy": 0.21396683752536774,
      "num_tokens": 93734711.0,
      "step": 50800
    },
    {
      "entropy": 5.48315920829773,
      "epoch": 4.268262969964293,
      "grad_norm": 1.125,
      "learning_rate": 0.00032958738349628063,
      "loss": 4.9527,
      "mean_token_accuracy": 0.207877579331398,
      "num_tokens": 93744496.0,
      "step": 50805
    },
    {
      "entropy": 5.447770977020264,
      "epoch": 4.268683049779458,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003295583311541243,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2136331543326378,
      "num_tokens": 93753610.0,
      "step": 50810
    },
    {
      "entropy": 5.523852920532226,
      "epoch": 4.269103129594623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032952927784549516,
      "loss": 4.9901,
      "mean_token_accuracy": 0.21518663614988326,
      "num_tokens": 93763046.0,
      "step": 50815
    },
    {
      "entropy": 5.443494701385498,
      "epoch": 4.269523209409788,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003295002235709079,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22040461301803588,
      "num_tokens": 93772030.0,
      "step": 50820
    },
    {
      "entropy": 5.429502534866333,
      "epoch": 4.269943289224953,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003294711683308773,
      "loss": 4.9011,
      "mean_token_accuracy": 0.21166953593492507,
      "num_tokens": 93781536.0,
      "step": 50825
    },
    {
      "entropy": 5.512565660476684,
      "epoch": 4.270363369040117,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032944211212591803,
      "loss": 4.9634,
      "mean_token_accuracy": 0.2143591821193695,
      "num_tokens": 93791543.0,
      "step": 50830
    },
    {
      "entropy": 5.516901206970215,
      "epoch": 4.2707834488552825,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003294130549565447,
      "loss": 4.9375,
      "mean_token_accuracy": 0.20926045924425124,
      "num_tokens": 93800932.0,
      "step": 50835
    },
    {
      "entropy": 5.473478698730469,
      "epoch": 4.271203528670448,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003293839968232722,
      "loss": 4.8548,
      "mean_token_accuracy": 0.22004181146621704,
      "num_tokens": 93808878.0,
      "step": 50840
    },
    {
      "entropy": 5.417050409317016,
      "epoch": 4.271623608485612,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032935493772661507,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21987452059984208,
      "num_tokens": 93819021.0,
      "step": 50845
    },
    {
      "entropy": 5.414792823791504,
      "epoch": 4.272043688300777,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032932587766708825,
      "loss": 4.8313,
      "mean_token_accuracy": 0.22402313202619553,
      "num_tokens": 93827729.0,
      "step": 50850
    },
    {
      "entropy": 5.3824310302734375,
      "epoch": 4.272463768115942,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003292968166452066,
      "loss": 4.8227,
      "mean_token_accuracy": 0.22110067456960678,
      "num_tokens": 93837816.0,
      "step": 50855
    },
    {
      "entropy": 5.4276745319366455,
      "epoch": 4.272883847931107,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032926775466148473,
      "loss": 4.9693,
      "mean_token_accuracy": 0.20657092481851577,
      "num_tokens": 93847883.0,
      "step": 50860
    },
    {
      "entropy": 5.456664896011352,
      "epoch": 4.273303927746272,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003292386917164375,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22310988157987593,
      "num_tokens": 93856420.0,
      "step": 50865
    },
    {
      "entropy": 5.442143726348877,
      "epoch": 4.2737240075614364,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003292096278105798,
      "loss": 4.8961,
      "mean_token_accuracy": 0.21424924731254577,
      "num_tokens": 93865204.0,
      "step": 50870
    },
    {
      "entropy": 5.401884698867798,
      "epoch": 4.274144087376602,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032918056294442647,
      "loss": 4.953,
      "mean_token_accuracy": 0.21848064810037612,
      "num_tokens": 93873998.0,
      "step": 50875
    },
    {
      "entropy": 5.455671787261963,
      "epoch": 4.274564167191766,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032915149711849233,
      "loss": 4.8963,
      "mean_token_accuracy": 0.21509873420000075,
      "num_tokens": 93882939.0,
      "step": 50880
    },
    {
      "entropy": 5.504365491867065,
      "epoch": 4.274984247006931,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003291224303332924,
      "loss": 4.9179,
      "mean_token_accuracy": 0.2181064948439598,
      "num_tokens": 93891607.0,
      "step": 50885
    },
    {
      "entropy": 5.457895946502686,
      "epoch": 4.2754043268220965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032909336258934134,
      "loss": 4.927,
      "mean_token_accuracy": 0.20847560167312623,
      "num_tokens": 93900384.0,
      "step": 50890
    },
    {
      "entropy": 5.438625478744507,
      "epoch": 4.275824406637261,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003290642938871543,
      "loss": 4.8947,
      "mean_token_accuracy": 0.21990025341510772,
      "num_tokens": 93909188.0,
      "step": 50895
    },
    {
      "entropy": 5.4732287406921385,
      "epoch": 4.276244486452426,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032903522422724613,
      "loss": 4.9288,
      "mean_token_accuracy": 0.21346823275089263,
      "num_tokens": 93917911.0,
      "step": 50900
    },
    {
      "entropy": 5.4847290992736815,
      "epoch": 4.276664566267591,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003290061536101318,
      "loss": 4.9415,
      "mean_token_accuracy": 0.20884594619274138,
      "num_tokens": 93926699.0,
      "step": 50905
    },
    {
      "entropy": 5.464923906326294,
      "epoch": 4.277084646082756,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003289770820363262,
      "loss": 4.8719,
      "mean_token_accuracy": 0.21946682631969452,
      "num_tokens": 93935540.0,
      "step": 50910
    },
    {
      "entropy": 5.429844713211059,
      "epoch": 4.277504725897921,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003289480095063444,
      "loss": 4.8979,
      "mean_token_accuracy": 0.211868080496788,
      "num_tokens": 93944168.0,
      "step": 50915
    },
    {
      "entropy": 5.44568281173706,
      "epoch": 4.277924805713085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032891893602070147,
      "loss": 4.864,
      "mean_token_accuracy": 0.21783745884895325,
      "num_tokens": 93952697.0,
      "step": 50920
    },
    {
      "entropy": 5.456588840484619,
      "epoch": 4.27834488552825,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003288898615799123,
      "loss": 4.8199,
      "mean_token_accuracy": 0.21121247708797455,
      "num_tokens": 93961864.0,
      "step": 50925
    },
    {
      "entropy": 5.445867919921875,
      "epoch": 4.278764965343416,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000328860786184492,
      "loss": 4.8947,
      "mean_token_accuracy": 0.21463760435581208,
      "num_tokens": 93971971.0,
      "step": 50930
    },
    {
      "entropy": 5.431161022186279,
      "epoch": 4.27918504515858,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003288317098349556,
      "loss": 4.8869,
      "mean_token_accuracy": 0.21514352560043334,
      "num_tokens": 93980589.0,
      "step": 50935
    },
    {
      "entropy": 5.417072439193726,
      "epoch": 4.279605124973745,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032880263253181816,
      "loss": 4.8936,
      "mean_token_accuracy": 0.21206253468990327,
      "num_tokens": 93990067.0,
      "step": 50940
    },
    {
      "entropy": 5.378760910034179,
      "epoch": 4.28002520478891,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003287735542755948,
      "loss": 4.8385,
      "mean_token_accuracy": 0.21448847353458406,
      "num_tokens": 93999452.0,
      "step": 50945
    },
    {
      "entropy": 5.38917555809021,
      "epoch": 4.280445284604075,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032874447506680063,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21398940086364746,
      "num_tokens": 94008181.0,
      "step": 50950
    },
    {
      "entropy": 5.46374659538269,
      "epoch": 4.28086536441924,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003287153949059507,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21281445473432542,
      "num_tokens": 94017018.0,
      "step": 50955
    },
    {
      "entropy": 5.515384101867676,
      "epoch": 4.281285444234404,
      "grad_norm": 1.125,
      "learning_rate": 0.00032868631379356027,
      "loss": 4.9628,
      "mean_token_accuracy": 0.2087954580783844,
      "num_tokens": 94026181.0,
      "step": 50960
    },
    {
      "entropy": 5.4786803245544435,
      "epoch": 4.28170552404957,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032865723173014445,
      "loss": 4.9356,
      "mean_token_accuracy": 0.20958052426576615,
      "num_tokens": 94035902.0,
      "step": 50965
    },
    {
      "entropy": 5.4797101497650145,
      "epoch": 4.282125603864734,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003286281487162184,
      "loss": 4.9101,
      "mean_token_accuracy": 0.22087534219026567,
      "num_tokens": 94045030.0,
      "step": 50970
    },
    {
      "entropy": 5.487666463851928,
      "epoch": 4.282545683679899,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003285990647522972,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21964920610189437,
      "num_tokens": 94054365.0,
      "step": 50975
    },
    {
      "entropy": 5.38326678276062,
      "epoch": 4.282965763495064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032856997983889624,
      "loss": 4.8459,
      "mean_token_accuracy": 0.22563223391771317,
      "num_tokens": 94063997.0,
      "step": 50980
    },
    {
      "entropy": 5.442962312698365,
      "epoch": 4.283385843310229,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032854089397653066,
      "loss": 4.9581,
      "mean_token_accuracy": 0.2152826115489006,
      "num_tokens": 94074276.0,
      "step": 50985
    },
    {
      "entropy": 5.435170650482178,
      "epoch": 4.283805923125394,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003285118071657157,
      "loss": 4.9175,
      "mean_token_accuracy": 0.21597391963005066,
      "num_tokens": 94083909.0,
      "step": 50990
    },
    {
      "entropy": 5.512300872802735,
      "epoch": 4.284226002940558,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032848271940696673,
      "loss": 4.9712,
      "mean_token_accuracy": 0.2097765699028969,
      "num_tokens": 94094082.0,
      "step": 50995
    },
    {
      "entropy": 5.4639795303344725,
      "epoch": 4.284646082755724,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003284536307007988,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21288820207118989,
      "num_tokens": 94102775.0,
      "step": 51000
    },
    {
      "epoch": 4.284646082755724,
      "eval_entropy": 5.198290802941948,
      "eval_loss": 5.037374496459961,
      "eval_mean_token_accuracy": 0.21564297522815973,
      "eval_num_tokens": 94102775.0,
      "eval_runtime": 27.1478,
      "eval_samples_per_second": 1376.393,
      "eval_steps_per_second": 172.058,
      "step": 51000
    },
    {
      "entropy": 5.4951094627380375,
      "epoch": 4.285066162570889,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003284245410477274,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21580720841884612,
      "num_tokens": 94111419.0,
      "step": 51005
    },
    {
      "entropy": 5.480131578445435,
      "epoch": 4.285486242386053,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003283954504482678,
      "loss": 4.916,
      "mean_token_accuracy": 0.21456512361764907,
      "num_tokens": 94119618.0,
      "step": 51010
    },
    {
      "entropy": 5.465358924865723,
      "epoch": 4.285906322201218,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032836635890293517,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21270590424537658,
      "num_tokens": 94128710.0,
      "step": 51015
    },
    {
      "entropy": 5.52176513671875,
      "epoch": 4.286326402016383,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032833726641224507,
      "loss": 4.9619,
      "mean_token_accuracy": 0.20716682970523834,
      "num_tokens": 94138257.0,
      "step": 51020
    },
    {
      "entropy": 5.431762742996216,
      "epoch": 4.286746481831548,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032830817297671273,
      "loss": 4.8498,
      "mean_token_accuracy": 0.2177225261926651,
      "num_tokens": 94147489.0,
      "step": 51025
    },
    {
      "entropy": 5.491260290145874,
      "epoch": 4.287166561646713,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003282790785968536,
      "loss": 4.9459,
      "mean_token_accuracy": 0.21854322850704194,
      "num_tokens": 94157055.0,
      "step": 51030
    },
    {
      "entropy": 5.440571737289429,
      "epoch": 4.2875866414618775,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032824998327318295,
      "loss": 4.9142,
      "mean_token_accuracy": 0.2200050339102745,
      "num_tokens": 94166564.0,
      "step": 51035
    },
    {
      "entropy": 5.43044638633728,
      "epoch": 4.288006721277043,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032822088700621625,
      "loss": 4.9422,
      "mean_token_accuracy": 0.211848621070385,
      "num_tokens": 94176266.0,
      "step": 51040
    },
    {
      "entropy": 5.477567291259765,
      "epoch": 4.288426801092207,
      "grad_norm": 1.078125,
      "learning_rate": 0.000328191789796469,
      "loss": 4.9375,
      "mean_token_accuracy": 0.21050227284431458,
      "num_tokens": 94185817.0,
      "step": 51045
    },
    {
      "entropy": 5.454865598678589,
      "epoch": 4.288846880907372,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032816269164445656,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21401975005865098,
      "num_tokens": 94194874.0,
      "step": 51050
    },
    {
      "entropy": 5.402228593826294,
      "epoch": 4.2892669607225375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003281335925506944,
      "loss": 4.8782,
      "mean_token_accuracy": 0.2206168442964554,
      "num_tokens": 94203981.0,
      "step": 51055
    },
    {
      "entropy": 5.493890857696533,
      "epoch": 4.289687040537702,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032810449251569805,
      "loss": 4.9597,
      "mean_token_accuracy": 0.20584422051906587,
      "num_tokens": 94213431.0,
      "step": 51060
    },
    {
      "entropy": 5.42995400428772,
      "epoch": 4.290107120352867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032807539153998294,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2183520033955574,
      "num_tokens": 94223038.0,
      "step": 51065
    },
    {
      "entropy": 5.506758165359497,
      "epoch": 4.290527200168032,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003280462896240646,
      "loss": 4.9716,
      "mean_token_accuracy": 0.21399559676647187,
      "num_tokens": 94232064.0,
      "step": 51070
    },
    {
      "entropy": 5.403184080123902,
      "epoch": 4.290947279983197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032801718676845856,
      "loss": 4.84,
      "mean_token_accuracy": 0.2200242683291435,
      "num_tokens": 94241396.0,
      "step": 51075
    },
    {
      "entropy": 5.303766679763794,
      "epoch": 4.291367359798362,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003279880829736803,
      "loss": 4.8159,
      "mean_token_accuracy": 0.22248194962739945,
      "num_tokens": 94250195.0,
      "step": 51080
    },
    {
      "entropy": 5.4152263641357425,
      "epoch": 4.291787439613526,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032795897824024553,
      "loss": 4.8876,
      "mean_token_accuracy": 0.20594078302383423,
      "num_tokens": 94259434.0,
      "step": 51085
    },
    {
      "entropy": 5.404571771621704,
      "epoch": 4.2922075194286915,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003279298725686697,
      "loss": 4.8491,
      "mean_token_accuracy": 0.22054219543933867,
      "num_tokens": 94268602.0,
      "step": 51090
    },
    {
      "entropy": 5.47986855506897,
      "epoch": 4.292627599243857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003279007659594684,
      "loss": 4.937,
      "mean_token_accuracy": 0.21788432449102402,
      "num_tokens": 94278037.0,
      "step": 51095
    },
    {
      "entropy": 5.382727861404419,
      "epoch": 4.293047679059021,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003278716584131573,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22415029555559157,
      "num_tokens": 94286822.0,
      "step": 51100
    },
    {
      "entropy": 5.458807325363159,
      "epoch": 4.293467758874186,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032784254993025195,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21362693309783937,
      "num_tokens": 94296837.0,
      "step": 51105
    },
    {
      "entropy": 5.382442903518677,
      "epoch": 4.293887838689351,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003278134405112682,
      "loss": 4.83,
      "mean_token_accuracy": 0.22783004492521286,
      "num_tokens": 94306096.0,
      "step": 51110
    },
    {
      "entropy": 5.433591747283936,
      "epoch": 4.294307918504516,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032778433015672126,
      "loss": 4.8846,
      "mean_token_accuracy": 0.21162329316139222,
      "num_tokens": 94314703.0,
      "step": 51115
    },
    {
      "entropy": 5.390170526504517,
      "epoch": 4.294727998319681,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032775521886712727,
      "loss": 4.8618,
      "mean_token_accuracy": 0.2100443199276924,
      "num_tokens": 94324061.0,
      "step": 51120
    },
    {
      "entropy": 5.5190530776977536,
      "epoch": 4.2951480781348454,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032772610664300175,
      "loss": 4.9597,
      "mean_token_accuracy": 0.21443410366773605,
      "num_tokens": 94333482.0,
      "step": 51125
    },
    {
      "entropy": 5.532102775573731,
      "epoch": 4.295568157950011,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003276969934848604,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21288614422082902,
      "num_tokens": 94343161.0,
      "step": 51130
    },
    {
      "entropy": 5.4792084217071535,
      "epoch": 4.295988237765175,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032766787939321885,
      "loss": 4.9488,
      "mean_token_accuracy": 0.2143415093421936,
      "num_tokens": 94353276.0,
      "step": 51135
    },
    {
      "entropy": 5.458649349212647,
      "epoch": 4.29640831758034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032763876436859305,
      "loss": 4.9977,
      "mean_token_accuracy": 0.20842737704515457,
      "num_tokens": 94361860.0,
      "step": 51140
    },
    {
      "entropy": 5.481810569763184,
      "epoch": 4.2968283973955055,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003276096484114986,
      "loss": 4.9007,
      "mean_token_accuracy": 0.21863940507173538,
      "num_tokens": 94371669.0,
      "step": 51145
    },
    {
      "entropy": 5.526703405380249,
      "epoch": 4.29724847721067,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032758053152245133,
      "loss": 4.9112,
      "mean_token_accuracy": 0.2156873181462288,
      "num_tokens": 94380524.0,
      "step": 51150
    },
    {
      "entropy": 5.473943185806275,
      "epoch": 4.297668557025835,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000327551413701967,
      "loss": 4.974,
      "mean_token_accuracy": 0.20893739610910417,
      "num_tokens": 94389238.0,
      "step": 51155
    },
    {
      "entropy": 5.34521450996399,
      "epoch": 4.298088636840999,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032752229495056144,
      "loss": 4.7887,
      "mean_token_accuracy": 0.22076240926980972,
      "num_tokens": 94398282.0,
      "step": 51160
    },
    {
      "entropy": 5.4173942565917965,
      "epoch": 4.298508716656165,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003274931752687505,
      "loss": 4.8858,
      "mean_token_accuracy": 0.21653555780649186,
      "num_tokens": 94407332.0,
      "step": 51165
    },
    {
      "entropy": 5.4573265552520756,
      "epoch": 4.29892879647133,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003274640546570499,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2132776066660881,
      "num_tokens": 94417530.0,
      "step": 51170
    },
    {
      "entropy": 5.4894084453582765,
      "epoch": 4.299348876286494,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032743493311597564,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20887168496847153,
      "num_tokens": 94426948.0,
      "step": 51175
    },
    {
      "entropy": 5.440796756744385,
      "epoch": 4.299768956101659,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003274058106460435,
      "loss": 4.8849,
      "mean_token_accuracy": 0.22413769215345383,
      "num_tokens": 94436860.0,
      "step": 51180
    },
    {
      "entropy": 5.437367725372314,
      "epoch": 4.300189035916825,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003273766872477695,
      "loss": 4.8241,
      "mean_token_accuracy": 0.21013121008872987,
      "num_tokens": 94445187.0,
      "step": 51185
    },
    {
      "entropy": 5.521466970443726,
      "epoch": 4.300609115731989,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003273475629216694,
      "loss": 4.9599,
      "mean_token_accuracy": 0.20968988984823228,
      "num_tokens": 94455420.0,
      "step": 51190
    },
    {
      "entropy": 5.463090181350708,
      "epoch": 4.301029195547154,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003273184376682592,
      "loss": 4.9582,
      "mean_token_accuracy": 0.2044214829802513,
      "num_tokens": 94465392.0,
      "step": 51195
    },
    {
      "entropy": 5.483523321151734,
      "epoch": 4.301449275362319,
      "grad_norm": 1.125,
      "learning_rate": 0.00032728931148805484,
      "loss": 4.9234,
      "mean_token_accuracy": 0.21858818978071212,
      "num_tokens": 94474906.0,
      "step": 51200
    },
    {
      "entropy": 5.453805685043335,
      "epoch": 4.301869355177484,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003272601843815722,
      "loss": 4.8596,
      "mean_token_accuracy": 0.21450824439525604,
      "num_tokens": 94484330.0,
      "step": 51205
    },
    {
      "entropy": 5.455346632003784,
      "epoch": 4.302289434992648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032723105634932737,
      "loss": 4.9137,
      "mean_token_accuracy": 0.20997219383716584,
      "num_tokens": 94493280.0,
      "step": 51210
    },
    {
      "entropy": 5.426219511032104,
      "epoch": 4.302709514807813,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003272019273918362,
      "loss": 4.9193,
      "mean_token_accuracy": 0.2189900204539299,
      "num_tokens": 94502821.0,
      "step": 51215
    },
    {
      "entropy": 5.398589277267456,
      "epoch": 4.303129594622979,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003271727975096148,
      "loss": 4.8941,
      "mean_token_accuracy": 0.21735986322164536,
      "num_tokens": 94511892.0,
      "step": 51220
    },
    {
      "entropy": 5.598267459869385,
      "epoch": 4.303549674438143,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003271436667031791,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20941093415021897,
      "num_tokens": 94520976.0,
      "step": 51225
    },
    {
      "entropy": 5.463815593719483,
      "epoch": 4.303969754253308,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003271145349730453,
      "loss": 4.8217,
      "mean_token_accuracy": 0.2258965939283371,
      "num_tokens": 94529801.0,
      "step": 51230
    },
    {
      "entropy": 5.461555147171021,
      "epoch": 4.304389834068473,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032708540231972925,
      "loss": 4.9015,
      "mean_token_accuracy": 0.22003995031118392,
      "num_tokens": 94540369.0,
      "step": 51235
    },
    {
      "entropy": 5.395655012130737,
      "epoch": 4.304809913883638,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032705626874374725,
      "loss": 4.83,
      "mean_token_accuracy": 0.22207614332437514,
      "num_tokens": 94548929.0,
      "step": 51240
    },
    {
      "entropy": 5.36048059463501,
      "epoch": 4.305229993698803,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003270271342456151,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2231200307607651,
      "num_tokens": 94557701.0,
      "step": 51245
    },
    {
      "entropy": 5.45658164024353,
      "epoch": 4.305650073513967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032699799882584906,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2207408294081688,
      "num_tokens": 94566733.0,
      "step": 51250
    },
    {
      "entropy": 5.426090860366822,
      "epoch": 4.306070153329133,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003269688624849653,
      "loss": 4.9143,
      "mean_token_accuracy": 0.2110556423664093,
      "num_tokens": 94574876.0,
      "step": 51255
    },
    {
      "entropy": 5.498019504547119,
      "epoch": 4.306490233144298,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032693972522347995,
      "loss": 4.9656,
      "mean_token_accuracy": 0.2147235542535782,
      "num_tokens": 94583759.0,
      "step": 51260
    },
    {
      "entropy": 5.398813962936401,
      "epoch": 4.306910312959462,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003269105870419091,
      "loss": 4.8287,
      "mean_token_accuracy": 0.21556118428707122,
      "num_tokens": 94593141.0,
      "step": 51265
    },
    {
      "entropy": 5.459956884384155,
      "epoch": 4.307330392774627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032688144794076884,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2195511743426323,
      "num_tokens": 94601832.0,
      "step": 51270
    },
    {
      "entropy": 5.421077108383178,
      "epoch": 4.307750472589792,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003268523079205755,
      "loss": 4.934,
      "mean_token_accuracy": 0.2096417099237442,
      "num_tokens": 94610872.0,
      "step": 51275
    },
    {
      "entropy": 5.517065477371216,
      "epoch": 4.308170552404957,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003268231669818453,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21002745777368545,
      "num_tokens": 94620722.0,
      "step": 51280
    },
    {
      "entropy": 5.431104564666748,
      "epoch": 4.308590632220122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003267940251250943,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22990762591362,
      "num_tokens": 94629999.0,
      "step": 51285
    },
    {
      "entropy": 5.4272266864776615,
      "epoch": 4.3090107120352865,
      "grad_norm": 1.125,
      "learning_rate": 0.0003267648823508388,
      "loss": 4.8785,
      "mean_token_accuracy": 0.21562185287475585,
      "num_tokens": 94638797.0,
      "step": 51290
    },
    {
      "entropy": 5.38621153831482,
      "epoch": 4.309430791850452,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003267357386595951,
      "loss": 4.8858,
      "mean_token_accuracy": 0.21475313752889633,
      "num_tokens": 94647591.0,
      "step": 51295
    },
    {
      "entropy": 5.529116535186768,
      "epoch": 4.309850871665616,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003267065940518794,
      "loss": 5.0124,
      "mean_token_accuracy": 0.21299785226583481,
      "num_tokens": 94656811.0,
      "step": 51300
    },
    {
      "entropy": 5.440731811523437,
      "epoch": 4.310270951480781,
      "grad_norm": 1.109375,
      "learning_rate": 0.000326677448528208,
      "loss": 4.9166,
      "mean_token_accuracy": 0.20928872376680374,
      "num_tokens": 94666073.0,
      "step": 51305
    },
    {
      "entropy": 5.395628786087036,
      "epoch": 4.3106910312959466,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003266483020890973,
      "loss": 4.8858,
      "mean_token_accuracy": 0.21730224788188934,
      "num_tokens": 94675602.0,
      "step": 51310
    },
    {
      "entropy": 5.504262018203735,
      "epoch": 4.311111111111111,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003266191547350634,
      "loss": 4.9975,
      "mean_token_accuracy": 0.20344124734401703,
      "num_tokens": 94685134.0,
      "step": 51315
    },
    {
      "entropy": 5.534199905395508,
      "epoch": 4.311531190926276,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032659000646662284,
      "loss": 4.9099,
      "mean_token_accuracy": 0.2070845767855644,
      "num_tokens": 94693625.0,
      "step": 51320
    },
    {
      "entropy": 5.362872362136841,
      "epoch": 4.3119512707414405,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032656085728429183,
      "loss": 4.7993,
      "mean_token_accuracy": 0.2219635307788849,
      "num_tokens": 94703285.0,
      "step": 51325
    },
    {
      "entropy": 5.374925231933593,
      "epoch": 4.312371350556606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003265317071885869,
      "loss": 4.8802,
      "mean_token_accuracy": 0.217352931201458,
      "num_tokens": 94713053.0,
      "step": 51330
    },
    {
      "entropy": 5.421394491195679,
      "epoch": 4.312791430371771,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003265025561800242,
      "loss": 4.903,
      "mean_token_accuracy": 0.2135825514793396,
      "num_tokens": 94722447.0,
      "step": 51335
    },
    {
      "entropy": 5.447473669052124,
      "epoch": 4.313211510186935,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003264734042591202,
      "loss": 4.9178,
      "mean_token_accuracy": 0.20962510704994203,
      "num_tokens": 94732511.0,
      "step": 51340
    },
    {
      "entropy": 5.46692533493042,
      "epoch": 4.3136315900021005,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032644425142639143,
      "loss": 4.8913,
      "mean_token_accuracy": 0.2177490621805191,
      "num_tokens": 94741352.0,
      "step": 51345
    },
    {
      "entropy": 5.45005054473877,
      "epoch": 4.314051669817266,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003264150976823542,
      "loss": 4.961,
      "mean_token_accuracy": 0.21009791195392608,
      "num_tokens": 94750350.0,
      "step": 51350
    },
    {
      "entropy": 5.43387885093689,
      "epoch": 4.31447174963243,
      "grad_norm": 1.125,
      "learning_rate": 0.000326385943027525,
      "loss": 4.8524,
      "mean_token_accuracy": 0.21698188334703444,
      "num_tokens": 94759635.0,
      "step": 51355
    },
    {
      "entropy": 5.3913013458251955,
      "epoch": 4.314891829447595,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032635678746242033,
      "loss": 4.8329,
      "mean_token_accuracy": 0.21967000514268875,
      "num_tokens": 94768674.0,
      "step": 51360
    },
    {
      "entropy": 5.384539842605591,
      "epoch": 4.31531190926276,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003263276309875566,
      "loss": 4.8463,
      "mean_token_accuracy": 0.2224358692765236,
      "num_tokens": 94778827.0,
      "step": 51365
    },
    {
      "entropy": 5.387426519393921,
      "epoch": 4.315731989077925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003262984736034502,
      "loss": 4.8124,
      "mean_token_accuracy": 0.22304930537939072,
      "num_tokens": 94787001.0,
      "step": 51370
    },
    {
      "entropy": 5.438317680358887,
      "epoch": 4.316152068893089,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003262693153106179,
      "loss": 4.9474,
      "mean_token_accuracy": 0.21080405563116072,
      "num_tokens": 94796142.0,
      "step": 51375
    },
    {
      "entropy": 5.5988246440887455,
      "epoch": 4.3165721487082545,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032624015610957605,
      "loss": 5.085,
      "mean_token_accuracy": 0.20316825211048126,
      "num_tokens": 94805080.0,
      "step": 51380
    },
    {
      "entropy": 5.506745672225952,
      "epoch": 4.31699222852342,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003262109960008411,
      "loss": 4.8632,
      "mean_token_accuracy": 0.21387710869312287,
      "num_tokens": 94813836.0,
      "step": 51385
    },
    {
      "entropy": 5.4850606441497805,
      "epoch": 4.317412308338584,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003261818349849298,
      "loss": 4.9845,
      "mean_token_accuracy": 0.20667244642972946,
      "num_tokens": 94823073.0,
      "step": 51390
    },
    {
      "entropy": 5.446786975860595,
      "epoch": 4.317832388153749,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003261526730623587,
      "loss": 4.9282,
      "mean_token_accuracy": 0.2083016201853752,
      "num_tokens": 94832560.0,
      "step": 51395
    },
    {
      "entropy": 5.449961233139038,
      "epoch": 4.3182524679689145,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003261235102336443,
      "loss": 4.8731,
      "mean_token_accuracy": 0.22024741470813752,
      "num_tokens": 94841922.0,
      "step": 51400
    },
    {
      "entropy": 5.479376745223999,
      "epoch": 4.318672547784079,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032609434649930323,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21076739579439163,
      "num_tokens": 94850987.0,
      "step": 51405
    },
    {
      "entropy": 5.488734483718872,
      "epoch": 4.319092627599244,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003260651818598521,
      "loss": 4.9399,
      "mean_token_accuracy": 0.2115508034825325,
      "num_tokens": 94860992.0,
      "step": 51410
    },
    {
      "entropy": 5.42077522277832,
      "epoch": 4.319512707414408,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003260360163158076,
      "loss": 4.8051,
      "mean_token_accuracy": 0.2176056608557701,
      "num_tokens": 94870447.0,
      "step": 51415
    },
    {
      "entropy": 5.455465412139892,
      "epoch": 4.319932787229574,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003260068498676864,
      "loss": 4.9012,
      "mean_token_accuracy": 0.21198573261499404,
      "num_tokens": 94879776.0,
      "step": 51420
    },
    {
      "entropy": 5.451696729660034,
      "epoch": 4.320352867044739,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032597768251600506,
      "loss": 4.9513,
      "mean_token_accuracy": 0.21682335138320924,
      "num_tokens": 94889127.0,
      "step": 51425
    },
    {
      "entropy": 5.499990940093994,
      "epoch": 4.320772946859903,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003259485142612804,
      "loss": 4.9888,
      "mean_token_accuracy": 0.2084311753511429,
      "num_tokens": 94899103.0,
      "step": 51430
    },
    {
      "entropy": 5.37331485748291,
      "epoch": 4.321193026675068,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032591934510402894,
      "loss": 4.7614,
      "mean_token_accuracy": 0.22413491755723952,
      "num_tokens": 94908000.0,
      "step": 51435
    },
    {
      "entropy": 5.4389997005462645,
      "epoch": 4.321613106490233,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032589017504476754,
      "loss": 4.9127,
      "mean_token_accuracy": 0.21694038361310958,
      "num_tokens": 94916863.0,
      "step": 51440
    },
    {
      "entropy": 5.453451633453369,
      "epoch": 4.322033186305398,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032586100408401285,
      "loss": 5.01,
      "mean_token_accuracy": 0.21832212954759597,
      "num_tokens": 94927172.0,
      "step": 51445
    },
    {
      "entropy": 5.530112218856812,
      "epoch": 4.322453266120563,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003258318322222817,
      "loss": 4.97,
      "mean_token_accuracy": 0.2099423587322235,
      "num_tokens": 94936531.0,
      "step": 51450
    },
    {
      "entropy": 5.444063758850097,
      "epoch": 4.322873345935728,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003258026594600908,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22313956171274185,
      "num_tokens": 94945867.0,
      "step": 51455
    },
    {
      "entropy": 5.4127647399902346,
      "epoch": 4.323293425750893,
      "grad_norm": 1.125,
      "learning_rate": 0.00032577348579795694,
      "loss": 4.8815,
      "mean_token_accuracy": 0.21978713124990462,
      "num_tokens": 94954389.0,
      "step": 51460
    },
    {
      "entropy": 5.442976379394532,
      "epoch": 4.323713505566057,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000325744311236397,
      "loss": 4.915,
      "mean_token_accuracy": 0.2198895111680031,
      "num_tokens": 94964133.0,
      "step": 51465
    },
    {
      "entropy": 5.486218118667603,
      "epoch": 4.324133585381222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032571513577592763,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21691394001245498,
      "num_tokens": 94973579.0,
      "step": 51470
    },
    {
      "entropy": 5.56401424407959,
      "epoch": 4.324553665196388,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003256859594170658,
      "loss": 5.002,
      "mean_token_accuracy": 0.20692049115896224,
      "num_tokens": 94983389.0,
      "step": 51475
    },
    {
      "entropy": 5.483445739746093,
      "epoch": 4.324973745011552,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003256567821603283,
      "loss": 4.8998,
      "mean_token_accuracy": 0.2109411209821701,
      "num_tokens": 94992607.0,
      "step": 51480
    },
    {
      "entropy": 5.469012975692749,
      "epoch": 4.325393824826717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032562760400623196,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21630772352218627,
      "num_tokens": 95002034.0,
      "step": 51485
    },
    {
      "entropy": 5.499039268493652,
      "epoch": 4.3258139046418815,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003255984249552937,
      "loss": 4.9674,
      "mean_token_accuracy": 0.20812161713838578,
      "num_tokens": 95010837.0,
      "step": 51490
    },
    {
      "entropy": 5.411073207855225,
      "epoch": 4.326233984457047,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032556924500803036,
      "loss": 4.8222,
      "mean_token_accuracy": 0.2181664690375328,
      "num_tokens": 95018856.0,
      "step": 51495
    },
    {
      "entropy": 5.482008361816407,
      "epoch": 4.326654064272212,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003255400641649589,
      "loss": 4.9824,
      "mean_token_accuracy": 0.20823485553264617,
      "num_tokens": 95028737.0,
      "step": 51500
    },
    {
      "entropy": 5.39846510887146,
      "epoch": 4.327074144087376,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032551088242659625,
      "loss": 4.8359,
      "mean_token_accuracy": 0.22110164910554886,
      "num_tokens": 95037112.0,
      "step": 51505
    },
    {
      "entropy": 5.518762588500977,
      "epoch": 4.327494223902542,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032548169979345937,
      "loss": 5.0841,
      "mean_token_accuracy": 0.20340307503938676,
      "num_tokens": 95047801.0,
      "step": 51510
    },
    {
      "entropy": 5.454804229736328,
      "epoch": 4.327914303717707,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003254525162660651,
      "loss": 4.8924,
      "mean_token_accuracy": 0.21986761689186096,
      "num_tokens": 95057102.0,
      "step": 51515
    },
    {
      "entropy": 5.461679458618164,
      "epoch": 4.328334383532871,
      "grad_norm": 1.46875,
      "learning_rate": 0.00032542333184493047,
      "loss": 4.8782,
      "mean_token_accuracy": 0.2198985442519188,
      "num_tokens": 95065657.0,
      "step": 51520
    },
    {
      "entropy": 5.472079658508301,
      "epoch": 4.328754463348036,
      "grad_norm": 1.125,
      "learning_rate": 0.00032539414653057246,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21608009040355683,
      "num_tokens": 95074862.0,
      "step": 51525
    },
    {
      "entropy": 5.434670925140381,
      "epoch": 4.329174543163201,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032536496032350813,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20647054463624953,
      "num_tokens": 95084112.0,
      "step": 51530
    },
    {
      "entropy": 5.52520341873169,
      "epoch": 4.329594622978366,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032533577322425454,
      "loss": 4.9798,
      "mean_token_accuracy": 0.21413507759571077,
      "num_tokens": 95092868.0,
      "step": 51535
    },
    {
      "entropy": 5.423083829879761,
      "epoch": 4.330014702793531,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003253065852333286,
      "loss": 4.8547,
      "mean_token_accuracy": 0.22252492606639862,
      "num_tokens": 95101702.0,
      "step": 51540
    },
    {
      "entropy": 5.415515661239624,
      "epoch": 4.3304347826086955,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032527739635124735,
      "loss": 4.8464,
      "mean_token_accuracy": 0.21788775771856309,
      "num_tokens": 95110258.0,
      "step": 51545
    },
    {
      "entropy": 5.418034648895263,
      "epoch": 4.330854862423861,
      "grad_norm": 1.140625,
      "learning_rate": 0.000325248206578528,
      "loss": 4.945,
      "mean_token_accuracy": 0.2117275282740593,
      "num_tokens": 95119538.0,
      "step": 51550
    },
    {
      "entropy": 5.4897106170654295,
      "epoch": 4.331274942239025,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032521901591568746,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21345512419939042,
      "num_tokens": 95128597.0,
      "step": 51555
    },
    {
      "entropy": 5.54432954788208,
      "epoch": 4.33169502205419,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032518982436324296,
      "loss": 4.9419,
      "mean_token_accuracy": 0.22581504583358764,
      "num_tokens": 95137729.0,
      "step": 51560
    },
    {
      "entropy": 5.539882659912109,
      "epoch": 4.3321151018693556,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003251606319217116,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21358068883419037,
      "num_tokens": 95147753.0,
      "step": 51565
    },
    {
      "entropy": 5.498455810546875,
      "epoch": 4.33253518168452,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032513143859161045,
      "loss": 4.9483,
      "mean_token_accuracy": 0.2138144075870514,
      "num_tokens": 95156806.0,
      "step": 51570
    },
    {
      "entropy": 5.384587669372559,
      "epoch": 4.332955261499685,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003251022443734566,
      "loss": 4.8647,
      "mean_token_accuracy": 0.22365618348121644,
      "num_tokens": 95165590.0,
      "step": 51575
    },
    {
      "entropy": 5.398341369628906,
      "epoch": 4.3333753413148495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003250730492677674,
      "loss": 4.8507,
      "mean_token_accuracy": 0.2206322431564331,
      "num_tokens": 95174377.0,
      "step": 51580
    },
    {
      "entropy": 5.446297121047974,
      "epoch": 4.333795421130015,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003250438532750599,
      "loss": 4.9578,
      "mean_token_accuracy": 0.209230999648571,
      "num_tokens": 95183351.0,
      "step": 51585
    },
    {
      "entropy": 5.429872655868531,
      "epoch": 4.33421550094518,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003250146563958512,
      "loss": 4.8607,
      "mean_token_accuracy": 0.2158204957842827,
      "num_tokens": 95193450.0,
      "step": 51590
    },
    {
      "entropy": 5.470662784576416,
      "epoch": 4.334635580760344,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003249854586306587,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21367547959089278,
      "num_tokens": 95203150.0,
      "step": 51595
    },
    {
      "entropy": 5.4923159122467045,
      "epoch": 4.3350556605755095,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003249562599799995,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22106506377458573,
      "num_tokens": 95212644.0,
      "step": 51600
    },
    {
      "entropy": 5.536326551437378,
      "epoch": 4.335475740390674,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003249270604443909,
      "loss": 5.0206,
      "mean_token_accuracy": 0.2015446439385414,
      "num_tokens": 95222544.0,
      "step": 51605
    },
    {
      "entropy": 5.472424411773682,
      "epoch": 4.335895820205839,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003248978600243501,
      "loss": 4.9388,
      "mean_token_accuracy": 0.2157145231962204,
      "num_tokens": 95231707.0,
      "step": 51610
    },
    {
      "entropy": 5.449455738067627,
      "epoch": 4.336315900021004,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003248686587203945,
      "loss": 4.8746,
      "mean_token_accuracy": 0.2213899686932564,
      "num_tokens": 95240623.0,
      "step": 51615
    },
    {
      "entropy": 5.413437938690185,
      "epoch": 4.336735979836169,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032483945653304124,
      "loss": 4.8496,
      "mean_token_accuracy": 0.21867106407880782,
      "num_tokens": 95250374.0,
      "step": 51620
    },
    {
      "entropy": 5.4733551979064945,
      "epoch": 4.337156059651334,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032481025346280763,
      "loss": 4.8904,
      "mean_token_accuracy": 0.2178541302680969,
      "num_tokens": 95258657.0,
      "step": 51625
    },
    {
      "entropy": 5.376325464248657,
      "epoch": 4.337576139466498,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003247810495102111,
      "loss": 4.835,
      "mean_token_accuracy": 0.21661980003118514,
      "num_tokens": 95267347.0,
      "step": 51630
    },
    {
      "entropy": 5.473505973815918,
      "epoch": 4.3379962192816635,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032475184467576886,
      "loss": 4.9373,
      "mean_token_accuracy": 0.20792390406131744,
      "num_tokens": 95276539.0,
      "step": 51635
    },
    {
      "entropy": 5.494429445266723,
      "epoch": 4.338416299096829,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003247226389599983,
      "loss": 4.9315,
      "mean_token_accuracy": 0.2231343001127243,
      "num_tokens": 95285821.0,
      "step": 51640
    },
    {
      "entropy": 5.508666086196899,
      "epoch": 4.338836378911993,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003246934323634168,
      "loss": 5.0101,
      "mean_token_accuracy": 0.2070164754986763,
      "num_tokens": 95294821.0,
      "step": 51645
    },
    {
      "entropy": 5.514852619171142,
      "epoch": 4.339256458727158,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003246642248865418,
      "loss": 4.9569,
      "mean_token_accuracy": 0.20591061264276506,
      "num_tokens": 95304823.0,
      "step": 51650
    },
    {
      "entropy": 5.42778058052063,
      "epoch": 4.339676538542323,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003246350165298905,
      "loss": 4.8801,
      "mean_token_accuracy": 0.2199082851409912,
      "num_tokens": 95313900.0,
      "step": 51655
    },
    {
      "entropy": 5.434550380706787,
      "epoch": 4.340096618357488,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003246058072939805,
      "loss": 4.9209,
      "mean_token_accuracy": 0.21499002426862718,
      "num_tokens": 95322711.0,
      "step": 51660
    },
    {
      "entropy": 5.513976049423218,
      "epoch": 4.340516698172653,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032457659717932924,
      "loss": 4.8816,
      "mean_token_accuracy": 0.21173354983329773,
      "num_tokens": 95332474.0,
      "step": 51665
    },
    {
      "entropy": 5.463698482513427,
      "epoch": 4.340936777987817,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000324547386186454,
      "loss": 4.8827,
      "mean_token_accuracy": 0.21348945796489716,
      "num_tokens": 95341541.0,
      "step": 51670
    },
    {
      "entropy": 5.4196594715118405,
      "epoch": 4.341356857802983,
      "grad_norm": 1.125,
      "learning_rate": 0.0003245181743158724,
      "loss": 4.8619,
      "mean_token_accuracy": 0.2192006677389145,
      "num_tokens": 95350115.0,
      "step": 51675
    },
    {
      "entropy": 5.415652132034301,
      "epoch": 4.341776937618148,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032448896156810184,
      "loss": 4.911,
      "mean_token_accuracy": 0.21896328181028366,
      "num_tokens": 95358894.0,
      "step": 51680
    },
    {
      "entropy": 5.384156084060669,
      "epoch": 4.342197017433312,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032445974794365973,
      "loss": 4.8347,
      "mean_token_accuracy": 0.22031533420085908,
      "num_tokens": 95367372.0,
      "step": 51685
    },
    {
      "entropy": 5.4544408321380615,
      "epoch": 4.342617097248477,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003244305334430637,
      "loss": 4.8585,
      "mean_token_accuracy": 0.22241962254047393,
      "num_tokens": 95376875.0,
      "step": 51690
    },
    {
      "entropy": 5.413222360610962,
      "epoch": 4.343037177063642,
      "grad_norm": 1.125,
      "learning_rate": 0.0003244013180668312,
      "loss": 4.9005,
      "mean_token_accuracy": 0.2146510124206543,
      "num_tokens": 95386572.0,
      "step": 51695
    },
    {
      "entropy": 5.3749549865722654,
      "epoch": 4.343457256878807,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003243721018154799,
      "loss": 4.8866,
      "mean_token_accuracy": 0.2159874439239502,
      "num_tokens": 95395618.0,
      "step": 51700
    },
    {
      "entropy": 5.5301408767700195,
      "epoch": 4.343877336693972,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032434288468952713,
      "loss": 4.998,
      "mean_token_accuracy": 0.20793575346469878,
      "num_tokens": 95405278.0,
      "step": 51705
    },
    {
      "entropy": 5.4682676792144775,
      "epoch": 4.344297416509137,
      "grad_norm": 1.125,
      "learning_rate": 0.0003243136666894905,
      "loss": 4.9538,
      "mean_token_accuracy": 0.21456776112318038,
      "num_tokens": 95415069.0,
      "step": 51710
    },
    {
      "entropy": 5.463176250457764,
      "epoch": 4.344717496324302,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032428444781588777,
      "loss": 4.9513,
      "mean_token_accuracy": 0.21746381968259812,
      "num_tokens": 95424048.0,
      "step": 51715
    },
    {
      "entropy": 5.444345235824585,
      "epoch": 4.345137576139466,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003242552280692364,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2229225978255272,
      "num_tokens": 95433353.0,
      "step": 51720
    },
    {
      "entropy": 5.400847291946411,
      "epoch": 4.345557655954631,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032422600745005397,
      "loss": 4.8113,
      "mean_token_accuracy": 0.21527298241853715,
      "num_tokens": 95442214.0,
      "step": 51725
    },
    {
      "entropy": 5.483308267593384,
      "epoch": 4.345977735769797,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003241967859588583,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2114060029387474,
      "num_tokens": 95451120.0,
      "step": 51730
    },
    {
      "entropy": 5.423619604110717,
      "epoch": 4.346397815584961,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032416756359616674,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2086505636572838,
      "num_tokens": 95460986.0,
      "step": 51735
    },
    {
      "entropy": 5.489164257049561,
      "epoch": 4.346817895400126,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003241383403624972,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21303655207157135,
      "num_tokens": 95469163.0,
      "step": 51740
    },
    {
      "entropy": 5.4499743461608885,
      "epoch": 4.3472379752152905,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003241091162583672,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21589256227016448,
      "num_tokens": 95478070.0,
      "step": 51745
    },
    {
      "entropy": 5.393898057937622,
      "epoch": 4.347658055030456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003240798912842946,
      "loss": 4.861,
      "mean_token_accuracy": 0.22610052227973937,
      "num_tokens": 95486893.0,
      "step": 51750
    },
    {
      "entropy": 5.328414249420166,
      "epoch": 4.348078134845621,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003240506654407969,
      "loss": 4.7798,
      "mean_token_accuracy": 0.22140767872333528,
      "num_tokens": 95495862.0,
      "step": 51755
    },
    {
      "entropy": 5.44094705581665,
      "epoch": 4.348498214660785,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032402143872839196,
      "loss": 4.9126,
      "mean_token_accuracy": 0.222537000477314,
      "num_tokens": 95504953.0,
      "step": 51760
    },
    {
      "entropy": 5.506960821151734,
      "epoch": 4.348918294475951,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003239922111475975,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21618363112211228,
      "num_tokens": 95513996.0,
      "step": 51765
    },
    {
      "entropy": 5.484166145324707,
      "epoch": 4.349338374291115,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032396298269893125,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2148881033062935,
      "num_tokens": 95523343.0,
      "step": 51770
    },
    {
      "entropy": 5.402021312713623,
      "epoch": 4.34975845410628,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003239337533829109,
      "loss": 4.8786,
      "mean_token_accuracy": 0.21554384827613832,
      "num_tokens": 95533173.0,
      "step": 51775
    },
    {
      "entropy": 5.39571442604065,
      "epoch": 4.350178533921445,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003239045232000544,
      "loss": 4.9201,
      "mean_token_accuracy": 0.20976729840040206,
      "num_tokens": 95542515.0,
      "step": 51780
    },
    {
      "entropy": 5.457283878326416,
      "epoch": 4.35059861373661,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003238752921508794,
      "loss": 4.9154,
      "mean_token_accuracy": 0.2165229767560959,
      "num_tokens": 95551880.0,
      "step": 51785
    },
    {
      "entropy": 5.554857778549194,
      "epoch": 4.351018693551775,
      "grad_norm": 1.125,
      "learning_rate": 0.0003238460602359037,
      "loss": 4.9692,
      "mean_token_accuracy": 0.20359882414340974,
      "num_tokens": 95561734.0,
      "step": 51790
    },
    {
      "entropy": 5.482613182067871,
      "epoch": 4.35143877336694,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003238168274556453,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22082646936178207,
      "num_tokens": 95571521.0,
      "step": 51795
    },
    {
      "entropy": 5.399557447433471,
      "epoch": 4.3518588531821045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032378759381062184,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21353477090597153,
      "num_tokens": 95580663.0,
      "step": 51800
    },
    {
      "entropy": 5.403800296783447,
      "epoch": 4.35227893299727,
      "grad_norm": 1.125,
      "learning_rate": 0.0003237583593013513,
      "loss": 4.9125,
      "mean_token_accuracy": 0.20815705955028535,
      "num_tokens": 95589251.0,
      "step": 51805
    },
    {
      "entropy": 5.482579040527344,
      "epoch": 4.352699012812434,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003237291239283516,
      "loss": 4.9715,
      "mean_token_accuracy": 0.21059896498918534,
      "num_tokens": 95598537.0,
      "step": 51810
    },
    {
      "entropy": 5.395714187622071,
      "epoch": 4.353119092627599,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003236998876921406,
      "loss": 4.7902,
      "mean_token_accuracy": 0.2217266246676445,
      "num_tokens": 95606883.0,
      "step": 51815
    },
    {
      "entropy": 5.564819955825806,
      "epoch": 4.353539172442764,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003236706505932361,
      "loss": 5.0773,
      "mean_token_accuracy": 0.203981713950634,
      "num_tokens": 95616814.0,
      "step": 51820
    },
    {
      "entropy": 5.45313606262207,
      "epoch": 4.353959252257929,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003236414126321561,
      "loss": 4.8429,
      "mean_token_accuracy": 0.21897368729114533,
      "num_tokens": 95625410.0,
      "step": 51825
    },
    {
      "entropy": 5.525550746917725,
      "epoch": 4.354379332073094,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003236121738094185,
      "loss": 5.0155,
      "mean_token_accuracy": 0.20938204377889633,
      "num_tokens": 95634624.0,
      "step": 51830
    },
    {
      "entropy": 5.489094305038452,
      "epoch": 4.3547994118882585,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032358293412554123,
      "loss": 4.9213,
      "mean_token_accuracy": 0.20997906476259232,
      "num_tokens": 95643859.0,
      "step": 51835
    },
    {
      "entropy": 5.510078430175781,
      "epoch": 4.355219491703424,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032355369358104234,
      "loss": 4.9618,
      "mean_token_accuracy": 0.2034471035003662,
      "num_tokens": 95653350.0,
      "step": 51840
    },
    {
      "entropy": 5.425216293334961,
      "epoch": 4.355639571518589,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032352445217643985,
      "loss": 4.8642,
      "mean_token_accuracy": 0.218943490087986,
      "num_tokens": 95662331.0,
      "step": 51845
    },
    {
      "entropy": 5.518051338195801,
      "epoch": 4.356059651333753,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032349520991225157,
      "loss": 5.0661,
      "mean_token_accuracy": 0.21234195828437805,
      "num_tokens": 95672359.0,
      "step": 51850
    },
    {
      "entropy": 5.476101589202881,
      "epoch": 4.3564797311489185,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032346596678899564,
      "loss": 4.8588,
      "mean_token_accuracy": 0.21512359529733657,
      "num_tokens": 95680887.0,
      "step": 51855
    },
    {
      "entropy": 5.5377525806427,
      "epoch": 4.356899810964083,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032343672280719005,
      "loss": 4.948,
      "mean_token_accuracy": 0.2092536374926567,
      "num_tokens": 95690574.0,
      "step": 51860
    },
    {
      "entropy": 5.472497749328613,
      "epoch": 4.357319890779248,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003234074779673529,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21283544600009918,
      "num_tokens": 95699879.0,
      "step": 51865
    },
    {
      "entropy": 5.45905065536499,
      "epoch": 4.357739970594413,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003233782322700022,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2161281242966652,
      "num_tokens": 95708757.0,
      "step": 51870
    },
    {
      "entropy": 5.45395450592041,
      "epoch": 4.358160050409578,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000323348985715656,
      "loss": 4.9626,
      "mean_token_accuracy": 0.21309740245342254,
      "num_tokens": 95717919.0,
      "step": 51875
    },
    {
      "entropy": 5.48255090713501,
      "epoch": 4.358580130224743,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003233197383048324,
      "loss": 4.8158,
      "mean_token_accuracy": 0.2238663375377655,
      "num_tokens": 95726230.0,
      "step": 51880
    },
    {
      "entropy": 5.36336030960083,
      "epoch": 4.359000210039907,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032329049003804953,
      "loss": 4.8483,
      "mean_token_accuracy": 0.21185460686683655,
      "num_tokens": 95735197.0,
      "step": 51885
    },
    {
      "entropy": 5.401708030700684,
      "epoch": 4.3594202898550725,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003232612409158255,
      "loss": 4.8835,
      "mean_token_accuracy": 0.21643887162208558,
      "num_tokens": 95744233.0,
      "step": 51890
    },
    {
      "entropy": 5.48811936378479,
      "epoch": 4.359840369670238,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003232319909386785,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22572125494480133,
      "num_tokens": 95752802.0,
      "step": 51895
    },
    {
      "entropy": 5.445374774932861,
      "epoch": 4.360260449485402,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003232027401071266,
      "loss": 4.8665,
      "mean_token_accuracy": 0.21483969390392305,
      "num_tokens": 95762521.0,
      "step": 51900
    },
    {
      "entropy": 5.464945077896118,
      "epoch": 4.360680529300567,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003231734884216879,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20874545276165007,
      "num_tokens": 95770746.0,
      "step": 51905
    },
    {
      "entropy": 5.512399625778198,
      "epoch": 4.361100609115732,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003231442358828807,
      "loss": 4.9711,
      "mean_token_accuracy": 0.2082071855664253,
      "num_tokens": 95780059.0,
      "step": 51910
    },
    {
      "entropy": 5.514664793014527,
      "epoch": 4.361520688930897,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003231149824912232,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2140289455652237,
      "num_tokens": 95790004.0,
      "step": 51915
    },
    {
      "entropy": 5.453298664093017,
      "epoch": 4.361940768746062,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003230857282472335,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22207137197256088,
      "num_tokens": 95799067.0,
      "step": 51920
    },
    {
      "entropy": 5.460177326202393,
      "epoch": 4.362360848561226,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003230564731514298,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21602569222450257,
      "num_tokens": 95808332.0,
      "step": 51925
    },
    {
      "entropy": 5.4228434562683105,
      "epoch": 4.362780928376392,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003230272172043305,
      "loss": 4.9065,
      "mean_token_accuracy": 0.20756284892559052,
      "num_tokens": 95817750.0,
      "step": 51930
    },
    {
      "entropy": 5.464637422561646,
      "epoch": 4.363201008191556,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032299796040645385,
      "loss": 4.9021,
      "mean_token_accuracy": 0.21588539183139802,
      "num_tokens": 95826636.0,
      "step": 51935
    },
    {
      "entropy": 5.483871173858643,
      "epoch": 4.363621088006721,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032296870275831804,
      "loss": 4.9004,
      "mean_token_accuracy": 0.20768955945968628,
      "num_tokens": 95836063.0,
      "step": 51940
    },
    {
      "entropy": 5.468503093719482,
      "epoch": 4.3640411678218864,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003229394442604413,
      "loss": 4.9731,
      "mean_token_accuracy": 0.21196958124637605,
      "num_tokens": 95845029.0,
      "step": 51945
    },
    {
      "entropy": 5.523165988922119,
      "epoch": 4.364461247637051,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000322910184913342,
      "loss": 5.0116,
      "mean_token_accuracy": 0.20695343762636184,
      "num_tokens": 95854588.0,
      "step": 51950
    },
    {
      "entropy": 5.480961179733276,
      "epoch": 4.364881327452216,
      "grad_norm": 1.125,
      "learning_rate": 0.0003228809247175385,
      "loss": 4.92,
      "mean_token_accuracy": 0.2198546200990677,
      "num_tokens": 95863327.0,
      "step": 51955
    },
    {
      "entropy": 5.491772556304932,
      "epoch": 4.365301407267381,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032285166367354904,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21488445103168488,
      "num_tokens": 95872808.0,
      "step": 51960
    },
    {
      "entropy": 5.414108800888061,
      "epoch": 4.365721487082546,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032282240178189194,
      "loss": 4.8616,
      "mean_token_accuracy": 0.21540625542402267,
      "num_tokens": 95881097.0,
      "step": 51965
    },
    {
      "entropy": 5.409906816482544,
      "epoch": 4.366141566897711,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003227931390430857,
      "loss": 4.8374,
      "mean_token_accuracy": 0.21900716423988342,
      "num_tokens": 95889874.0,
      "step": 51970
    },
    {
      "entropy": 5.441257476806641,
      "epoch": 4.366561646712875,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032276387545764854,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22171589285135268,
      "num_tokens": 95899348.0,
      "step": 51975
    },
    {
      "entropy": 5.422741413116455,
      "epoch": 4.36698172652804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003227346110260989,
      "loss": 4.8285,
      "mean_token_accuracy": 0.2211180329322815,
      "num_tokens": 95907914.0,
      "step": 51980
    },
    {
      "entropy": 5.485193586349487,
      "epoch": 4.367401806343205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032270534574895524,
      "loss": 5.0508,
      "mean_token_accuracy": 0.20505081564188005,
      "num_tokens": 95917096.0,
      "step": 51985
    },
    {
      "entropy": 5.426369667053223,
      "epoch": 4.36782188615837,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032267607962673607,
      "loss": 4.8395,
      "mean_token_accuracy": 0.21925582587718964,
      "num_tokens": 95926429.0,
      "step": 51990
    },
    {
      "entropy": 5.4606701850891115,
      "epoch": 4.368241965973535,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032264681265995956,
      "loss": 4.8697,
      "mean_token_accuracy": 0.2202382802963257,
      "num_tokens": 95936451.0,
      "step": 51995
    },
    {
      "entropy": 5.498107147216797,
      "epoch": 4.3686620457886995,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003226175448491443,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21141891777515412,
      "num_tokens": 95945828.0,
      "step": 52000
    },
    {
      "entropy": 5.503820848464966,
      "epoch": 4.369082125603865,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003225882761948088,
      "loss": 4.928,
      "mean_token_accuracy": 0.20868726074695587,
      "num_tokens": 95955350.0,
      "step": 52005
    },
    {
      "entropy": 5.516413593292237,
      "epoch": 4.36950220541903,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032255900669747145,
      "loss": 4.8765,
      "mean_token_accuracy": 0.218747179210186,
      "num_tokens": 95963954.0,
      "step": 52010
    },
    {
      "entropy": 5.4267871379852295,
      "epoch": 4.369922285234194,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003225297363576508,
      "loss": 4.807,
      "mean_token_accuracy": 0.21605663895606994,
      "num_tokens": 95973398.0,
      "step": 52015
    },
    {
      "entropy": 5.472176122665405,
      "epoch": 4.37034236504936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032250046517586536,
      "loss": 4.9318,
      "mean_token_accuracy": 0.20848995447158813,
      "num_tokens": 95982647.0,
      "step": 52020
    },
    {
      "entropy": 5.455416297912597,
      "epoch": 4.370762444864524,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003224711931526336,
      "loss": 4.9147,
      "mean_token_accuracy": 0.21991997808218003,
      "num_tokens": 95991601.0,
      "step": 52025
    },
    {
      "entropy": 5.5525092601776125,
      "epoch": 4.371182524679689,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003224419202884742,
      "loss": 5.0604,
      "mean_token_accuracy": 0.21296733915805816,
      "num_tokens": 96002369.0,
      "step": 52030
    },
    {
      "entropy": 5.4748142719268795,
      "epoch": 4.371602604494854,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003224126465839055,
      "loss": 4.9059,
      "mean_token_accuracy": 0.21917108446359634,
      "num_tokens": 96012653.0,
      "step": 52035
    },
    {
      "entropy": 5.437061786651611,
      "epoch": 4.372022684310019,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003223833720394462,
      "loss": 4.8599,
      "mean_token_accuracy": 0.2154367282986641,
      "num_tokens": 96021859.0,
      "step": 52040
    },
    {
      "entropy": 5.433590269088745,
      "epoch": 4.372442764125184,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032235409665561495,
      "loss": 4.8764,
      "mean_token_accuracy": 0.21674347072839736,
      "num_tokens": 96031061.0,
      "step": 52045
    },
    {
      "entropy": 5.500615835189819,
      "epoch": 4.372862843940348,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032232482043293017,
      "loss": 5.0086,
      "mean_token_accuracy": 0.21124292463064193,
      "num_tokens": 96039970.0,
      "step": 52050
    },
    {
      "entropy": 5.508480644226074,
      "epoch": 4.3732829237555135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032229554337191057,
      "loss": 4.9559,
      "mean_token_accuracy": 0.20899228900671005,
      "num_tokens": 96049323.0,
      "step": 52055
    },
    {
      "entropy": 5.44127721786499,
      "epoch": 4.373703003570679,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032226626547307476,
      "loss": 4.8686,
      "mean_token_accuracy": 0.21536618173122407,
      "num_tokens": 96058355.0,
      "step": 52060
    },
    {
      "entropy": 5.469861030578613,
      "epoch": 4.374123083385843,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032223698673694145,
      "loss": 4.9569,
      "mean_token_accuracy": 0.21078307628631593,
      "num_tokens": 96066888.0,
      "step": 52065
    },
    {
      "entropy": 5.457766056060791,
      "epoch": 4.374543163201008,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032220770716402924,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22203691601753234,
      "num_tokens": 96075983.0,
      "step": 52070
    },
    {
      "entropy": 5.4403034210205075,
      "epoch": 4.374963243016173,
      "grad_norm": 1.125,
      "learning_rate": 0.00032217842675485677,
      "loss": 4.8208,
      "mean_token_accuracy": 0.2272222712635994,
      "num_tokens": 96084906.0,
      "step": 52075
    },
    {
      "entropy": 5.414588260650635,
      "epoch": 4.375383322831338,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003221491455099428,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21551796644926072,
      "num_tokens": 96094956.0,
      "step": 52080
    },
    {
      "entropy": 5.450351190567017,
      "epoch": 4.375803402646503,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000322119863429806,
      "loss": 4.8783,
      "mean_token_accuracy": 0.21894556730985643,
      "num_tokens": 96105168.0,
      "step": 52085
    },
    {
      "entropy": 5.377452278137207,
      "epoch": 4.3762234824616675,
      "grad_norm": 1.09375,
      "learning_rate": 0.000322090580514965,
      "loss": 4.7684,
      "mean_token_accuracy": 0.2195052832365036,
      "num_tokens": 96114737.0,
      "step": 52090
    },
    {
      "entropy": 5.426179504394531,
      "epoch": 4.376643562276833,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003220612967659387,
      "loss": 4.9175,
      "mean_token_accuracy": 0.20944809317588806,
      "num_tokens": 96124690.0,
      "step": 52095
    },
    {
      "entropy": 5.430619096755981,
      "epoch": 4.377063642091997,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003220320121832457,
      "loss": 4.8334,
      "mean_token_accuracy": 0.22414392828941346,
      "num_tokens": 96133713.0,
      "step": 52100
    },
    {
      "entropy": 5.4073097705841064,
      "epoch": 4.377483721907162,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003220027267674049,
      "loss": 4.863,
      "mean_token_accuracy": 0.21651387512683867,
      "num_tokens": 96143560.0,
      "step": 52105
    },
    {
      "entropy": 5.3421995639801025,
      "epoch": 4.3779038017223275,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032197344051893494,
      "loss": 4.7721,
      "mean_token_accuracy": 0.23602833598852158,
      "num_tokens": 96152344.0,
      "step": 52110
    },
    {
      "entropy": 5.358415269851685,
      "epoch": 4.378323881537492,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003219441534383547,
      "loss": 4.8569,
      "mean_token_accuracy": 0.2153329595923424,
      "num_tokens": 96161413.0,
      "step": 52115
    },
    {
      "entropy": 5.4477437973022464,
      "epoch": 4.378743961352657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003219148655261829,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2128266543149948,
      "num_tokens": 96171067.0,
      "step": 52120
    },
    {
      "entropy": 5.505917882919311,
      "epoch": 4.379164041167822,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003218855767829385,
      "loss": 4.9031,
      "mean_token_accuracy": 0.21445335447788239,
      "num_tokens": 96180974.0,
      "step": 52125
    },
    {
      "entropy": 5.539471292495728,
      "epoch": 4.379584120982987,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032185628720914013,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21120190471410752,
      "num_tokens": 96190353.0,
      "step": 52130
    },
    {
      "entropy": 5.493329048156738,
      "epoch": 4.380004200798152,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032182699680530685,
      "loss": 5.0153,
      "mean_token_accuracy": 0.20717872083187103,
      "num_tokens": 96200070.0,
      "step": 52135
    },
    {
      "entropy": 5.479125928878784,
      "epoch": 4.380424280613316,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032179770557195744,
      "loss": 4.9937,
      "mean_token_accuracy": 0.20355050265789032,
      "num_tokens": 96209028.0,
      "step": 52140
    },
    {
      "entropy": 5.491712522506714,
      "epoch": 4.3808443604284815,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003217684135096107,
      "loss": 4.8977,
      "mean_token_accuracy": 0.212109012901783,
      "num_tokens": 96218890.0,
      "step": 52145
    },
    {
      "entropy": 5.384298944473267,
      "epoch": 4.381264440243647,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032173912061878566,
      "loss": 4.7979,
      "mean_token_accuracy": 0.22198835462331773,
      "num_tokens": 96229152.0,
      "step": 52150
    },
    {
      "entropy": 5.36123251914978,
      "epoch": 4.381684520058811,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032170982690000116,
      "loss": 4.8793,
      "mean_token_accuracy": 0.21615341305732727,
      "num_tokens": 96238816.0,
      "step": 52155
    },
    {
      "entropy": 5.378158473968506,
      "epoch": 4.382104599873976,
      "grad_norm": 1.125,
      "learning_rate": 0.00032168053235377616,
      "loss": 4.8156,
      "mean_token_accuracy": 0.2184726968407631,
      "num_tokens": 96247798.0,
      "step": 52160
    },
    {
      "entropy": 5.444741344451904,
      "epoch": 4.382524679689141,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032165123698062946,
      "loss": 4.9058,
      "mean_token_accuracy": 0.2173535481095314,
      "num_tokens": 96255658.0,
      "step": 52165
    },
    {
      "entropy": 5.479947900772094,
      "epoch": 4.382944759504306,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003216219407810802,
      "loss": 4.8627,
      "mean_token_accuracy": 0.2178246095776558,
      "num_tokens": 96263924.0,
      "step": 52170
    },
    {
      "entropy": 5.419219589233398,
      "epoch": 4.383364839319471,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003215926437556473,
      "loss": 4.8174,
      "mean_token_accuracy": 0.21468324214220047,
      "num_tokens": 96273688.0,
      "step": 52175
    },
    {
      "entropy": 5.439648866653442,
      "epoch": 4.383784919134635,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032156334590484966,
      "loss": 4.9014,
      "mean_token_accuracy": 0.21824454367160798,
      "num_tokens": 96282680.0,
      "step": 52180
    },
    {
      "entropy": 5.476385831832886,
      "epoch": 4.384204998949801,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032153404722920633,
      "loss": 4.9105,
      "mean_token_accuracy": 0.2113177701830864,
      "num_tokens": 96291740.0,
      "step": 52185
    },
    {
      "entropy": 5.523378324508667,
      "epoch": 4.384625078764965,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032150474772923624,
      "loss": 4.9429,
      "mean_token_accuracy": 0.21210426837205887,
      "num_tokens": 96300462.0,
      "step": 52190
    },
    {
      "entropy": 5.521164894104004,
      "epoch": 4.38504515858013,
      "grad_norm": 1.515625,
      "learning_rate": 0.00032147544740545854,
      "loss": 5.0027,
      "mean_token_accuracy": 0.2094910368323326,
      "num_tokens": 96309833.0,
      "step": 52195
    },
    {
      "entropy": 5.4727483749389645,
      "epoch": 4.3854652383952955,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032144614625839225,
      "loss": 4.9134,
      "mean_token_accuracy": 0.21589014381170274,
      "num_tokens": 96318836.0,
      "step": 52200
    },
    {
      "entropy": 5.427656364440918,
      "epoch": 4.38588531821046,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032141684428855637,
      "loss": 4.8556,
      "mean_token_accuracy": 0.21991558223962784,
      "num_tokens": 96328723.0,
      "step": 52205
    },
    {
      "entropy": 5.473079967498779,
      "epoch": 4.386305398025625,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032138754149647,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21583349257707596,
      "num_tokens": 96337933.0,
      "step": 52210
    },
    {
      "entropy": 5.4463001728057865,
      "epoch": 4.386725477840789,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032135823788265227,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22216875851154327,
      "num_tokens": 96347533.0,
      "step": 52215
    },
    {
      "entropy": 5.531159782409668,
      "epoch": 4.387145557655955,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032132893344762216,
      "loss": 4.9542,
      "mean_token_accuracy": 0.21664543449878693,
      "num_tokens": 96358070.0,
      "step": 52220
    },
    {
      "entropy": 5.460342073440552,
      "epoch": 4.38756563747112,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003212996281918989,
      "loss": 4.9205,
      "mean_token_accuracy": 0.21925655603408814,
      "num_tokens": 96367325.0,
      "step": 52225
    },
    {
      "entropy": 5.480633020401001,
      "epoch": 4.387985717286284,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032127032211600153,
      "loss": 5.0105,
      "mean_token_accuracy": 0.20565512776374817,
      "num_tokens": 96376256.0,
      "step": 52230
    },
    {
      "entropy": 5.498672437667847,
      "epoch": 4.388405797101449,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032124101522044925,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21540682315826415,
      "num_tokens": 96385485.0,
      "step": 52235
    },
    {
      "entropy": 5.5253692150115965,
      "epoch": 4.388825876916614,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003212117075057612,
      "loss": 4.8772,
      "mean_token_accuracy": 0.21531635969877244,
      "num_tokens": 96395093.0,
      "step": 52240
    },
    {
      "entropy": 5.458129215240478,
      "epoch": 4.389245956731779,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003211823989724565,
      "loss": 4.9465,
      "mean_token_accuracy": 0.2182396411895752,
      "num_tokens": 96405113.0,
      "step": 52245
    },
    {
      "entropy": 5.44935188293457,
      "epoch": 4.389666036546944,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032115308962105436,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21516552418470383,
      "num_tokens": 96414356.0,
      "step": 52250
    },
    {
      "entropy": 5.460012817382813,
      "epoch": 4.3900861163621085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003211237794520741,
      "loss": 4.859,
      "mean_token_accuracy": 0.21580223739147186,
      "num_tokens": 96424552.0,
      "step": 52255
    },
    {
      "entropy": 5.475487232208252,
      "epoch": 4.390506196177274,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003210944684660348,
      "loss": 4.909,
      "mean_token_accuracy": 0.21978201419115068,
      "num_tokens": 96434625.0,
      "step": 52260
    },
    {
      "entropy": 5.443752193450928,
      "epoch": 4.390926275992438,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003210651566634557,
      "loss": 4.9447,
      "mean_token_accuracy": 0.2077163726091385,
      "num_tokens": 96443630.0,
      "step": 52265
    },
    {
      "entropy": 5.4640673160552975,
      "epoch": 4.391346355807603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003210358440448561,
      "loss": 4.9655,
      "mean_token_accuracy": 0.21499472707509995,
      "num_tokens": 96452812.0,
      "step": 52270
    },
    {
      "entropy": 5.504287576675415,
      "epoch": 4.391766435622769,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032100653061075524,
      "loss": 4.8562,
      "mean_token_accuracy": 0.21397818773984909,
      "num_tokens": 96462617.0,
      "step": 52275
    },
    {
      "entropy": 5.465566253662109,
      "epoch": 4.392186515437933,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003209772163616723,
      "loss": 4.8939,
      "mean_token_accuracy": 0.21288462579250336,
      "num_tokens": 96471362.0,
      "step": 52280
    },
    {
      "entropy": 5.4539165019989015,
      "epoch": 4.392606595253098,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003209479012981267,
      "loss": 4.8963,
      "mean_token_accuracy": 0.21960276514291763,
      "num_tokens": 96480192.0,
      "step": 52285
    },
    {
      "entropy": 5.45131106376648,
      "epoch": 4.393026675068263,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032091858542063775,
      "loss": 4.8973,
      "mean_token_accuracy": 0.21583859473466874,
      "num_tokens": 96488826.0,
      "step": 52290
    },
    {
      "entropy": 5.50287971496582,
      "epoch": 4.393446754883428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003208892687297246,
      "loss": 4.9659,
      "mean_token_accuracy": 0.212163507938385,
      "num_tokens": 96498306.0,
      "step": 52295
    },
    {
      "entropy": 5.485382175445556,
      "epoch": 4.393866834698593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003208599512259068,
      "loss": 4.9779,
      "mean_token_accuracy": 0.20965215265750886,
      "num_tokens": 96507641.0,
      "step": 52300
    },
    {
      "entropy": 5.4937238693237305,
      "epoch": 4.394286914513757,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032083063290970346,
      "loss": 4.9232,
      "mean_token_accuracy": 0.21216667890548707,
      "num_tokens": 96516141.0,
      "step": 52305
    },
    {
      "entropy": 5.448025274276733,
      "epoch": 4.3947069943289225,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003208013137816341,
      "loss": 4.8512,
      "mean_token_accuracy": 0.22109986543655397,
      "num_tokens": 96525291.0,
      "step": 52310
    },
    {
      "entropy": 5.372561025619507,
      "epoch": 4.395127074144088,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000320771993842218,
      "loss": 4.7801,
      "mean_token_accuracy": 0.22215869277715683,
      "num_tokens": 96535014.0,
      "step": 52315
    },
    {
      "entropy": 5.470380020141602,
      "epoch": 4.395547153959252,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003207426730919747,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21297538727521897,
      "num_tokens": 96544926.0,
      "step": 52320
    },
    {
      "entropy": 5.4339134216308596,
      "epoch": 4.395967233774417,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032071335153142347,
      "loss": 4.8832,
      "mean_token_accuracy": 0.21309303790330886,
      "num_tokens": 96553875.0,
      "step": 52325
    },
    {
      "entropy": 5.433984231948853,
      "epoch": 4.396387313589582,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032068402916108377,
      "loss": 4.9329,
      "mean_token_accuracy": 0.20813489854335784,
      "num_tokens": 96563357.0,
      "step": 52330
    },
    {
      "entropy": 5.395068073272705,
      "epoch": 4.396807393404747,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032065470598147496,
      "loss": 4.8148,
      "mean_token_accuracy": 0.21663892716169358,
      "num_tokens": 96572819.0,
      "step": 52335
    },
    {
      "entropy": 5.505472421646118,
      "epoch": 4.397227473219912,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003206253819931166,
      "loss": 4.9697,
      "mean_token_accuracy": 0.20969041734933852,
      "num_tokens": 96582069.0,
      "step": 52340
    },
    {
      "entropy": 5.425909519195557,
      "epoch": 4.3976475530350765,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032059605719652804,
      "loss": 4.8359,
      "mean_token_accuracy": 0.22496277689933777,
      "num_tokens": 96590762.0,
      "step": 52345
    },
    {
      "entropy": 5.530546998977661,
      "epoch": 4.398067632850242,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032056673159222885,
      "loss": 5.0672,
      "mean_token_accuracy": 0.20438509732484816,
      "num_tokens": 96599282.0,
      "step": 52350
    },
    {
      "entropy": 5.472009611129761,
      "epoch": 4.398487712665406,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032053740518073843,
      "loss": 4.8934,
      "mean_token_accuracy": 0.21251263618469238,
      "num_tokens": 96608299.0,
      "step": 52355
    },
    {
      "entropy": 5.493935012817383,
      "epoch": 4.398907792480571,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003205080779625763,
      "loss": 4.9397,
      "mean_token_accuracy": 0.21784404516220093,
      "num_tokens": 96618814.0,
      "step": 52360
    },
    {
      "entropy": 5.477106094360352,
      "epoch": 4.3993278722957365,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003204787499382621,
      "loss": 4.9594,
      "mean_token_accuracy": 0.2138746678829193,
      "num_tokens": 96628256.0,
      "step": 52365
    },
    {
      "entropy": 5.385229015350342,
      "epoch": 4.399747952110901,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032044942110831515,
      "loss": 4.8091,
      "mean_token_accuracy": 0.22291209250688554,
      "num_tokens": 96637558.0,
      "step": 52370
    },
    {
      "entropy": 5.524699354171753,
      "epoch": 4.400168031926066,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032042009147325516,
      "loss": 5.001,
      "mean_token_accuracy": 0.20886730551719665,
      "num_tokens": 96646650.0,
      "step": 52375
    },
    {
      "entropy": 5.5008673667907715,
      "epoch": 4.40058811174123,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032039076103360163,
      "loss": 4.9161,
      "mean_token_accuracy": 0.21209233850240708,
      "num_tokens": 96656116.0,
      "step": 52380
    },
    {
      "entropy": 5.434337902069092,
      "epoch": 4.401008191556396,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003203614297898742,
      "loss": 4.915,
      "mean_token_accuracy": 0.21908442229032515,
      "num_tokens": 96664853.0,
      "step": 52385
    },
    {
      "entropy": 5.477983856201172,
      "epoch": 4.401428271371561,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003203320977425924,
      "loss": 4.9437,
      "mean_token_accuracy": 0.2070666506886482,
      "num_tokens": 96673945.0,
      "step": 52390
    },
    {
      "entropy": 5.537311029434204,
      "epoch": 4.401848351186725,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003203027648922757,
      "loss": 5.0203,
      "mean_token_accuracy": 0.208792744576931,
      "num_tokens": 96683786.0,
      "step": 52395
    },
    {
      "entropy": 5.465467309951782,
      "epoch": 4.4022684310018905,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000320273431239444,
      "loss": 4.8855,
      "mean_token_accuracy": 0.2179328143596649,
      "num_tokens": 96692249.0,
      "step": 52400
    },
    {
      "entropy": 5.51190152168274,
      "epoch": 4.402688510817056,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003202440967846167,
      "loss": 4.9749,
      "mean_token_accuracy": 0.2085494413971901,
      "num_tokens": 96701642.0,
      "step": 52405
    },
    {
      "entropy": 5.42705659866333,
      "epoch": 4.40310859063222,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003202147615283136,
      "loss": 4.9022,
      "mean_token_accuracy": 0.2232041984796524,
      "num_tokens": 96709512.0,
      "step": 52410
    },
    {
      "entropy": 5.427405929565429,
      "epoch": 4.403528670447385,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003201854254710542,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21417910009622573,
      "num_tokens": 96719028.0,
      "step": 52415
    },
    {
      "entropy": 5.517612314224243,
      "epoch": 4.40394875026255,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032015608861335823,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21636366993188857,
      "num_tokens": 96728366.0,
      "step": 52420
    },
    {
      "entropy": 5.403967809677124,
      "epoch": 4.404368830077715,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032012675095574545,
      "loss": 4.85,
      "mean_token_accuracy": 0.22276408076286316,
      "num_tokens": 96737810.0,
      "step": 52425
    },
    {
      "entropy": 5.30035891532898,
      "epoch": 4.404788909892879,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032009741249873545,
      "loss": 4.718,
      "mean_token_accuracy": 0.23417620807886125,
      "num_tokens": 96746574.0,
      "step": 52430
    },
    {
      "entropy": 5.498630857467651,
      "epoch": 4.405208989708044,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003200680732428481,
      "loss": 5.0039,
      "mean_token_accuracy": 0.21099905520677567,
      "num_tokens": 96755683.0,
      "step": 52435
    },
    {
      "entropy": 5.400561285018921,
      "epoch": 4.40562906952321,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000320038733188603,
      "loss": 4.901,
      "mean_token_accuracy": 0.2153201848268509,
      "num_tokens": 96765543.0,
      "step": 52440
    },
    {
      "entropy": 5.449112892150879,
      "epoch": 4.406049149338374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032000939233651994,
      "loss": 4.8616,
      "mean_token_accuracy": 0.21920222491025926,
      "num_tokens": 96774079.0,
      "step": 52445
    },
    {
      "entropy": 5.487109184265137,
      "epoch": 4.406469229153539,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031998005068711866,
      "loss": 4.9569,
      "mean_token_accuracy": 0.2076313689351082,
      "num_tokens": 96782432.0,
      "step": 52450
    },
    {
      "entropy": 5.485706090927124,
      "epoch": 4.4068893089687045,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000319950708240919,
      "loss": 4.882,
      "mean_token_accuracy": 0.2151985913515091,
      "num_tokens": 96792060.0,
      "step": 52455
    },
    {
      "entropy": 5.419200897216797,
      "epoch": 4.407309388783869,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003199213649984406,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2221128150820732,
      "num_tokens": 96801110.0,
      "step": 52460
    },
    {
      "entropy": 5.417022848129273,
      "epoch": 4.407729468599034,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003198920209602034,
      "loss": 4.9071,
      "mean_token_accuracy": 0.21776987314224244,
      "num_tokens": 96810110.0,
      "step": 52465
    },
    {
      "entropy": 5.414833116531372,
      "epoch": 4.408149548414198,
      "grad_norm": 1.125,
      "learning_rate": 0.00031986267612672716,
      "loss": 4.9023,
      "mean_token_accuracy": 0.2198183938860893,
      "num_tokens": 96820216.0,
      "step": 52470
    },
    {
      "entropy": 5.415157175064087,
      "epoch": 4.408569628229364,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031983333049853175,
      "loss": 4.8219,
      "mean_token_accuracy": 0.21779033690690994,
      "num_tokens": 96829308.0,
      "step": 52475
    },
    {
      "entropy": 5.367703294754028,
      "epoch": 4.408989708044529,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031980398407613694,
      "loss": 4.7747,
      "mean_token_accuracy": 0.22167675644159318,
      "num_tokens": 96839128.0,
      "step": 52480
    },
    {
      "entropy": 5.427528429031372,
      "epoch": 4.409409787859693,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031977463686006256,
      "loss": 4.9326,
      "mean_token_accuracy": 0.20488783568143845,
      "num_tokens": 96847666.0,
      "step": 52485
    },
    {
      "entropy": 5.513604784011841,
      "epoch": 4.409829867674858,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003197452888508286,
      "loss": 4.9057,
      "mean_token_accuracy": 0.22756165862083436,
      "num_tokens": 96857043.0,
      "step": 52490
    },
    {
      "entropy": 5.49125108718872,
      "epoch": 4.410249947490023,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003197159400489549,
      "loss": 4.9523,
      "mean_token_accuracy": 0.21076682657003404,
      "num_tokens": 96866281.0,
      "step": 52495
    },
    {
      "entropy": 5.461493921279907,
      "epoch": 4.410670027305188,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003196865904549613,
      "loss": 4.8453,
      "mean_token_accuracy": 0.2242945432662964,
      "num_tokens": 96875457.0,
      "step": 52500
    },
    {
      "entropy": 5.433376598358154,
      "epoch": 4.411090107120353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003196572400693678,
      "loss": 4.9684,
      "mean_token_accuracy": 0.2078787550330162,
      "num_tokens": 96884315.0,
      "step": 52505
    },
    {
      "entropy": 5.491872358322143,
      "epoch": 4.4115101869355176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031962788889269427,
      "loss": 4.9918,
      "mean_token_accuracy": 0.2140519693493843,
      "num_tokens": 96893599.0,
      "step": 52510
    },
    {
      "entropy": 5.506499338150024,
      "epoch": 4.411930266750683,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003195985369254608,
      "loss": 4.8551,
      "mean_token_accuracy": 0.21921105980873107,
      "num_tokens": 96902698.0,
      "step": 52515
    },
    {
      "entropy": 5.408009147644043,
      "epoch": 4.412350346565847,
      "grad_norm": 1.1875,
      "learning_rate": 0.000319569184168187,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22440190017223358,
      "num_tokens": 96911300.0,
      "step": 52520
    },
    {
      "entropy": 5.448168182373047,
      "epoch": 4.412770426381012,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003195398306213932,
      "loss": 4.9707,
      "mean_token_accuracy": 0.2001977562904358,
      "num_tokens": 96920486.0,
      "step": 52525
    },
    {
      "entropy": 5.438336610794067,
      "epoch": 4.413190506196178,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003195104762855992,
      "loss": 4.9191,
      "mean_token_accuracy": 0.21874054819345473,
      "num_tokens": 96930771.0,
      "step": 52530
    },
    {
      "entropy": 5.431320428848267,
      "epoch": 4.413610586011342,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000319481121161325,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22496256828308106,
      "num_tokens": 96939837.0,
      "step": 52535
    },
    {
      "entropy": 5.462357664108277,
      "epoch": 4.414030665826507,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031945176524909066,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2147974282503128,
      "num_tokens": 96948328.0,
      "step": 52540
    },
    {
      "entropy": 5.5086887836456295,
      "epoch": 4.4144507456416715,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031942240854941617,
      "loss": 4.9452,
      "mean_token_accuracy": 0.21699891090393067,
      "num_tokens": 96957665.0,
      "step": 52545
    },
    {
      "entropy": 5.46887788772583,
      "epoch": 4.414870825456837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031939305106282157,
      "loss": 4.875,
      "mean_token_accuracy": 0.2166127786040306,
      "num_tokens": 96966810.0,
      "step": 52550
    },
    {
      "entropy": 5.475343322753906,
      "epoch": 4.415290905272002,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031936369278982694,
      "loss": 4.9494,
      "mean_token_accuracy": 0.21445146203041077,
      "num_tokens": 96976416.0,
      "step": 52555
    },
    {
      "entropy": 5.430989074707031,
      "epoch": 4.415710985087166,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003193343337309524,
      "loss": 4.8429,
      "mean_token_accuracy": 0.22219238430261612,
      "num_tokens": 96985350.0,
      "step": 52560
    },
    {
      "entropy": 5.428889369964599,
      "epoch": 4.4161310649023315,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000319304973886718,
      "loss": 4.8806,
      "mean_token_accuracy": 0.2140945866703987,
      "num_tokens": 96994307.0,
      "step": 52565
    },
    {
      "entropy": 5.437466430664062,
      "epoch": 4.416551144717497,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031927561325764366,
      "loss": 4.8646,
      "mean_token_accuracy": 0.21598258912563323,
      "num_tokens": 97003165.0,
      "step": 52570
    },
    {
      "entropy": 5.441289567947388,
      "epoch": 4.416971224532661,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003192462518442497,
      "loss": 4.8094,
      "mean_token_accuracy": 0.22778561860322952,
      "num_tokens": 97011960.0,
      "step": 52575
    },
    {
      "entropy": 5.423130035400391,
      "epoch": 4.417391304347826,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003192168896470562,
      "loss": 4.9023,
      "mean_token_accuracy": 0.22177961468696594,
      "num_tokens": 97021353.0,
      "step": 52580
    },
    {
      "entropy": 5.4962475299835205,
      "epoch": 4.417811384162991,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003191875266665832,
      "loss": 4.86,
      "mean_token_accuracy": 0.21257856339216233,
      "num_tokens": 97030356.0,
      "step": 52585
    },
    {
      "entropy": 5.442424964904785,
      "epoch": 4.418231463978156,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031915816290335096,
      "loss": 4.9274,
      "mean_token_accuracy": 0.21520068645477294,
      "num_tokens": 97039352.0,
      "step": 52590
    },
    {
      "entropy": 5.416506242752075,
      "epoch": 4.418651543793321,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003191287983578796,
      "loss": 4.8871,
      "mean_token_accuracy": 0.21582849621772765,
      "num_tokens": 97048122.0,
      "step": 52595
    },
    {
      "entropy": 5.423711347579956,
      "epoch": 4.4190716236084855,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003190994330306893,
      "loss": 4.9181,
      "mean_token_accuracy": 0.22115711718797684,
      "num_tokens": 97056853.0,
      "step": 52600
    },
    {
      "entropy": 5.471658229827881,
      "epoch": 4.419491703423651,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003190700669223003,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21379654109477997,
      "num_tokens": 97065713.0,
      "step": 52605
    },
    {
      "entropy": 5.4787928581237795,
      "epoch": 4.419911783238815,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031904070003323264,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2232297584414482,
      "num_tokens": 97074647.0,
      "step": 52610
    },
    {
      "entropy": 5.452570962905884,
      "epoch": 4.42033186305398,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003190113323640067,
      "loss": 4.9585,
      "mean_token_accuracy": 0.21841602623462678,
      "num_tokens": 97084204.0,
      "step": 52615
    },
    {
      "entropy": 5.3085309028625485,
      "epoch": 4.4207519428691455,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003189819639151427,
      "loss": 4.7043,
      "mean_token_accuracy": 0.23018197864294052,
      "num_tokens": 97093868.0,
      "step": 52620
    },
    {
      "entropy": 5.489178133010864,
      "epoch": 4.42117202268431,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031895259468716085,
      "loss": 4.9633,
      "mean_token_accuracy": 0.20867136269807815,
      "num_tokens": 97105124.0,
      "step": 52625
    },
    {
      "entropy": 5.465738344192505,
      "epoch": 4.421592102499475,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031892322468058147,
      "loss": 4.9077,
      "mean_token_accuracy": 0.21601202785968782,
      "num_tokens": 97114378.0,
      "step": 52630
    },
    {
      "entropy": 5.457678508758545,
      "epoch": 4.422012182314639,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031889385389592467,
      "loss": 4.9368,
      "mean_token_accuracy": 0.21531140357255935,
      "num_tokens": 97122999.0,
      "step": 52635
    },
    {
      "entropy": 5.4859458923339846,
      "epoch": 4.422432262129805,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000318864482333711,
      "loss": 4.9624,
      "mean_token_accuracy": 0.2098172962665558,
      "num_tokens": 97132493.0,
      "step": 52640
    },
    {
      "entropy": 5.423301839828492,
      "epoch": 4.42285234194497,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003188351099944605,
      "loss": 4.877,
      "mean_token_accuracy": 0.21761366873979568,
      "num_tokens": 97141051.0,
      "step": 52645
    },
    {
      "entropy": 5.407134389877319,
      "epoch": 4.423272421760134,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003188057368786936,
      "loss": 4.8851,
      "mean_token_accuracy": 0.21508648991584778,
      "num_tokens": 97150282.0,
      "step": 52650
    },
    {
      "entropy": 5.419349145889282,
      "epoch": 4.4236925015752995,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031877636298693066,
      "loss": 4.9019,
      "mean_token_accuracy": 0.217318157851696,
      "num_tokens": 97159967.0,
      "step": 52655
    },
    {
      "entropy": 5.484866619110107,
      "epoch": 4.424112581390464,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000318746988319692,
      "loss": 4.8787,
      "mean_token_accuracy": 0.21672368347644805,
      "num_tokens": 97169316.0,
      "step": 52660
    },
    {
      "entropy": 5.524976634979248,
      "epoch": 4.424532661205629,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031871761287749793,
      "loss": 4.902,
      "mean_token_accuracy": 0.215748693048954,
      "num_tokens": 97178513.0,
      "step": 52665
    },
    {
      "entropy": 5.361628246307373,
      "epoch": 4.424952741020794,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003186882366608688,
      "loss": 4.7861,
      "mean_token_accuracy": 0.2228020876646042,
      "num_tokens": 97187264.0,
      "step": 52670
    },
    {
      "entropy": 5.3517671585083,
      "epoch": 4.425372820835959,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031865885967032514,
      "loss": 4.7589,
      "mean_token_accuracy": 0.22999098151922226,
      "num_tokens": 97196601.0,
      "step": 52675
    },
    {
      "entropy": 5.423547792434692,
      "epoch": 4.425792900651124,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003186294819063873,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22788713425397872,
      "num_tokens": 97205416.0,
      "step": 52680
    },
    {
      "entropy": 5.43200101852417,
      "epoch": 4.426212980466288,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003186001033695756,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2335409179329872,
      "num_tokens": 97214351.0,
      "step": 52685
    },
    {
      "entropy": 5.482290267944336,
      "epoch": 4.426633060281453,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003185707240604105,
      "loss": 4.9686,
      "mean_token_accuracy": 0.21545757800340654,
      "num_tokens": 97222623.0,
      "step": 52690
    },
    {
      "entropy": 5.459488582611084,
      "epoch": 4.427053140096619,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031854134397941254,
      "loss": 4.9084,
      "mean_token_accuracy": 0.2134867489337921,
      "num_tokens": 97231884.0,
      "step": 52695
    },
    {
      "entropy": 5.497443103790284,
      "epoch": 4.427473219911783,
      "grad_norm": 1.0625,
      "learning_rate": 0.000318511963127102,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21707534193992614,
      "num_tokens": 97241604.0,
      "step": 52700
    },
    {
      "entropy": 5.474092817306518,
      "epoch": 4.427893299726948,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003184825815039995,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21290517896413802,
      "num_tokens": 97250385.0,
      "step": 52705
    },
    {
      "entropy": 5.4097248077392575,
      "epoch": 4.428313379542113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003184531991106254,
      "loss": 4.8486,
      "mean_token_accuracy": 0.2202730268239975,
      "num_tokens": 97260014.0,
      "step": 52710
    },
    {
      "entropy": 5.445449876785278,
      "epoch": 4.428733459357278,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003184238159475003,
      "loss": 4.8967,
      "mean_token_accuracy": 0.2194458231329918,
      "num_tokens": 97269617.0,
      "step": 52715
    },
    {
      "entropy": 5.513185310363769,
      "epoch": 4.429153539172443,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003183944320151447,
      "loss": 4.9312,
      "mean_token_accuracy": 0.20756967216730118,
      "num_tokens": 97278867.0,
      "step": 52720
    },
    {
      "entropy": 5.45350136756897,
      "epoch": 4.429573618987607,
      "grad_norm": 1.09375,
      "learning_rate": 0.000318365047314079,
      "loss": 4.932,
      "mean_token_accuracy": 0.21024617105722426,
      "num_tokens": 97288281.0,
      "step": 52725
    },
    {
      "entropy": 5.388107347488403,
      "epoch": 4.429993698802773,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031833566184482384,
      "loss": 4.8183,
      "mean_token_accuracy": 0.2260854035615921,
      "num_tokens": 97297859.0,
      "step": 52730
    },
    {
      "entropy": 5.490490341186524,
      "epoch": 4.430413778617938,
      "grad_norm": 1.125,
      "learning_rate": 0.00031830627560789975,
      "loss": 4.8764,
      "mean_token_accuracy": 0.2178186446428299,
      "num_tokens": 97306821.0,
      "step": 52735
    },
    {
      "entropy": 5.380337715148926,
      "epoch": 4.430833858433102,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003182768886038272,
      "loss": 4.83,
      "mean_token_accuracy": 0.22362343072891236,
      "num_tokens": 97315354.0,
      "step": 52740
    },
    {
      "entropy": 5.412058925628662,
      "epoch": 4.431253938248267,
      "grad_norm": 1.140625,
      "learning_rate": 0.000318247500833127,
      "loss": 4.8969,
      "mean_token_accuracy": 0.2187800019979477,
      "num_tokens": 97324489.0,
      "step": 52745
    },
    {
      "entropy": 5.403015184402466,
      "epoch": 4.431674018063432,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031821811229631954,
      "loss": 4.8857,
      "mean_token_accuracy": 0.2134777843952179,
      "num_tokens": 97333590.0,
      "step": 52750
    },
    {
      "entropy": 5.428277063369751,
      "epoch": 4.432094097878597,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003181887229939254,
      "loss": 4.8581,
      "mean_token_accuracy": 0.22157347649335862,
      "num_tokens": 97343817.0,
      "step": 52755
    },
    {
      "entropy": 5.396492862701416,
      "epoch": 4.432514177693762,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031815933292646535,
      "loss": 4.8756,
      "mean_token_accuracy": 0.21787190437316895,
      "num_tokens": 97353871.0,
      "step": 52760
    },
    {
      "entropy": 5.419612407684326,
      "epoch": 4.432934257508927,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031812994209445996,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2195767268538475,
      "num_tokens": 97363336.0,
      "step": 52765
    },
    {
      "entropy": 5.465654516220093,
      "epoch": 4.433354337324092,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031810055049842986,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22152291983366013,
      "num_tokens": 97371931.0,
      "step": 52770
    },
    {
      "entropy": 5.353368854522705,
      "epoch": 4.433774417139256,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003180711581388957,
      "loss": 4.7291,
      "mean_token_accuracy": 0.2363284334540367,
      "num_tokens": 97380552.0,
      "step": 52775
    },
    {
      "entropy": 5.401114225387573,
      "epoch": 4.434194496954421,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031804176501637807,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22030171900987625,
      "num_tokens": 97390217.0,
      "step": 52780
    },
    {
      "entropy": 5.399984502792359,
      "epoch": 4.434614576769587,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031801237113139775,
      "loss": 4.7992,
      "mean_token_accuracy": 0.21502979397773742,
      "num_tokens": 97398973.0,
      "step": 52785
    },
    {
      "entropy": 5.407816076278687,
      "epoch": 4.435034656584751,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003179829764844755,
      "loss": 4.8098,
      "mean_token_accuracy": 0.22759129106998444,
      "num_tokens": 97407922.0,
      "step": 52790
    },
    {
      "entropy": 5.525967025756836,
      "epoch": 4.435454736399916,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031795358107613187,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21190026998519898,
      "num_tokens": 97417036.0,
      "step": 52795
    },
    {
      "entropy": 5.517636823654175,
      "epoch": 4.4358748162150805,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031792418490688773,
      "loss": 4.9438,
      "mean_token_accuracy": 0.21037257015705108,
      "num_tokens": 97426590.0,
      "step": 52800
    },
    {
      "entropy": 5.457188272476197,
      "epoch": 4.436294896030246,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003178947879772637,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2123267576098442,
      "num_tokens": 97435878.0,
      "step": 52805
    },
    {
      "entropy": 5.387064361572266,
      "epoch": 4.436714975845411,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003178653902877807,
      "loss": 4.8216,
      "mean_token_accuracy": 0.2214900881052017,
      "num_tokens": 97444726.0,
      "step": 52810
    },
    {
      "entropy": 5.404617261886597,
      "epoch": 4.437135055660575,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003178359918389593,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22360433787107467,
      "num_tokens": 97453713.0,
      "step": 52815
    },
    {
      "entropy": 5.451097631454468,
      "epoch": 4.4375551354757405,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031780659263132033,
      "loss": 4.9578,
      "mean_token_accuracy": 0.2066273346543312,
      "num_tokens": 97463980.0,
      "step": 52820
    },
    {
      "entropy": 5.456020164489746,
      "epoch": 4.437975215290905,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031777719266538467,
      "loss": 4.8919,
      "mean_token_accuracy": 0.21130561083555222,
      "num_tokens": 97472857.0,
      "step": 52825
    },
    {
      "entropy": 5.44453239440918,
      "epoch": 4.43839529510607,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031774779194167305,
      "loss": 4.9225,
      "mean_token_accuracy": 0.2153715744614601,
      "num_tokens": 97481463.0,
      "step": 52830
    },
    {
      "entropy": 5.377542924880982,
      "epoch": 4.438815374921235,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003177183904607063,
      "loss": 4.799,
      "mean_token_accuracy": 0.2222690239548683,
      "num_tokens": 97490192.0,
      "step": 52835
    },
    {
      "entropy": 5.439916133880615,
      "epoch": 4.4392354547364,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003176889882230052,
      "loss": 4.9337,
      "mean_token_accuracy": 0.21288691610097885,
      "num_tokens": 97499211.0,
      "step": 52840
    },
    {
      "entropy": 5.458006429672241,
      "epoch": 4.439655534551565,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031765958522909075,
      "loss": 4.9679,
      "mean_token_accuracy": 0.214278681576252,
      "num_tokens": 97508600.0,
      "step": 52845
    },
    {
      "entropy": 5.566705179214478,
      "epoch": 4.44007561436673,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031763018147948374,
      "loss": 5.0327,
      "mean_token_accuracy": 0.2063508227467537,
      "num_tokens": 97518059.0,
      "step": 52850
    },
    {
      "entropy": 5.47790904045105,
      "epoch": 4.4404956941818945,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000317600776974705,
      "loss": 4.8952,
      "mean_token_accuracy": 0.22214568704366683,
      "num_tokens": 97526630.0,
      "step": 52855
    },
    {
      "entropy": 5.412748718261719,
      "epoch": 4.44091577399706,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031757137171527537,
      "loss": 4.8104,
      "mean_token_accuracy": 0.21993699967861174,
      "num_tokens": 97536118.0,
      "step": 52860
    },
    {
      "entropy": 5.42773666381836,
      "epoch": 4.441335853812224,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003175419657017158,
      "loss": 4.9329,
      "mean_token_accuracy": 0.213163098692894,
      "num_tokens": 97545285.0,
      "step": 52865
    },
    {
      "entropy": 5.385793972015381,
      "epoch": 4.441755933627389,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003175125589345473,
      "loss": 4.9037,
      "mean_token_accuracy": 0.21895668059587478,
      "num_tokens": 97553984.0,
      "step": 52870
    },
    {
      "entropy": 5.454912948608398,
      "epoch": 4.442176013442554,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003174831514142906,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21570868492126466,
      "num_tokens": 97563106.0,
      "step": 52875
    },
    {
      "entropy": 5.441811037063599,
      "epoch": 4.442596093257719,
      "grad_norm": 1.25,
      "learning_rate": 0.0003174537431414668,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21436492055654527,
      "num_tokens": 97572710.0,
      "step": 52880
    },
    {
      "entropy": 5.490273380279541,
      "epoch": 4.443016173072884,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003174243341165968,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20936359763145446,
      "num_tokens": 97582644.0,
      "step": 52885
    },
    {
      "entropy": 5.515843725204467,
      "epoch": 4.443436252888048,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031739492434020163,
      "loss": 4.9703,
      "mean_token_accuracy": 0.20948172062635423,
      "num_tokens": 97592409.0,
      "step": 52890
    },
    {
      "entropy": 5.497490787506104,
      "epoch": 4.443856332703214,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003173655138128021,
      "loss": 4.919,
      "mean_token_accuracy": 0.2072652131319046,
      "num_tokens": 97601254.0,
      "step": 52895
    },
    {
      "entropy": 5.305904340744019,
      "epoch": 4.444276412518379,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003173361025349194,
      "loss": 4.7892,
      "mean_token_accuracy": 0.22000620812177657,
      "num_tokens": 97610065.0,
      "step": 52900
    },
    {
      "entropy": 5.359373283386231,
      "epoch": 4.444696492333543,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031730669050707437,
      "loss": 4.881,
      "mean_token_accuracy": 0.2240329921245575,
      "num_tokens": 97619488.0,
      "step": 52905
    },
    {
      "entropy": 5.4111326217651365,
      "epoch": 4.4451165721487085,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031727727772978815,
      "loss": 4.8577,
      "mean_token_accuracy": 0.22283989042043686,
      "num_tokens": 97629087.0,
      "step": 52910
    },
    {
      "entropy": 5.510238122940064,
      "epoch": 4.445536651963873,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031724786420358174,
      "loss": 4.8793,
      "mean_token_accuracy": 0.21372707486152648,
      "num_tokens": 97638264.0,
      "step": 52915
    },
    {
      "entropy": 5.439018201828003,
      "epoch": 4.445956731779038,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031721844992897615,
      "loss": 4.9182,
      "mean_token_accuracy": 0.21023769974708556,
      "num_tokens": 97646847.0,
      "step": 52920
    },
    {
      "entropy": 5.498687267303467,
      "epoch": 4.446376811594203,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031718903490649247,
      "loss": 4.9232,
      "mean_token_accuracy": 0.21308021247386932,
      "num_tokens": 97655548.0,
      "step": 52925
    },
    {
      "entropy": 5.403440475463867,
      "epoch": 4.446796891409368,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003171596191366517,
      "loss": 4.8862,
      "mean_token_accuracy": 0.21682589054107665,
      "num_tokens": 97665478.0,
      "step": 52930
    },
    {
      "entropy": 5.387605381011963,
      "epoch": 4.447216971224533,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031713020261997505,
      "loss": 4.8221,
      "mean_token_accuracy": 0.22976971417665482,
      "num_tokens": 97673934.0,
      "step": 52935
    },
    {
      "entropy": 5.439334106445313,
      "epoch": 4.447637051039697,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003171007853569835,
      "loss": 4.8907,
      "mean_token_accuracy": 0.21080342978239058,
      "num_tokens": 97683502.0,
      "step": 52940
    },
    {
      "entropy": 5.558388376235962,
      "epoch": 4.448057130854862,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003170713673481982,
      "loss": 4.9655,
      "mean_token_accuracy": 0.21658090651035308,
      "num_tokens": 97692490.0,
      "step": 52945
    },
    {
      "entropy": 5.4215775489807125,
      "epoch": 4.448477210670028,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031704194859414036,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21398452818393707,
      "num_tokens": 97701589.0,
      "step": 52950
    },
    {
      "entropy": 5.394556999206543,
      "epoch": 4.448897290485192,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031701252909533106,
      "loss": 4.8603,
      "mean_token_accuracy": 0.22544134259223939,
      "num_tokens": 97710343.0,
      "step": 52955
    },
    {
      "entropy": 5.465783405303955,
      "epoch": 4.449317370300357,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003169831088522914,
      "loss": 4.9361,
      "mean_token_accuracy": 0.217232908308506,
      "num_tokens": 97719921.0,
      "step": 52960
    },
    {
      "entropy": 5.437366104125976,
      "epoch": 4.449737450115522,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031695368786554255,
      "loss": 4.9264,
      "mean_token_accuracy": 0.20964955687522888,
      "num_tokens": 97728766.0,
      "step": 52965
    },
    {
      "entropy": 5.41498646736145,
      "epoch": 4.450157529930687,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031692426613560574,
      "loss": 4.905,
      "mean_token_accuracy": 0.21556129604578017,
      "num_tokens": 97737904.0,
      "step": 52970
    },
    {
      "entropy": 5.519631338119507,
      "epoch": 4.450577609745852,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031689484366300227,
      "loss": 4.9768,
      "mean_token_accuracy": 0.20318000465631486,
      "num_tokens": 97747713.0,
      "step": 52975
    },
    {
      "entropy": 5.551828289031983,
      "epoch": 4.450997689561016,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003168654204482531,
      "loss": 4.978,
      "mean_token_accuracy": 0.21177821159362792,
      "num_tokens": 97758167.0,
      "step": 52980
    },
    {
      "entropy": 5.501491212844849,
      "epoch": 4.451417769376182,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031683599649187953,
      "loss": 4.9439,
      "mean_token_accuracy": 0.21458910703659057,
      "num_tokens": 97767045.0,
      "step": 52985
    },
    {
      "entropy": 5.479863500595092,
      "epoch": 4.451837849191346,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003168065717944029,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21055239588022232,
      "num_tokens": 97775893.0,
      "step": 52990
    },
    {
      "entropy": 5.443522882461548,
      "epoch": 4.452257929006511,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003167771463563444,
      "loss": 4.8786,
      "mean_token_accuracy": 0.2134924814105034,
      "num_tokens": 97785318.0,
      "step": 52995
    },
    {
      "entropy": 5.452366733551026,
      "epoch": 4.452678008821676,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003167477201782252,
      "loss": 4.8701,
      "mean_token_accuracy": 0.21735883951187135,
      "num_tokens": 97794290.0,
      "step": 53000
    },
    {
      "entropy": 5.506940460205078,
      "epoch": 4.453098088636841,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003167182932605666,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21549326330423355,
      "num_tokens": 97802735.0,
      "step": 53005
    },
    {
      "entropy": 5.535981893539429,
      "epoch": 4.453518168452006,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003166888656038899,
      "loss": 5.001,
      "mean_token_accuracy": 0.20935046970844268,
      "num_tokens": 97813088.0,
      "step": 53010
    },
    {
      "entropy": 5.474702215194702,
      "epoch": 4.453938248267171,
      "grad_norm": 1.046875,
      "learning_rate": 0.00031665943720871656,
      "loss": 4.9321,
      "mean_token_accuracy": 0.21474050283432006,
      "num_tokens": 97822765.0,
      "step": 53015
    },
    {
      "entropy": 5.4788800239562985,
      "epoch": 4.454358328082336,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031663000807556765,
      "loss": 4.883,
      "mean_token_accuracy": 0.2139604866504669,
      "num_tokens": 97832228.0,
      "step": 53020
    },
    {
      "entropy": 5.53567590713501,
      "epoch": 4.454778407897501,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003166005782049646,
      "loss": 4.9542,
      "mean_token_accuracy": 0.2152375817298889,
      "num_tokens": 97840858.0,
      "step": 53025
    },
    {
      "entropy": 5.458631181716919,
      "epoch": 4.455198487712665,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031657114759742867,
      "loss": 4.9552,
      "mean_token_accuracy": 0.21127670109272004,
      "num_tokens": 97849451.0,
      "step": 53030
    },
    {
      "entropy": 5.483599710464477,
      "epoch": 4.45561856752783,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003165417162534813,
      "loss": 4.9273,
      "mean_token_accuracy": 0.20972698479890822,
      "num_tokens": 97858269.0,
      "step": 53035
    },
    {
      "entropy": 5.375989866256714,
      "epoch": 4.456038647342995,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003165122841736439,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22930460423231125,
      "num_tokens": 97867419.0,
      "step": 53040
    },
    {
      "entropy": 5.446200847625732,
      "epoch": 4.45645872715816,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031648285135843767,
      "loss": 4.9498,
      "mean_token_accuracy": 0.21701389849185942,
      "num_tokens": 97876991.0,
      "step": 53045
    },
    {
      "entropy": 5.570157527923584,
      "epoch": 4.456878806973325,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003164534178083841,
      "loss": 5.0263,
      "mean_token_accuracy": 0.2040518894791603,
      "num_tokens": 97885131.0,
      "step": 53050
    },
    {
      "entropy": 5.465914583206176,
      "epoch": 4.4572988867884895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003164239835240046,
      "loss": 4.8422,
      "mean_token_accuracy": 0.21671239733695985,
      "num_tokens": 97894003.0,
      "step": 53055
    },
    {
      "entropy": 5.519971323013306,
      "epoch": 4.457718966603655,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003163945485058206,
      "loss": 5.0415,
      "mean_token_accuracy": 0.21562414169311522,
      "num_tokens": 97904143.0,
      "step": 53060
    },
    {
      "entropy": 5.53253960609436,
      "epoch": 4.45813904641882,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003163651127543534,
      "loss": 4.9606,
      "mean_token_accuracy": 0.21282820552587509,
      "num_tokens": 97913794.0,
      "step": 53065
    },
    {
      "entropy": 5.486976861953735,
      "epoch": 4.458559126233984,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003163356762701246,
      "loss": 4.8985,
      "mean_token_accuracy": 0.2166242465376854,
      "num_tokens": 97922881.0,
      "step": 53070
    },
    {
      "entropy": 5.410591697692871,
      "epoch": 4.4589792060491495,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031630623905365554,
      "loss": 4.8852,
      "mean_token_accuracy": 0.21914468705654144,
      "num_tokens": 97931647.0,
      "step": 53075
    },
    {
      "entropy": 5.3040543556213375,
      "epoch": 4.459399285864314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003162768011054677,
      "loss": 4.7934,
      "mean_token_accuracy": 0.23215265870094298,
      "num_tokens": 97941061.0,
      "step": 53080
    },
    {
      "entropy": 5.3896262645721436,
      "epoch": 4.459819365679479,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031624736242608266,
      "loss": 4.8037,
      "mean_token_accuracy": 0.2185826525092125,
      "num_tokens": 97949913.0,
      "step": 53085
    },
    {
      "entropy": 5.536066198348999,
      "epoch": 4.460239445494644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003162179230160218,
      "loss": 5.0347,
      "mean_token_accuracy": 0.20899847596883775,
      "num_tokens": 97959632.0,
      "step": 53090
    },
    {
      "entropy": 5.48373589515686,
      "epoch": 4.460659525309809,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003161884828758066,
      "loss": 4.8892,
      "mean_token_accuracy": 0.21736445873975754,
      "num_tokens": 97968688.0,
      "step": 53095
    },
    {
      "entropy": 5.425414991378784,
      "epoch": 4.461079605124974,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003161590420059587,
      "loss": 4.86,
      "mean_token_accuracy": 0.22178848683834076,
      "num_tokens": 97978645.0,
      "step": 53100
    },
    {
      "entropy": 5.442144870758057,
      "epoch": 4.461499684940138,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003161296004069995,
      "loss": 4.8691,
      "mean_token_accuracy": 0.2188206896185875,
      "num_tokens": 97987562.0,
      "step": 53105
    },
    {
      "entropy": 5.438197422027588,
      "epoch": 4.4619197647553035,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003161001580794507,
      "loss": 4.8685,
      "mean_token_accuracy": 0.2182604894042015,
      "num_tokens": 97995815.0,
      "step": 53110
    },
    {
      "entropy": 5.385009908676148,
      "epoch": 4.462339844570469,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003160707150238337,
      "loss": 4.8939,
      "mean_token_accuracy": 0.21594319939613343,
      "num_tokens": 98005207.0,
      "step": 53115
    },
    {
      "entropy": 5.449315166473388,
      "epoch": 4.462759924385633,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003160412712406702,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21854565888643265,
      "num_tokens": 98014113.0,
      "step": 53120
    },
    {
      "entropy": 5.437417650222779,
      "epoch": 4.463180004200798,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003160118267304817,
      "loss": 4.8912,
      "mean_token_accuracy": 0.22020445615053177,
      "num_tokens": 98023062.0,
      "step": 53125
    },
    {
      "entropy": 5.57524471282959,
      "epoch": 4.463600084015963,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003159823814937897,
      "loss": 4.9871,
      "mean_token_accuracy": 0.2121908634901047,
      "num_tokens": 98033297.0,
      "step": 53130
    },
    {
      "entropy": 5.495550060272217,
      "epoch": 4.464020163831128,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031595293553111605,
      "loss": 4.9216,
      "mean_token_accuracy": 0.21283296048641204,
      "num_tokens": 98042822.0,
      "step": 53135
    },
    {
      "entropy": 5.429848146438599,
      "epoch": 4.464440243646293,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031592348884298217,
      "loss": 4.959,
      "mean_token_accuracy": 0.21605680286884307,
      "num_tokens": 98052714.0,
      "step": 53140
    },
    {
      "entropy": 5.484619426727295,
      "epoch": 4.4648603234614574,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003158940414299097,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21166001856327057,
      "num_tokens": 98061506.0,
      "step": 53145
    },
    {
      "entropy": 5.387711858749389,
      "epoch": 4.465280403276623,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031586459329242036,
      "loss": 4.7829,
      "mean_token_accuracy": 0.22444144487380982,
      "num_tokens": 98071295.0,
      "step": 53150
    },
    {
      "entropy": 5.431887483596801,
      "epoch": 4.465700483091787,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003158351444310359,
      "loss": 4.8737,
      "mean_token_accuracy": 0.21118629574775696,
      "num_tokens": 98080087.0,
      "step": 53155
    },
    {
      "entropy": 5.426849794387818,
      "epoch": 4.466120562906952,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031580569484627774,
      "loss": 4.9044,
      "mean_token_accuracy": 0.21496160924434662,
      "num_tokens": 98089164.0,
      "step": 53160
    },
    {
      "entropy": 5.3966718196868895,
      "epoch": 4.4665406427221175,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031577624453866783,
      "loss": 4.8397,
      "mean_token_accuracy": 0.2234419971704483,
      "num_tokens": 98098423.0,
      "step": 53165
    },
    {
      "entropy": 5.480627584457397,
      "epoch": 4.466960722537282,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003157467935087277,
      "loss": 4.8735,
      "mean_token_accuracy": 0.21604054123163224,
      "num_tokens": 98107391.0,
      "step": 53170
    },
    {
      "entropy": 5.365064525604248,
      "epoch": 4.467380802352447,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031571734175697913,
      "loss": 4.8111,
      "mean_token_accuracy": 0.2248449757695198,
      "num_tokens": 98116062.0,
      "step": 53175
    },
    {
      "entropy": 5.357926511764527,
      "epoch": 4.467800882167612,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003156878892839438,
      "loss": 4.8438,
      "mean_token_accuracy": 0.21995015293359757,
      "num_tokens": 98124686.0,
      "step": 53180
    },
    {
      "entropy": 5.466133785247803,
      "epoch": 4.468220961982777,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031565843609014345,
      "loss": 4.9359,
      "mean_token_accuracy": 0.2136639326810837,
      "num_tokens": 98133767.0,
      "step": 53185
    },
    {
      "entropy": 5.448021459579468,
      "epoch": 4.468641041797942,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003156289821760998,
      "loss": 4.8549,
      "mean_token_accuracy": 0.2160454124212265,
      "num_tokens": 98143018.0,
      "step": 53190
    },
    {
      "entropy": 5.474397373199463,
      "epoch": 4.469061121613106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031559952754233475,
      "loss": 4.8857,
      "mean_token_accuracy": 0.21651339530944824,
      "num_tokens": 98152452.0,
      "step": 53195
    },
    {
      "entropy": 5.401927185058594,
      "epoch": 4.469481201428271,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031557007218936987,
      "loss": 4.8891,
      "mean_token_accuracy": 0.21298731565475465,
      "num_tokens": 98161274.0,
      "step": 53200
    },
    {
      "entropy": 5.396775531768799,
      "epoch": 4.469901281243437,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031554061611772715,
      "loss": 4.8307,
      "mean_token_accuracy": 0.21682164072990417,
      "num_tokens": 98170510.0,
      "step": 53205
    },
    {
      "entropy": 5.433926391601562,
      "epoch": 4.470321361058601,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031551115932792817,
      "loss": 4.8849,
      "mean_token_accuracy": 0.2176467627286911,
      "num_tokens": 98180124.0,
      "step": 53210
    },
    {
      "entropy": 5.487809467315674,
      "epoch": 4.470741440873766,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031548170182049495,
      "loss": 4.888,
      "mean_token_accuracy": 0.2176934912800789,
      "num_tokens": 98189857.0,
      "step": 53215
    },
    {
      "entropy": 5.42241473197937,
      "epoch": 4.471161520688931,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031545224359594927,
      "loss": 4.8702,
      "mean_token_accuracy": 0.22773234099149703,
      "num_tokens": 98198488.0,
      "step": 53220
    },
    {
      "entropy": 5.432129907608032,
      "epoch": 4.471581600504096,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003154227846548128,
      "loss": 4.9088,
      "mean_token_accuracy": 0.2190661460161209,
      "num_tokens": 98208105.0,
      "step": 53225
    },
    {
      "entropy": 5.568310022354126,
      "epoch": 4.472001680319261,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031539332499760757,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20130769908428192,
      "num_tokens": 98217289.0,
      "step": 53230
    },
    {
      "entropy": 5.422289276123047,
      "epoch": 4.472421760134425,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003153638646248554,
      "loss": 4.8666,
      "mean_token_accuracy": 0.21465734541416168,
      "num_tokens": 98226442.0,
      "step": 53235
    },
    {
      "entropy": 5.396089506149292,
      "epoch": 4.472841839949591,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003153344035370782,
      "loss": 4.8254,
      "mean_token_accuracy": 0.22231490761041642,
      "num_tokens": 98235600.0,
      "step": 53240
    },
    {
      "entropy": 5.477881574630738,
      "epoch": 4.473261919764755,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031530494173479776,
      "loss": 4.8789,
      "mean_token_accuracy": 0.20900893062353135,
      "num_tokens": 98244201.0,
      "step": 53245
    },
    {
      "entropy": 5.413246202468872,
      "epoch": 4.47368199957992,
      "grad_norm": 1.140625,
      "learning_rate": 0.000315275479218536,
      "loss": 4.8618,
      "mean_token_accuracy": 0.21578338891267776,
      "num_tokens": 98252738.0,
      "step": 53250
    },
    {
      "entropy": 5.487430715560913,
      "epoch": 4.474102079395085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031524601598881496,
      "loss": 4.9817,
      "mean_token_accuracy": 0.20874630212783812,
      "num_tokens": 98261739.0,
      "step": 53255
    },
    {
      "entropy": 5.428723955154419,
      "epoch": 4.47452215921025,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031521655204615635,
      "loss": 4.8662,
      "mean_token_accuracy": 0.21388293206691741,
      "num_tokens": 98271151.0,
      "step": 53260
    },
    {
      "entropy": 5.5254364013671875,
      "epoch": 4.474942239025415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003151870873910823,
      "loss": 4.9408,
      "mean_token_accuracy": 0.21761344522237777,
      "num_tokens": 98279975.0,
      "step": 53265
    },
    {
      "entropy": 5.482453346252441,
      "epoch": 4.475362318840579,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031515762202411466,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21160122007131577,
      "num_tokens": 98289232.0,
      "step": 53270
    },
    {
      "entropy": 5.441144037246704,
      "epoch": 4.475782398655745,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031512815594577547,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2232329860329628,
      "num_tokens": 98298035.0,
      "step": 53275
    },
    {
      "entropy": 5.587204885482788,
      "epoch": 4.47620247847091,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031509868915658665,
      "loss": 4.9682,
      "mean_token_accuracy": 0.21614712476730347,
      "num_tokens": 98306206.0,
      "step": 53280
    },
    {
      "entropy": 5.371900177001953,
      "epoch": 4.476622558286074,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031506922165707016,
      "loss": 4.7973,
      "mean_token_accuracy": 0.22184006720781327,
      "num_tokens": 98315480.0,
      "step": 53285
    },
    {
      "entropy": 5.415388011932373,
      "epoch": 4.477042638101239,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031503975344774807,
      "loss": 4.8569,
      "mean_token_accuracy": 0.2248992145061493,
      "num_tokens": 98324051.0,
      "step": 53290
    },
    {
      "entropy": 5.426820087432861,
      "epoch": 4.477462717916404,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031501028452914233,
      "loss": 4.9628,
      "mean_token_accuracy": 0.21723170578479767,
      "num_tokens": 98333652.0,
      "step": 53295
    },
    {
      "entropy": 5.455460166931152,
      "epoch": 4.477882797731569,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031498081490177504,
      "loss": 4.855,
      "mean_token_accuracy": 0.21625391989946366,
      "num_tokens": 98343087.0,
      "step": 53300
    },
    {
      "entropy": 5.495666551589966,
      "epoch": 4.478302877546734,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003149513445661682,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21022555679082872,
      "num_tokens": 98353301.0,
      "step": 53305
    },
    {
      "entropy": 5.505917882919311,
      "epoch": 4.4787229573618985,
      "grad_norm": 1.125,
      "learning_rate": 0.00031492187352284386,
      "loss": 4.963,
      "mean_token_accuracy": 0.21269094347953796,
      "num_tokens": 98362561.0,
      "step": 53310
    },
    {
      "entropy": 5.425669574737549,
      "epoch": 4.479143037177064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031489240177232406,
      "loss": 4.8064,
      "mean_token_accuracy": 0.22136002480983735,
      "num_tokens": 98371115.0,
      "step": 53315
    },
    {
      "entropy": 5.406041765213013,
      "epoch": 4.479563116992228,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003148629293151309,
      "loss": 4.9062,
      "mean_token_accuracy": 0.21190346777439117,
      "num_tokens": 98380504.0,
      "step": 53320
    },
    {
      "entropy": 5.406209945678711,
      "epoch": 4.479983196807393,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031483345615178647,
      "loss": 4.9216,
      "mean_token_accuracy": 0.2190944030880928,
      "num_tokens": 98390366.0,
      "step": 53325
    },
    {
      "entropy": 5.4192756652832035,
      "epoch": 4.4804032766225586,
      "grad_norm": 1.125,
      "learning_rate": 0.000314803982282813,
      "loss": 4.8946,
      "mean_token_accuracy": 0.22215379327535628,
      "num_tokens": 98398716.0,
      "step": 53330
    },
    {
      "entropy": 5.4296973705291744,
      "epoch": 4.480823356437723,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031477450770873235,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22195022255182267,
      "num_tokens": 98407490.0,
      "step": 53335
    },
    {
      "entropy": 5.431981801986694,
      "epoch": 4.481243436252888,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003147450324300668,
      "loss": 4.8989,
      "mean_token_accuracy": 0.21665359288454056,
      "num_tokens": 98416910.0,
      "step": 53340
    },
    {
      "entropy": 5.42791600227356,
      "epoch": 4.481663516068053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031471555644733854,
      "loss": 4.908,
      "mean_token_accuracy": 0.21492082327604295,
      "num_tokens": 98426607.0,
      "step": 53345
    },
    {
      "entropy": 5.4607813358306885,
      "epoch": 4.482083595883218,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003146860797610695,
      "loss": 4.9404,
      "mean_token_accuracy": 0.22275031059980394,
      "num_tokens": 98435536.0,
      "step": 53350
    },
    {
      "entropy": 5.462203168869019,
      "epoch": 4.482503675698383,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003146566023717821,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21607382744550704,
      "num_tokens": 98444055.0,
      "step": 53355
    },
    {
      "entropy": 5.380508279800415,
      "epoch": 4.482923755513547,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031462712427999834,
      "loss": 4.8365,
      "mean_token_accuracy": 0.2190446063876152,
      "num_tokens": 98452448.0,
      "step": 53360
    },
    {
      "entropy": 5.4369830131530765,
      "epoch": 4.4833438353287125,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031459764548624053,
      "loss": 4.9139,
      "mean_token_accuracy": 0.21510887891054153,
      "num_tokens": 98462413.0,
      "step": 53365
    },
    {
      "entropy": 5.447090291976929,
      "epoch": 4.483763915143878,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003145681659910308,
      "loss": 4.925,
      "mean_token_accuracy": 0.21292521059513092,
      "num_tokens": 98471410.0,
      "step": 53370
    },
    {
      "entropy": 5.5315016746521,
      "epoch": 4.484183994959042,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003145386857948914,
      "loss": 5.0118,
      "mean_token_accuracy": 0.21311484575271605,
      "num_tokens": 98480877.0,
      "step": 53375
    },
    {
      "entropy": 5.422076797485351,
      "epoch": 4.484604074774207,
      "grad_norm": 1.125,
      "learning_rate": 0.0003145092048983446,
      "loss": 4.7922,
      "mean_token_accuracy": 0.2215873509645462,
      "num_tokens": 98489570.0,
      "step": 53380
    },
    {
      "entropy": 5.411387538909912,
      "epoch": 4.485024154589372,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003144797233019125,
      "loss": 4.8913,
      "mean_token_accuracy": 0.21614248901605607,
      "num_tokens": 98498708.0,
      "step": 53385
    },
    {
      "entropy": 5.4638265609741214,
      "epoch": 4.485444234404537,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031445024100611743,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21098807007074355,
      "num_tokens": 98508034.0,
      "step": 53390
    },
    {
      "entropy": 5.482037162780761,
      "epoch": 4.485864314219702,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003144207580114817,
      "loss": 4.9744,
      "mean_token_accuracy": 0.2122819557785988,
      "num_tokens": 98516827.0,
      "step": 53395
    },
    {
      "entropy": 5.472130632400512,
      "epoch": 4.4862843940348665,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003143912743185275,
      "loss": 4.9834,
      "mean_token_accuracy": 0.20642450153827668,
      "num_tokens": 98525947.0,
      "step": 53400
    },
    {
      "entropy": 5.39224681854248,
      "epoch": 4.486704473850032,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031436178992777714,
      "loss": 4.8414,
      "mean_token_accuracy": 0.2190713629126549,
      "num_tokens": 98534706.0,
      "step": 53405
    },
    {
      "entropy": 5.47880220413208,
      "epoch": 4.487124553665196,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000314332304839753,
      "loss": 4.9273,
      "mean_token_accuracy": 0.21314302384853362,
      "num_tokens": 98544108.0,
      "step": 53410
    },
    {
      "entropy": 5.523857164382934,
      "epoch": 4.487544633480361,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031430281905497724,
      "loss": 4.9251,
      "mean_token_accuracy": 0.21778662502765656,
      "num_tokens": 98553630.0,
      "step": 53415
    },
    {
      "entropy": 5.467495059967041,
      "epoch": 4.4879647132955265,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003142733325739723,
      "loss": 4.8383,
      "mean_token_accuracy": 0.21941766738891602,
      "num_tokens": 98563081.0,
      "step": 53420
    },
    {
      "entropy": 5.40680685043335,
      "epoch": 4.488384793110691,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031424384539726057,
      "loss": 4.8202,
      "mean_token_accuracy": 0.22245641499757768,
      "num_tokens": 98572450.0,
      "step": 53425
    },
    {
      "entropy": 5.404620409011841,
      "epoch": 4.488804872925856,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003142143575253643,
      "loss": 4.8482,
      "mean_token_accuracy": 0.21935250759124755,
      "num_tokens": 98581536.0,
      "step": 53430
    },
    {
      "entropy": 5.39507098197937,
      "epoch": 4.48922495274102,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003141848689588059,
      "loss": 4.832,
      "mean_token_accuracy": 0.22560716718435286,
      "num_tokens": 98590513.0,
      "step": 53435
    },
    {
      "entropy": 5.3705517768859865,
      "epoch": 4.489645032556186,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003141553796981077,
      "loss": 4.8846,
      "mean_token_accuracy": 0.216270112991333,
      "num_tokens": 98600692.0,
      "step": 53440
    },
    {
      "entropy": 5.422472953796387,
      "epoch": 4.490065112371351,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031412588974379216,
      "loss": 4.8714,
      "mean_token_accuracy": 0.21503733843564987,
      "num_tokens": 98609195.0,
      "step": 53445
    },
    {
      "entropy": 5.463728380203247,
      "epoch": 4.490485192186515,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003140963990963816,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21156463474035264,
      "num_tokens": 98618183.0,
      "step": 53450
    },
    {
      "entropy": 5.422085905075074,
      "epoch": 4.49090527200168,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003140669077563984,
      "loss": 4.7984,
      "mean_token_accuracy": 0.22239491045475007,
      "num_tokens": 98626969.0,
      "step": 53455
    },
    {
      "entropy": 5.543868350982666,
      "epoch": 4.491325351816846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031403741572436506,
      "loss": 5.0421,
      "mean_token_accuracy": 0.2021412268280983,
      "num_tokens": 98636111.0,
      "step": 53460
    },
    {
      "entropy": 5.44164457321167,
      "epoch": 4.49174543163201,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031400792300080403,
      "loss": 4.9405,
      "mean_token_accuracy": 0.21015286743640899,
      "num_tokens": 98645683.0,
      "step": 53465
    },
    {
      "entropy": 5.5404431343078615,
      "epoch": 4.492165511447175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031397842958623775,
      "loss": 5.0095,
      "mean_token_accuracy": 0.20464869737625122,
      "num_tokens": 98654839.0,
      "step": 53470
    },
    {
      "entropy": 5.509757566452026,
      "epoch": 4.49258559126234,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003139489354811887,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20894163548946382,
      "num_tokens": 98663149.0,
      "step": 53475
    },
    {
      "entropy": 5.468457412719727,
      "epoch": 4.493005671077505,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031391944068617924,
      "loss": 4.8554,
      "mean_token_accuracy": 0.2224610447883606,
      "num_tokens": 98673302.0,
      "step": 53480
    },
    {
      "entropy": 5.4172797203063965,
      "epoch": 4.493425750892669,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031388994520173185,
      "loss": 4.8769,
      "mean_token_accuracy": 0.21398339718580245,
      "num_tokens": 98682154.0,
      "step": 53485
    },
    {
      "entropy": 5.433697652816773,
      "epoch": 4.493845830707834,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031386044902836924,
      "loss": 4.9213,
      "mean_token_accuracy": 0.20991375297307968,
      "num_tokens": 98690736.0,
      "step": 53490
    },
    {
      "entropy": 5.482519435882568,
      "epoch": 4.494265910523,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031383095216661375,
      "loss": 4.9309,
      "mean_token_accuracy": 0.2096562311053276,
      "num_tokens": 98699216.0,
      "step": 53495
    },
    {
      "entropy": 5.5122710227966305,
      "epoch": 4.494685990338164,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003138014546169879,
      "loss": 4.9558,
      "mean_token_accuracy": 0.20465680062770844,
      "num_tokens": 98708935.0,
      "step": 53500
    },
    {
      "entropy": 5.587067365646362,
      "epoch": 4.495106070153329,
      "grad_norm": 0.984375,
      "learning_rate": 0.00031377195638001426,
      "loss": 5.099,
      "mean_token_accuracy": 0.2039380893111229,
      "num_tokens": 98718870.0,
      "step": 53505
    },
    {
      "entropy": 5.404705429077149,
      "epoch": 4.495526149968494,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003137424574562154,
      "loss": 4.8256,
      "mean_token_accuracy": 0.2156616061925888,
      "num_tokens": 98728825.0,
      "step": 53510
    },
    {
      "entropy": 5.408054494857788,
      "epoch": 4.495946229783659,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003137129578461139,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22686939537525178,
      "num_tokens": 98738031.0,
      "step": 53515
    },
    {
      "entropy": 5.443265008926391,
      "epoch": 4.496366309598824,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003136834575502323,
      "loss": 4.8903,
      "mean_token_accuracy": 0.22019112408161162,
      "num_tokens": 98747443.0,
      "step": 53520
    },
    {
      "entropy": 5.516780996322632,
      "epoch": 4.496786389413988,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003136539565690931,
      "loss": 4.9746,
      "mean_token_accuracy": 0.21408906430006028,
      "num_tokens": 98755949.0,
      "step": 53525
    },
    {
      "entropy": 5.545882511138916,
      "epoch": 4.497206469229154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000313624454903219,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21764040738344193,
      "num_tokens": 98764330.0,
      "step": 53530
    },
    {
      "entropy": 5.370059204101563,
      "epoch": 4.497626549044319,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003135949525531325,
      "loss": 4.7195,
      "mean_token_accuracy": 0.2297000005841255,
      "num_tokens": 98772976.0,
      "step": 53535
    },
    {
      "entropy": 5.444423055648803,
      "epoch": 4.498046628859483,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003135654495193564,
      "loss": 4.855,
      "mean_token_accuracy": 0.214237679541111,
      "num_tokens": 98782135.0,
      "step": 53540
    },
    {
      "entropy": 5.430327367782593,
      "epoch": 4.498466708674648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003135359458024131,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2109684929251671,
      "num_tokens": 98791832.0,
      "step": 53545
    },
    {
      "entropy": 5.57566819190979,
      "epoch": 4.498886788489813,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031350644140282544,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20885028392076493,
      "num_tokens": 98801307.0,
      "step": 53550
    },
    {
      "entropy": 5.557552146911621,
      "epoch": 4.499306868304978,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031347693632111594,
      "loss": 4.9167,
      "mean_token_accuracy": 0.2127078577876091,
      "num_tokens": 98809706.0,
      "step": 53555
    },
    {
      "entropy": 5.4722600936889645,
      "epoch": 4.499726948120143,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003134474305578074,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21330588161945344,
      "num_tokens": 98818656.0,
      "step": 53560
    },
    {
      "entropy": 5.423731851577759,
      "epoch": 4.5001470279353075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031341792411342237,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2130034014582634,
      "num_tokens": 98827600.0,
      "step": 53565
    },
    {
      "entropy": 5.439676952362061,
      "epoch": 4.500567107750473,
      "grad_norm": 1.125,
      "learning_rate": 0.0003133884169884836,
      "loss": 4.917,
      "mean_token_accuracy": 0.21497246325016023,
      "num_tokens": 98836111.0,
      "step": 53570
    },
    {
      "entropy": 5.553131151199341,
      "epoch": 4.500987187565637,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003133589091835139,
      "loss": 5.0,
      "mean_token_accuracy": 0.21071190536022186,
      "num_tokens": 98845245.0,
      "step": 53575
    },
    {
      "entropy": 5.434970712661743,
      "epoch": 4.501407267380802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031332940069903574,
      "loss": 4.7788,
      "mean_token_accuracy": 0.22707661390304565,
      "num_tokens": 98854043.0,
      "step": 53580
    },
    {
      "entropy": 5.397494697570801,
      "epoch": 4.5018273471959676,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031329989153557205,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21413897424936296,
      "num_tokens": 98864042.0,
      "step": 53585
    },
    {
      "entropy": 5.44750747680664,
      "epoch": 4.502247427011132,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003132703816936455,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21424990445375441,
      "num_tokens": 98874072.0,
      "step": 53590
    },
    {
      "entropy": 5.470082235336304,
      "epoch": 4.502667506826297,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003132408711737788,
      "loss": 4.9129,
      "mean_token_accuracy": 0.21542454659938812,
      "num_tokens": 98883934.0,
      "step": 53595
    },
    {
      "entropy": 5.428835678100586,
      "epoch": 4.5030875866414615,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031321135997649483,
      "loss": 4.8863,
      "mean_token_accuracy": 0.21056631505489348,
      "num_tokens": 98893101.0,
      "step": 53600
    },
    {
      "entropy": 5.510517740249634,
      "epoch": 4.503507666456627,
      "grad_norm": 1.125,
      "learning_rate": 0.0003131818481023162,
      "loss": 4.9623,
      "mean_token_accuracy": 0.21569174230098725,
      "num_tokens": 98902423.0,
      "step": 53605
    },
    {
      "entropy": 5.456219911575317,
      "epoch": 4.503927746271792,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031315233555176587,
      "loss": 4.8761,
      "mean_token_accuracy": 0.2144124239683151,
      "num_tokens": 98910741.0,
      "step": 53610
    },
    {
      "entropy": 5.415083742141723,
      "epoch": 4.504347826086956,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003131228223253665,
      "loss": 4.8815,
      "mean_token_accuracy": 0.21781588941812516,
      "num_tokens": 98919216.0,
      "step": 53615
    },
    {
      "entropy": 5.457391262054443,
      "epoch": 4.5047679059021215,
      "grad_norm": 1.046875,
      "learning_rate": 0.000313093308423641,
      "loss": 4.8816,
      "mean_token_accuracy": 0.2267578884959221,
      "num_tokens": 98928444.0,
      "step": 53620
    },
    {
      "entropy": 5.406866264343262,
      "epoch": 4.505187985717287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003130637938471121,
      "loss": 4.845,
      "mean_token_accuracy": 0.22499793767929077,
      "num_tokens": 98937534.0,
      "step": 53625
    },
    {
      "entropy": 5.5008971214294435,
      "epoch": 4.505608065532451,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003130342785963027,
      "loss": 5.0151,
      "mean_token_accuracy": 0.20594724714756013,
      "num_tokens": 98946434.0,
      "step": 53630
    },
    {
      "entropy": 5.419393014907837,
      "epoch": 4.506028145347616,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031300476267173574,
      "loss": 4.7984,
      "mean_token_accuracy": 0.22749440670013427,
      "num_tokens": 98955474.0,
      "step": 53635
    },
    {
      "entropy": 5.383840894699096,
      "epoch": 4.506448225162781,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003129752460739339,
      "loss": 4.8348,
      "mean_token_accuracy": 0.2244684249162674,
      "num_tokens": 98963963.0,
      "step": 53640
    },
    {
      "entropy": 5.458723258972168,
      "epoch": 4.506868304977946,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003129457288034202,
      "loss": 4.8899,
      "mean_token_accuracy": 0.21847936809062957,
      "num_tokens": 98972871.0,
      "step": 53645
    },
    {
      "entropy": 5.499466371536255,
      "epoch": 4.50728838479311,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031291621086071734,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2169748514890671,
      "num_tokens": 98982513.0,
      "step": 53650
    },
    {
      "entropy": 5.457868003845215,
      "epoch": 4.5077084646082755,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031288669224634847,
      "loss": 4.9163,
      "mean_token_accuracy": 0.2160566046833992,
      "num_tokens": 98992639.0,
      "step": 53655
    },
    {
      "entropy": 5.450006771087646,
      "epoch": 4.508128544423441,
      "grad_norm": 1.125,
      "learning_rate": 0.0003128571729608363,
      "loss": 4.8649,
      "mean_token_accuracy": 0.22328678369522095,
      "num_tokens": 99001206.0,
      "step": 53660
    },
    {
      "entropy": 5.507611608505249,
      "epoch": 4.508548624238605,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003128276530047037,
      "loss": 4.9663,
      "mean_token_accuracy": 0.2155882552266121,
      "num_tokens": 99009928.0,
      "step": 53665
    },
    {
      "entropy": 5.436125993728638,
      "epoch": 4.50896870405377,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003127981323784739,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22028865814208984,
      "num_tokens": 99018389.0,
      "step": 53670
    },
    {
      "entropy": 5.47079176902771,
      "epoch": 4.5093887838689355,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031276861108266954,
      "loss": 4.9333,
      "mean_token_accuracy": 0.2091022700071335,
      "num_tokens": 99027863.0,
      "step": 53675
    },
    {
      "entropy": 5.492423439025879,
      "epoch": 4.5098088636841,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031273908911781364,
      "loss": 4.999,
      "mean_token_accuracy": 0.20750705152750015,
      "num_tokens": 99037753.0,
      "step": 53680
    },
    {
      "entropy": 5.415361928939819,
      "epoch": 4.510228943499265,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00031270956648442923,
      "loss": 4.7506,
      "mean_token_accuracy": 0.22594424784183503,
      "num_tokens": 99046571.0,
      "step": 53685
    },
    {
      "entropy": 5.434928464889526,
      "epoch": 4.510649023314429,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003126800431830392,
      "loss": 4.8675,
      "mean_token_accuracy": 0.22050926685333253,
      "num_tokens": 99054454.0,
      "step": 53690
    },
    {
      "entropy": 5.439047622680664,
      "epoch": 4.511069103129595,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003126505192141667,
      "loss": 4.9016,
      "mean_token_accuracy": 0.21631849259138108,
      "num_tokens": 99064657.0,
      "step": 53695
    },
    {
      "entropy": 5.429821395874024,
      "epoch": 4.51148918294476,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003126209945783347,
      "loss": 4.8508,
      "mean_token_accuracy": 0.21676862090826035,
      "num_tokens": 99074591.0,
      "step": 53700
    },
    {
      "entropy": 5.461758232116699,
      "epoch": 4.511909262759924,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031259146927606606,
      "loss": 4.9734,
      "mean_token_accuracy": 0.21272041350603105,
      "num_tokens": 99083777.0,
      "step": 53705
    },
    {
      "entropy": 5.431796598434448,
      "epoch": 4.512329342575089,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003125619433078839,
      "loss": 4.9329,
      "mean_token_accuracy": 0.21305366158485411,
      "num_tokens": 99092314.0,
      "step": 53710
    },
    {
      "entropy": 5.5615315437316895,
      "epoch": 4.512749422390254,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031253241667431134,
      "loss": 4.9937,
      "mean_token_accuracy": 0.21273328214883805,
      "num_tokens": 99101787.0,
      "step": 53715
    },
    {
      "entropy": 5.4874824523925785,
      "epoch": 4.513169502205419,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003125028893758712,
      "loss": 4.8873,
      "mean_token_accuracy": 0.21263317912817,
      "num_tokens": 99112032.0,
      "step": 53720
    },
    {
      "entropy": 5.512672853469849,
      "epoch": 4.513589582020584,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031247336141308686,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21278815120458602,
      "num_tokens": 99121671.0,
      "step": 53725
    },
    {
      "entropy": 5.5145152568817135,
      "epoch": 4.514009661835749,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003124438327864811,
      "loss": 4.973,
      "mean_token_accuracy": 0.20837592631578444,
      "num_tokens": 99132045.0,
      "step": 53730
    },
    {
      "entropy": 5.496447277069092,
      "epoch": 4.514429741650914,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031241430349657716,
      "loss": 5.0047,
      "mean_token_accuracy": 0.2004175066947937,
      "num_tokens": 99141821.0,
      "step": 53735
    },
    {
      "entropy": 5.446555042266846,
      "epoch": 4.514849821466079,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003123847735438981,
      "loss": 4.8363,
      "mean_token_accuracy": 0.2174776628613472,
      "num_tokens": 99151302.0,
      "step": 53740
    },
    {
      "entropy": 5.532415294647217,
      "epoch": 4.515269901281243,
      "grad_norm": 1.140625,
      "learning_rate": 0.000312355242928967,
      "loss": 5.0267,
      "mean_token_accuracy": 0.20553721934556962,
      "num_tokens": 99161209.0,
      "step": 53745
    },
    {
      "entropy": 5.484852647781372,
      "epoch": 4.515689981096409,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031232571165230703,
      "loss": 4.891,
      "mean_token_accuracy": 0.21525015085935592,
      "num_tokens": 99170828.0,
      "step": 53750
    },
    {
      "entropy": 5.455737161636352,
      "epoch": 4.516110060911573,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031229617971444124,
      "loss": 4.8733,
      "mean_token_accuracy": 0.21705916076898574,
      "num_tokens": 99178993.0,
      "step": 53755
    },
    {
      "entropy": 5.404990339279175,
      "epoch": 4.516530140726738,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031226664711589294,
      "loss": 4.8772,
      "mean_token_accuracy": 0.21104240119457246,
      "num_tokens": 99187855.0,
      "step": 53760
    },
    {
      "entropy": 5.420843887329101,
      "epoch": 4.5169502205419025,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031223711385718506,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2188960537314415,
      "num_tokens": 99197178.0,
      "step": 53765
    },
    {
      "entropy": 5.495751237869262,
      "epoch": 4.517370300357068,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000312207579938841,
      "loss": 4.9215,
      "mean_token_accuracy": 0.20843950510025025,
      "num_tokens": 99205992.0,
      "step": 53770
    },
    {
      "entropy": 5.500760555267334,
      "epoch": 4.517790380172233,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031217804536138374,
      "loss": 4.9374,
      "mean_token_accuracy": 0.2245072156190872,
      "num_tokens": 99214479.0,
      "step": 53775
    },
    {
      "entropy": 5.450986385345459,
      "epoch": 4.518210459987397,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003121485101253365,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2191474974155426,
      "num_tokens": 99224015.0,
      "step": 53780
    },
    {
      "entropy": 5.436478996276856,
      "epoch": 4.518630539802563,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031211897423122267,
      "loss": 4.8887,
      "mean_token_accuracy": 0.21863277405500411,
      "num_tokens": 99233240.0,
      "step": 53785
    },
    {
      "entropy": 5.465553855895996,
      "epoch": 4.519050619617728,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031208943767956527,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21224070638418197,
      "num_tokens": 99242984.0,
      "step": 53790
    },
    {
      "entropy": 5.477675724029541,
      "epoch": 4.519470699432892,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003120599004708876,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2142531767487526,
      "num_tokens": 99251789.0,
      "step": 53795
    },
    {
      "entropy": 5.435706377029419,
      "epoch": 4.519890779248057,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003120303626057128,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22310758680105208,
      "num_tokens": 99260365.0,
      "step": 53800
    },
    {
      "entropy": 5.448882055282593,
      "epoch": 4.520310859063222,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003120008240845642,
      "loss": 4.9015,
      "mean_token_accuracy": 0.21904115676879882,
      "num_tokens": 99269401.0,
      "step": 53805
    },
    {
      "entropy": 5.4023909091949465,
      "epoch": 4.520730938878387,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031197128490796514,
      "loss": 4.9131,
      "mean_token_accuracy": 0.2159987911581993,
      "num_tokens": 99278181.0,
      "step": 53810
    },
    {
      "entropy": 5.489183568954468,
      "epoch": 4.521151018693551,
      "grad_norm": 0.984375,
      "learning_rate": 0.00031194174507643874,
      "loss": 4.8471,
      "mean_token_accuracy": 0.22173991352319716,
      "num_tokens": 99287463.0,
      "step": 53815
    },
    {
      "entropy": 5.522190046310425,
      "epoch": 4.5215710985087165,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031191220459050834,
      "loss": 4.9023,
      "mean_token_accuracy": 0.21701218485832213,
      "num_tokens": 99296102.0,
      "step": 53820
    },
    {
      "entropy": 5.503693294525147,
      "epoch": 4.521991178323882,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003118826634506973,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21071493476629258,
      "num_tokens": 99306103.0,
      "step": 53825
    },
    {
      "entropy": 5.475738620758056,
      "epoch": 4.522411258139046,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031185312165752884,
      "loss": 4.8961,
      "mean_token_accuracy": 0.21821934431791307,
      "num_tokens": 99314920.0,
      "step": 53830
    },
    {
      "entropy": 5.455300283432007,
      "epoch": 4.522831337954211,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031182357921152626,
      "loss": 4.9408,
      "mean_token_accuracy": 0.20739528089761733,
      "num_tokens": 99324351.0,
      "step": 53835
    },
    {
      "entropy": 5.422160291671753,
      "epoch": 4.523251417769377,
      "grad_norm": 1.15625,
      "learning_rate": 0.000311794036113213,
      "loss": 4.9076,
      "mean_token_accuracy": 0.21968408524990082,
      "num_tokens": 99333396.0,
      "step": 53840
    },
    {
      "entropy": 5.432770490646362,
      "epoch": 4.523671497584541,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003117644923631123,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22153882682323456,
      "num_tokens": 99342830.0,
      "step": 53845
    },
    {
      "entropy": 5.4199512004852295,
      "epoch": 4.524091577399706,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003117349479617476,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22963169813156128,
      "num_tokens": 99351786.0,
      "step": 53850
    },
    {
      "entropy": 5.520252752304077,
      "epoch": 4.5245116572148705,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003117054029096422,
      "loss": 4.9559,
      "mean_token_accuracy": 0.2167395070195198,
      "num_tokens": 99361737.0,
      "step": 53855
    },
    {
      "entropy": 5.430084657669068,
      "epoch": 4.524931737030036,
      "grad_norm": 1.125,
      "learning_rate": 0.0003116758572073194,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2208741471171379,
      "num_tokens": 99371435.0,
      "step": 53860
    },
    {
      "entropy": 5.501640367507934,
      "epoch": 4.525351816845201,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031164631085530277,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21964914500713348,
      "num_tokens": 99380312.0,
      "step": 53865
    },
    {
      "entropy": 5.507298278808594,
      "epoch": 4.525771896660365,
      "grad_norm": 1.125,
      "learning_rate": 0.00031161676385411556,
      "loss": 4.9603,
      "mean_token_accuracy": 0.21613744348287584,
      "num_tokens": 99389573.0,
      "step": 53870
    },
    {
      "entropy": 5.497798347473145,
      "epoch": 4.5261919764755305,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003115872162042813,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21717147827148436,
      "num_tokens": 99399202.0,
      "step": 53875
    },
    {
      "entropy": 5.435253143310547,
      "epoch": 4.526612056290695,
      "grad_norm": 1.25,
      "learning_rate": 0.0003115576679063233,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2200694590806961,
      "num_tokens": 99407615.0,
      "step": 53880
    },
    {
      "entropy": 5.46064395904541,
      "epoch": 4.52703213610586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031152811896076506,
      "loss": 4.9077,
      "mean_token_accuracy": 0.21852820068597795,
      "num_tokens": 99416293.0,
      "step": 53885
    },
    {
      "entropy": 5.491093301773072,
      "epoch": 4.527452215921025,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003114985693681299,
      "loss": 4.917,
      "mean_token_accuracy": 0.2180670514702797,
      "num_tokens": 99425753.0,
      "step": 53890
    },
    {
      "entropy": 5.538564014434814,
      "epoch": 4.52787229573619,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031146901912894147,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21837958842515945,
      "num_tokens": 99434383.0,
      "step": 53895
    },
    {
      "entropy": 5.442461442947388,
      "epoch": 4.528292375551355,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031143946824372314,
      "loss": 4.9346,
      "mean_token_accuracy": 0.21127525568008423,
      "num_tokens": 99443663.0,
      "step": 53900
    },
    {
      "entropy": 5.438608026504516,
      "epoch": 4.52871245536652,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003114099167129984,
      "loss": 4.8925,
      "mean_token_accuracy": 0.21368128359317778,
      "num_tokens": 99454200.0,
      "step": 53905
    },
    {
      "entropy": 5.51953501701355,
      "epoch": 4.5291325351816845,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031138036453729075,
      "loss": 4.9441,
      "mean_token_accuracy": 0.2108748137950897,
      "num_tokens": 99462867.0,
      "step": 53910
    },
    {
      "entropy": 5.588230848312378,
      "epoch": 4.52955261499685,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003113508117171236,
      "loss": 5.0068,
      "mean_token_accuracy": 0.21835036873817443,
      "num_tokens": 99472882.0,
      "step": 53915
    },
    {
      "entropy": 5.512640476226807,
      "epoch": 4.529972694812014,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003113212582530206,
      "loss": 4.9634,
      "mean_token_accuracy": 0.2072130873799324,
      "num_tokens": 99482043.0,
      "step": 53920
    },
    {
      "entropy": 5.429971551895141,
      "epoch": 4.530392774627179,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003112917041455052,
      "loss": 4.897,
      "mean_token_accuracy": 0.2201188921928406,
      "num_tokens": 99491659.0,
      "step": 53925
    },
    {
      "entropy": 5.518028306961059,
      "epoch": 4.530812854442344,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003112621493951009,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22013195008039474,
      "num_tokens": 99501947.0,
      "step": 53930
    },
    {
      "entropy": 5.48227972984314,
      "epoch": 4.531232934257509,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003112325940023314,
      "loss": 4.88,
      "mean_token_accuracy": 0.2116911917924881,
      "num_tokens": 99510740.0,
      "step": 53935
    },
    {
      "entropy": 5.437436294555664,
      "epoch": 4.531653014072674,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003112030379677201,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21496642678976058,
      "num_tokens": 99520519.0,
      "step": 53940
    },
    {
      "entropy": 5.463752555847168,
      "epoch": 4.532073093887838,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003111734812917906,
      "loss": 4.8533,
      "mean_token_accuracy": 0.22301608175039292,
      "num_tokens": 99529161.0,
      "step": 53945
    },
    {
      "entropy": 5.42420916557312,
      "epoch": 4.532493173703004,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031114392397506654,
      "loss": 4.8843,
      "mean_token_accuracy": 0.21334288120269776,
      "num_tokens": 99538400.0,
      "step": 53950
    },
    {
      "entropy": 5.48500919342041,
      "epoch": 4.532913253518169,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003111143660180714,
      "loss": 4.9449,
      "mean_token_accuracy": 0.21174023449420928,
      "num_tokens": 99548063.0,
      "step": 53955
    },
    {
      "entropy": 5.453989791870117,
      "epoch": 4.533333333333333,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031108480742132895,
      "loss": 4.8899,
      "mean_token_accuracy": 0.2228432387113571,
      "num_tokens": 99557043.0,
      "step": 53960
    },
    {
      "entropy": 5.5076395034790036,
      "epoch": 4.5337534131484984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003110552481853627,
      "loss": 4.9018,
      "mean_token_accuracy": 0.217052686214447,
      "num_tokens": 99564931.0,
      "step": 53965
    },
    {
      "entropy": 5.526358604431152,
      "epoch": 4.534173492963663,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003110256883106963,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2178504839539528,
      "num_tokens": 99574405.0,
      "step": 53970
    },
    {
      "entropy": 5.521481704711914,
      "epoch": 4.534593572778828,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031099612779785337,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21847476661205292,
      "num_tokens": 99583976.0,
      "step": 53975
    },
    {
      "entropy": 5.538613510131836,
      "epoch": 4.535013652593992,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00031096656664735754,
      "loss": 4.9543,
      "mean_token_accuracy": 0.21098721772432327,
      "num_tokens": 99593626.0,
      "step": 53980
    },
    {
      "entropy": 5.386105298995972,
      "epoch": 4.535433732409158,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003109370048597325,
      "loss": 4.7839,
      "mean_token_accuracy": 0.22718895971775055,
      "num_tokens": 99602058.0,
      "step": 53985
    },
    {
      "entropy": 5.441739511489868,
      "epoch": 4.535853812224323,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000310907442435502,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21877770572900773,
      "num_tokens": 99611071.0,
      "step": 53990
    },
    {
      "entropy": 5.472273254394532,
      "epoch": 4.536273892039487,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031087787937518957,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21501127630472183,
      "num_tokens": 99620550.0,
      "step": 53995
    },
    {
      "entropy": 5.486808204650879,
      "epoch": 4.536693971854652,
      "grad_norm": 1.125,
      "learning_rate": 0.000310848315679319,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2281252384185791,
      "num_tokens": 99628943.0,
      "step": 54000
    },
    {
      "epoch": 4.536693971854652,
      "eval_entropy": 5.2217363815597535,
      "eval_loss": 5.019283294677734,
      "eval_mean_token_accuracy": 0.21826504009951875,
      "eval_num_tokens": 99628943.0,
      "eval_runtime": 27.4899,
      "eval_samples_per_second": 1359.265,
      "eval_steps_per_second": 169.917,
      "step": 54000
    },
    {
      "entropy": 5.471318244934082,
      "epoch": 4.537114051669818,
      "grad_norm": 1.140625,
      "learning_rate": 0.000310818751348414,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2154093340039253,
      "num_tokens": 99638046.0,
      "step": 54005
    },
    {
      "entropy": 5.481273555755616,
      "epoch": 4.537534131484982,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003107891863829983,
      "loss": 5.0298,
      "mean_token_accuracy": 0.2012087270617485,
      "num_tokens": 99648455.0,
      "step": 54010
    },
    {
      "entropy": 5.5072088718414305,
      "epoch": 4.537954211300147,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031075962078359556,
      "loss": 4.8596,
      "mean_token_accuracy": 0.2222466915845871,
      "num_tokens": 99657373.0,
      "step": 54015
    },
    {
      "entropy": 5.536188411712646,
      "epoch": 4.5383742911153115,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003107300545507295,
      "loss": 4.9683,
      "mean_token_accuracy": 0.21598615646362304,
      "num_tokens": 99667180.0,
      "step": 54020
    },
    {
      "entropy": 5.401660442352295,
      "epoch": 4.538794370930477,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031070048768492393,
      "loss": 4.7843,
      "mean_token_accuracy": 0.2241109147667885,
      "num_tokens": 99676329.0,
      "step": 54025
    },
    {
      "entropy": 5.4051487922668455,
      "epoch": 4.539214450745642,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003106709201867027,
      "loss": 4.7503,
      "mean_token_accuracy": 0.23732140213251113,
      "num_tokens": 99685134.0,
      "step": 54030
    },
    {
      "entropy": 5.356300783157349,
      "epoch": 4.539634530560806,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003106413520565894,
      "loss": 4.9189,
      "mean_token_accuracy": 0.21409253925085067,
      "num_tokens": 99695406.0,
      "step": 54035
    },
    {
      "entropy": 5.356435775756836,
      "epoch": 4.540054610375972,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031061178329510784,
      "loss": 4.7744,
      "mean_token_accuracy": 0.2289700984954834,
      "num_tokens": 99704571.0,
      "step": 54040
    },
    {
      "entropy": 5.4368335723876955,
      "epoch": 4.540474690191136,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031058221390278194,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21406402587890624,
      "num_tokens": 99713720.0,
      "step": 54045
    },
    {
      "entropy": 5.6165026187896725,
      "epoch": 4.540894770006301,
      "grad_norm": 1.125,
      "learning_rate": 0.0003105526438801355,
      "loss": 5.062,
      "mean_token_accuracy": 0.20800909996032715,
      "num_tokens": 99722956.0,
      "step": 54050
    },
    {
      "entropy": 5.558380365371704,
      "epoch": 4.541314849821466,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031052307322769224,
      "loss": 4.9632,
      "mean_token_accuracy": 0.20629569739103318,
      "num_tokens": 99731891.0,
      "step": 54055
    },
    {
      "entropy": 5.451576519012451,
      "epoch": 4.541734929636631,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000310493501945976,
      "loss": 4.8497,
      "mean_token_accuracy": 0.2220842033624649,
      "num_tokens": 99740632.0,
      "step": 54060
    },
    {
      "entropy": 5.491143321990966,
      "epoch": 4.542155009451796,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003104639300355107,
      "loss": 5.0499,
      "mean_token_accuracy": 0.1999162182211876,
      "num_tokens": 99749843.0,
      "step": 54065
    },
    {
      "entropy": 5.408870506286621,
      "epoch": 4.542575089266961,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003104343574968202,
      "loss": 4.8522,
      "mean_token_accuracy": 0.222443063557148,
      "num_tokens": 99758801.0,
      "step": 54070
    },
    {
      "entropy": 5.506881427764893,
      "epoch": 4.5429951690821255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003104047843304282,
      "loss": 4.9213,
      "mean_token_accuracy": 0.21733839064836502,
      "num_tokens": 99768177.0,
      "step": 54075
    },
    {
      "entropy": 5.452262210845947,
      "epoch": 4.543415248897291,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031037521053685867,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22179689705371858,
      "num_tokens": 99777885.0,
      "step": 54080
    },
    {
      "entropy": 5.357145166397094,
      "epoch": 4.543835328712455,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031034563611663556,
      "loss": 4.8436,
      "mean_token_accuracy": 0.22470791190862655,
      "num_tokens": 99788536.0,
      "step": 54085
    },
    {
      "entropy": 5.387142515182495,
      "epoch": 4.54425540852762,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031031606107028266,
      "loss": 4.8456,
      "mean_token_accuracy": 0.2232941597700119,
      "num_tokens": 99797861.0,
      "step": 54090
    },
    {
      "entropy": 5.379414749145508,
      "epoch": 4.544675488342785,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031028648539832386,
      "loss": 4.7377,
      "mean_token_accuracy": 0.23285213857889175,
      "num_tokens": 99806974.0,
      "step": 54095
    },
    {
      "entropy": 5.510574626922607,
      "epoch": 4.54509556815795,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031025690910128326,
      "loss": 4.964,
      "mean_token_accuracy": 0.21155446767807007,
      "num_tokens": 99816625.0,
      "step": 54100
    },
    {
      "entropy": 5.489735078811646,
      "epoch": 4.545515647973115,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031022733217968463,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21795652359724044,
      "num_tokens": 99825563.0,
      "step": 54105
    },
    {
      "entropy": 5.474232769012451,
      "epoch": 4.5459357277882795,
      "grad_norm": 1.0625,
      "learning_rate": 0.000310197754634052,
      "loss": 4.9336,
      "mean_token_accuracy": 0.2148524507880211,
      "num_tokens": 99834532.0,
      "step": 54110
    },
    {
      "entropy": 5.433786249160766,
      "epoch": 4.546355807603445,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003101681764649092,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2186695948243141,
      "num_tokens": 99844821.0,
      "step": 54115
    },
    {
      "entropy": 5.408127784729004,
      "epoch": 4.54677588741861,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031013859767278025,
      "loss": 4.821,
      "mean_token_accuracy": 0.22095987647771836,
      "num_tokens": 99853324.0,
      "step": 54120
    },
    {
      "entropy": 5.502726745605469,
      "epoch": 4.547195967233774,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003101090182581892,
      "loss": 4.9417,
      "mean_token_accuracy": 0.21624071300029754,
      "num_tokens": 99863237.0,
      "step": 54125
    },
    {
      "entropy": 5.383314609527588,
      "epoch": 4.5476160470489395,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031007943822165997,
      "loss": 4.7835,
      "mean_token_accuracy": 0.21878862828016282,
      "num_tokens": 99872623.0,
      "step": 54130
    },
    {
      "entropy": 5.539159107208252,
      "epoch": 4.548036126864104,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003100498575637165,
      "loss": 4.9631,
      "mean_token_accuracy": 0.21134042143821716,
      "num_tokens": 99881504.0,
      "step": 54135
    },
    {
      "entropy": 5.534261798858642,
      "epoch": 4.548456206679269,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003100202762848829,
      "loss": 4.9862,
      "mean_token_accuracy": 0.2101243853569031,
      "num_tokens": 99890764.0,
      "step": 54140
    },
    {
      "entropy": 5.461968326568604,
      "epoch": 4.548876286494434,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003099906943856831,
      "loss": 4.9022,
      "mean_token_accuracy": 0.20731922686100007,
      "num_tokens": 99900019.0,
      "step": 54145
    },
    {
      "entropy": 5.476276159286499,
      "epoch": 4.549296366309599,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030996111186664115,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21245089471340178,
      "num_tokens": 99910027.0,
      "step": 54150
    },
    {
      "entropy": 5.422049570083618,
      "epoch": 4.549716446124764,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030993152872828115,
      "loss": 4.8415,
      "mean_token_accuracy": 0.21370947659015654,
      "num_tokens": 99919422.0,
      "step": 54155
    },
    {
      "entropy": 5.417487001419067,
      "epoch": 4.550136525939928,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003099019449711271,
      "loss": 4.8527,
      "mean_token_accuracy": 0.2156989872455597,
      "num_tokens": 99929084.0,
      "step": 54160
    },
    {
      "entropy": 5.483035659790039,
      "epoch": 4.5505566057550935,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003098723605957031,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21619779467582703,
      "num_tokens": 99938272.0,
      "step": 54165
    },
    {
      "entropy": 5.47108416557312,
      "epoch": 4.550976685570259,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030984277560253305,
      "loss": 4.9979,
      "mean_token_accuracy": 0.21197667121887206,
      "num_tokens": 99947604.0,
      "step": 54170
    },
    {
      "entropy": 5.4871275424957275,
      "epoch": 4.551396765385423,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030981318999214125,
      "loss": 4.9604,
      "mean_token_accuracy": 0.22033741772174836,
      "num_tokens": 99955776.0,
      "step": 54175
    },
    {
      "entropy": 5.4923077583312985,
      "epoch": 4.551816845200588,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003097836037650518,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21928345263004304,
      "num_tokens": 99965087.0,
      "step": 54180
    },
    {
      "entropy": 5.389701843261719,
      "epoch": 4.552236925015753,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003097540169217886,
      "loss": 4.8426,
      "mean_token_accuracy": 0.2216482549905777,
      "num_tokens": 99974380.0,
      "step": 54185
    },
    {
      "entropy": 5.471138572692871,
      "epoch": 4.552657004830918,
      "grad_norm": 1.109375,
      "learning_rate": 0.000309724429462876,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21636261343955993,
      "num_tokens": 99983324.0,
      "step": 54190
    },
    {
      "entropy": 5.465923881530761,
      "epoch": 4.553077084646083,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003096948413888379,
      "loss": 4.8862,
      "mean_token_accuracy": 0.20996962189674379,
      "num_tokens": 99991830.0,
      "step": 54195
    },
    {
      "entropy": 5.44795708656311,
      "epoch": 4.553497164461247,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030966525270019864,
      "loss": 4.815,
      "mean_token_accuracy": 0.22860635966062545,
      "num_tokens": 100000728.0,
      "step": 54200
    },
    {
      "entropy": 5.494551515579223,
      "epoch": 4.553917244276413,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030963566339748224,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2159940481185913,
      "num_tokens": 100010262.0,
      "step": 54205
    },
    {
      "entropy": 5.535422515869141,
      "epoch": 4.554337324091577,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003096060734812129,
      "loss": 4.9932,
      "mean_token_accuracy": 0.20765456706285476,
      "num_tokens": 100019861.0,
      "step": 54210
    },
    {
      "entropy": 5.473788499832153,
      "epoch": 4.554757403906742,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003095764829519148,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22132470160722734,
      "num_tokens": 100029417.0,
      "step": 54215
    },
    {
      "entropy": 5.43595175743103,
      "epoch": 4.5551774837219074,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003095468918101121,
      "loss": 4.9445,
      "mean_token_accuracy": 0.21131829917430878,
      "num_tokens": 100039666.0,
      "step": 54220
    },
    {
      "entropy": 5.4640449523925785,
      "epoch": 4.555597563537072,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000309517300056329,
      "loss": 4.8697,
      "mean_token_accuracy": 0.21944156438112258,
      "num_tokens": 100049147.0,
      "step": 54225
    },
    {
      "entropy": 5.438245630264282,
      "epoch": 4.556017643352237,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030948770769108965,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2073511466383934,
      "num_tokens": 100059049.0,
      "step": 54230
    },
    {
      "entropy": 5.485520553588867,
      "epoch": 4.556437723167402,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030945811471491837,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21914099156856537,
      "num_tokens": 100068570.0,
      "step": 54235
    },
    {
      "entropy": 5.3222795009613035,
      "epoch": 4.556857802982567,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003094285211283393,
      "loss": 4.7054,
      "mean_token_accuracy": 0.23149035274982452,
      "num_tokens": 100077083.0,
      "step": 54240
    },
    {
      "entropy": 5.426255607604981,
      "epoch": 4.557277882797732,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003093989269318767,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22315551936626435,
      "num_tokens": 100085972.0,
      "step": 54245
    },
    {
      "entropy": 5.476176118850708,
      "epoch": 4.557697962612896,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003093693321260549,
      "loss": 4.9149,
      "mean_token_accuracy": 0.2196134150028229,
      "num_tokens": 100095352.0,
      "step": 54250
    },
    {
      "entropy": 5.489451599121094,
      "epoch": 4.558118042428061,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030933973671139805,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21209125220775604,
      "num_tokens": 100104518.0,
      "step": 54255
    },
    {
      "entropy": 5.447358655929565,
      "epoch": 4.558538122243226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003093101406884304,
      "loss": 4.7766,
      "mean_token_accuracy": 0.2210245832800865,
      "num_tokens": 100113805.0,
      "step": 54260
    },
    {
      "entropy": 5.371789979934692,
      "epoch": 4.558958202058391,
      "grad_norm": 1.125,
      "learning_rate": 0.0003092805440576763,
      "loss": 4.8738,
      "mean_token_accuracy": 0.21502485275268554,
      "num_tokens": 100123291.0,
      "step": 54265
    },
    {
      "entropy": 5.422336673736572,
      "epoch": 4.559378281873556,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030925094681966006,
      "loss": 4.9321,
      "mean_token_accuracy": 0.21184980124235153,
      "num_tokens": 100132790.0,
      "step": 54270
    },
    {
      "entropy": 5.549043941497803,
      "epoch": 4.5597983616887205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030922134897490587,
      "loss": 5.0142,
      "mean_token_accuracy": 0.21313155442476273,
      "num_tokens": 100142680.0,
      "step": 54275
    },
    {
      "entropy": 5.457962942123413,
      "epoch": 4.560218441503886,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030919175052393814,
      "loss": 4.8667,
      "mean_token_accuracy": 0.21954360455274582,
      "num_tokens": 100152010.0,
      "step": 54280
    },
    {
      "entropy": 5.44225172996521,
      "epoch": 4.560638521319051,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003091621514672812,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2183779686689377,
      "num_tokens": 100161045.0,
      "step": 54285
    },
    {
      "entropy": 5.4798542022705075,
      "epoch": 4.561058601134215,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030913255180545923,
      "loss": 4.9008,
      "mean_token_accuracy": 0.21242113560438156,
      "num_tokens": 100170304.0,
      "step": 54290
    },
    {
      "entropy": 5.546559476852417,
      "epoch": 4.561478680949381,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003091029515389967,
      "loss": 4.9622,
      "mean_token_accuracy": 0.21548850685358048,
      "num_tokens": 100179768.0,
      "step": 54295
    },
    {
      "entropy": 5.529477453231811,
      "epoch": 4.561898760764545,
      "grad_norm": 1.125,
      "learning_rate": 0.00030907335066841796,
      "loss": 4.9299,
      "mean_token_accuracy": 0.2154596522450447,
      "num_tokens": 100189822.0,
      "step": 54300
    },
    {
      "entropy": 5.41951003074646,
      "epoch": 4.56231884057971,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003090437491942474,
      "loss": 4.8581,
      "mean_token_accuracy": 0.21240567564964294,
      "num_tokens": 100199392.0,
      "step": 54305
    },
    {
      "entropy": 5.4980274677276615,
      "epoch": 4.562738920394875,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003090141471170093,
      "loss": 5.0116,
      "mean_token_accuracy": 0.21250563859939575,
      "num_tokens": 100209012.0,
      "step": 54310
    },
    {
      "entropy": 5.486895179748535,
      "epoch": 4.56315900021004,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003089845444372281,
      "loss": 4.8946,
      "mean_token_accuracy": 0.20784790068864822,
      "num_tokens": 100218570.0,
      "step": 54315
    },
    {
      "entropy": 5.476863813400269,
      "epoch": 4.563579080025205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003089549411554283,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22265715003013611,
      "num_tokens": 100228140.0,
      "step": 54320
    },
    {
      "entropy": 5.457001972198486,
      "epoch": 4.563999159840369,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030892533727213415,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21140570044517518,
      "num_tokens": 100237697.0,
      "step": 54325
    },
    {
      "entropy": 5.415836048126221,
      "epoch": 4.5644192396555345,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030889573278787005,
      "loss": 4.9196,
      "mean_token_accuracy": 0.2159065291285515,
      "num_tokens": 100247550.0,
      "step": 54330
    },
    {
      "entropy": 5.473118162155151,
      "epoch": 4.5648393194707,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003088661277031605,
      "loss": 4.9143,
      "mean_token_accuracy": 0.2186002478003502,
      "num_tokens": 100256611.0,
      "step": 54335
    },
    {
      "entropy": 5.481097030639648,
      "epoch": 4.565259399285864,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030883652201853,
      "loss": 4.9212,
      "mean_token_accuracy": 0.21079082041978836,
      "num_tokens": 100266032.0,
      "step": 54340
    },
    {
      "entropy": 5.490805053710938,
      "epoch": 4.565679479101029,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030880691573450294,
      "loss": 4.9195,
      "mean_token_accuracy": 0.2165882483124733,
      "num_tokens": 100274733.0,
      "step": 54345
    },
    {
      "entropy": 5.445133829116822,
      "epoch": 4.566099558916195,
      "grad_norm": 1.0,
      "learning_rate": 0.00030877730885160373,
      "loss": 4.8661,
      "mean_token_accuracy": 0.22421473562717437,
      "num_tokens": 100284551.0,
      "step": 54350
    },
    {
      "entropy": 5.452758359909057,
      "epoch": 4.566519638731359,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003087477013703569,
      "loss": 4.8856,
      "mean_token_accuracy": 0.21668392419815063,
      "num_tokens": 100292978.0,
      "step": 54355
    },
    {
      "entropy": 5.464300680160522,
      "epoch": 4.566939718546524,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003087180932912869,
      "loss": 4.9124,
      "mean_token_accuracy": 0.21565820425748825,
      "num_tokens": 100302125.0,
      "step": 54360
    },
    {
      "entropy": 5.486041164398193,
      "epoch": 4.5673597983616885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030868848461491823,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22611035257577897,
      "num_tokens": 100310551.0,
      "step": 54365
    },
    {
      "entropy": 5.444546222686768,
      "epoch": 4.567779878176854,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003086588753417754,
      "loss": 4.829,
      "mean_token_accuracy": 0.21875597387552262,
      "num_tokens": 100319861.0,
      "step": 54370
    },
    {
      "entropy": 5.413014602661133,
      "epoch": 4.568199957992018,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030862926547238294,
      "loss": 4.8827,
      "mean_token_accuracy": 0.21705504059791564,
      "num_tokens": 100328881.0,
      "step": 54375
    },
    {
      "entropy": 5.394482707977295,
      "epoch": 4.568620037807183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030859965500726544,
      "loss": 4.766,
      "mean_token_accuracy": 0.23114999681711196,
      "num_tokens": 100337757.0,
      "step": 54380
    },
    {
      "entropy": 5.402694034576416,
      "epoch": 4.5690401176223485,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030857004394694725,
      "loss": 4.958,
      "mean_token_accuracy": 0.21230340600013733,
      "num_tokens": 100347426.0,
      "step": 54385
    },
    {
      "entropy": 5.518881559371948,
      "epoch": 4.569460197437513,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003085404322919531,
      "loss": 4.9837,
      "mean_token_accuracy": 0.21032562851905823,
      "num_tokens": 100357518.0,
      "step": 54390
    },
    {
      "entropy": 5.484267997741699,
      "epoch": 4.569880277252678,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003085108200428074,
      "loss": 4.8665,
      "mean_token_accuracy": 0.22159137427806855,
      "num_tokens": 100366077.0,
      "step": 54395
    },
    {
      "entropy": 5.45523362159729,
      "epoch": 4.570300357067843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003084812072000348,
      "loss": 4.878,
      "mean_token_accuracy": 0.2210783541202545,
      "num_tokens": 100375252.0,
      "step": 54400
    },
    {
      "entropy": 5.551794958114624,
      "epoch": 4.570720436883008,
      "grad_norm": 1.125,
      "learning_rate": 0.00030845159376415986,
      "loss": 4.9816,
      "mean_token_accuracy": 0.20783021301031113,
      "num_tokens": 100385199.0,
      "step": 54405
    },
    {
      "entropy": 5.41440896987915,
      "epoch": 4.571140516698173,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030842197973570713,
      "loss": 4.8627,
      "mean_token_accuracy": 0.219854037463665,
      "num_tokens": 100394991.0,
      "step": 54410
    },
    {
      "entropy": 5.4324182033538815,
      "epoch": 4.571560596513337,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003083923651152013,
      "loss": 4.8784,
      "mean_token_accuracy": 0.21975968778133392,
      "num_tokens": 100403933.0,
      "step": 54415
    },
    {
      "entropy": 5.541505193710327,
      "epoch": 4.5719806763285025,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030836274990316687,
      "loss": 4.9212,
      "mean_token_accuracy": 0.21514364033937455,
      "num_tokens": 100413079.0,
      "step": 54420
    },
    {
      "entropy": 5.5662308692932125,
      "epoch": 4.572400756143667,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030833313410012845,
      "loss": 5.1306,
      "mean_token_accuracy": 0.20363088101148605,
      "num_tokens": 100421689.0,
      "step": 54425
    },
    {
      "entropy": 5.468038940429688,
      "epoch": 4.572820835958832,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003083035177066108,
      "loss": 4.8478,
      "mean_token_accuracy": 0.22843518406152724,
      "num_tokens": 100430127.0,
      "step": 54430
    },
    {
      "entropy": 5.5754598617553714,
      "epoch": 4.573240915773997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003082739007231385,
      "loss": 4.9801,
      "mean_token_accuracy": 0.2164631277322769,
      "num_tokens": 100438785.0,
      "step": 54435
    },
    {
      "entropy": 5.497809743881225,
      "epoch": 4.573660995589162,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003082442831502361,
      "loss": 4.9048,
      "mean_token_accuracy": 0.2223639726638794,
      "num_tokens": 100447759.0,
      "step": 54440
    },
    {
      "entropy": 5.465541028976441,
      "epoch": 4.574081075404327,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003082146649884284,
      "loss": 4.889,
      "mean_token_accuracy": 0.2180471047759056,
      "num_tokens": 100457290.0,
      "step": 54445
    },
    {
      "entropy": 5.450444602966309,
      "epoch": 4.574501155219492,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030818504623824,
      "loss": 4.8833,
      "mean_token_accuracy": 0.2211957573890686,
      "num_tokens": 100467192.0,
      "step": 54450
    },
    {
      "entropy": 5.507277584075927,
      "epoch": 4.574921235034656,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003081554269001956,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2187766343355179,
      "num_tokens": 100476672.0,
      "step": 54455
    },
    {
      "entropy": 5.615111303329468,
      "epoch": 4.575341314849822,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003081258069748199,
      "loss": 5.048,
      "mean_token_accuracy": 0.20612335205078125,
      "num_tokens": 100487239.0,
      "step": 54460
    },
    {
      "entropy": 5.477368688583374,
      "epoch": 4.575761394664986,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003080961864626376,
      "loss": 4.8816,
      "mean_token_accuracy": 0.21713481843471527,
      "num_tokens": 100495892.0,
      "step": 54465
    },
    {
      "entropy": 5.380749464035034,
      "epoch": 4.576181474480151,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030806656536417335,
      "loss": 4.8302,
      "mean_token_accuracy": 0.22542784810066224,
      "num_tokens": 100504076.0,
      "step": 54470
    },
    {
      "entropy": 5.5083033561706545,
      "epoch": 4.5766015542953165,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003080369436799519,
      "loss": 4.9723,
      "mean_token_accuracy": 0.21776233166456221,
      "num_tokens": 100513992.0,
      "step": 54475
    },
    {
      "entropy": 5.5132661819458,
      "epoch": 4.577021634110481,
      "grad_norm": 1.125,
      "learning_rate": 0.0003080073214104981,
      "loss": 4.9282,
      "mean_token_accuracy": 0.21429613530635833,
      "num_tokens": 100523088.0,
      "step": 54480
    },
    {
      "entropy": 5.569762086868286,
      "epoch": 4.577441713925646,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030797769855633655,
      "loss": 4.9828,
      "mean_token_accuracy": 0.21054159849882126,
      "num_tokens": 100532025.0,
      "step": 54485
    },
    {
      "entropy": 5.4490406036376955,
      "epoch": 4.57786179374081,
      "grad_norm": 1.25,
      "learning_rate": 0.00030794807511799207,
      "loss": 4.8387,
      "mean_token_accuracy": 0.21415667086839676,
      "num_tokens": 100540115.0,
      "step": 54490
    },
    {
      "entropy": 5.53407850265503,
      "epoch": 4.578281873555976,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003079184510959893,
      "loss": 5.027,
      "mean_token_accuracy": 0.20709847360849382,
      "num_tokens": 100549776.0,
      "step": 54495
    },
    {
      "entropy": 5.442603874206543,
      "epoch": 4.578701953371141,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003078888264908532,
      "loss": 4.8891,
      "mean_token_accuracy": 0.21560741364955902,
      "num_tokens": 100558858.0,
      "step": 54500
    },
    {
      "entropy": 5.487362670898437,
      "epoch": 4.579122033186305,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030785920130310847,
      "loss": 4.8983,
      "mean_token_accuracy": 0.2271055668592453,
      "num_tokens": 100568106.0,
      "step": 54505
    },
    {
      "entropy": 5.602463912963867,
      "epoch": 4.57954211300147,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030782957553327997,
      "loss": 5.0515,
      "mean_token_accuracy": 0.21170361936092377,
      "num_tokens": 100578760.0,
      "step": 54510
    },
    {
      "entropy": 5.470852994918824,
      "epoch": 4.579962192816636,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030779994918189236,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21684931516647338,
      "num_tokens": 100587475.0,
      "step": 54515
    },
    {
      "entropy": 5.451191139221192,
      "epoch": 4.5803822726318,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030777032224947056,
      "loss": 4.8911,
      "mean_token_accuracy": 0.21435958296060562,
      "num_tokens": 100595974.0,
      "step": 54520
    },
    {
      "entropy": 5.521852207183838,
      "epoch": 4.580802352446965,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003077406947365394,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21006111651659012,
      "num_tokens": 100606747.0,
      "step": 54525
    },
    {
      "entropy": 5.474239206314087,
      "epoch": 4.5812224322621296,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003077110666436236,
      "loss": 4.895,
      "mean_token_accuracy": 0.21838807910680771,
      "num_tokens": 100616531.0,
      "step": 54530
    },
    {
      "entropy": 5.459987306594849,
      "epoch": 4.581642512077295,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003076814379712482,
      "loss": 4.9228,
      "mean_token_accuracy": 0.20916089415550232,
      "num_tokens": 100625528.0,
      "step": 54535
    },
    {
      "entropy": 5.379468011856079,
      "epoch": 4.582062591892459,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003076518087199379,
      "loss": 4.7077,
      "mean_token_accuracy": 0.23449985235929488,
      "num_tokens": 100634731.0,
      "step": 54540
    },
    {
      "entropy": 5.446372079849243,
      "epoch": 4.582482671707624,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030762217889021765,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22653906792402267,
      "num_tokens": 100644956.0,
      "step": 54545
    },
    {
      "entropy": 5.456447601318359,
      "epoch": 4.58290275152279,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030759254848261233,
      "loss": 4.857,
      "mean_token_accuracy": 0.2155323475599289,
      "num_tokens": 100654860.0,
      "step": 54550
    },
    {
      "entropy": 5.456289529800415,
      "epoch": 4.583322831337954,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030756291749764675,
      "loss": 4.9168,
      "mean_token_accuracy": 0.21232270300388337,
      "num_tokens": 100664157.0,
      "step": 54555
    },
    {
      "entropy": 5.499687385559082,
      "epoch": 4.583742911153119,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003075332859358459,
      "loss": 4.9633,
      "mean_token_accuracy": 0.210037262737751,
      "num_tokens": 100674133.0,
      "step": 54560
    },
    {
      "entropy": 5.428631258010864,
      "epoch": 4.584162990968284,
      "grad_norm": 1.25,
      "learning_rate": 0.0003075036537977345,
      "loss": 4.8347,
      "mean_token_accuracy": 0.22274686992168427,
      "num_tokens": 100682448.0,
      "step": 54565
    },
    {
      "entropy": 5.42307071685791,
      "epoch": 4.584583070783449,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030747402108383776,
      "loss": 4.8726,
      "mean_token_accuracy": 0.2246316984295845,
      "num_tokens": 100691005.0,
      "step": 54570
    },
    {
      "entropy": 5.489992141723633,
      "epoch": 4.585003150598614,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003074443877946804,
      "loss": 5.0241,
      "mean_token_accuracy": 0.21400773525238037,
      "num_tokens": 100700628.0,
      "step": 54575
    },
    {
      "entropy": 5.467999219894409,
      "epoch": 4.585423230413778,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003074147539307875,
      "loss": 4.8359,
      "mean_token_accuracy": 0.21748716682195662,
      "num_tokens": 100709362.0,
      "step": 54580
    },
    {
      "entropy": 5.491313076019287,
      "epoch": 4.5858433102289435,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030738511949268384,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21492698937654495,
      "num_tokens": 100718645.0,
      "step": 54585
    },
    {
      "entropy": 5.468999004364013,
      "epoch": 4.586263390044109,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030735548448089446,
      "loss": 4.9093,
      "mean_token_accuracy": 0.21172546744346618,
      "num_tokens": 100727064.0,
      "step": 54590
    },
    {
      "entropy": 5.4507135391235355,
      "epoch": 4.586683469859273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030732584889594437,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2242371052503586,
      "num_tokens": 100735964.0,
      "step": 54595
    },
    {
      "entropy": 5.491916084289551,
      "epoch": 4.587103549674438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003072962127383585,
      "loss": 4.8876,
      "mean_token_accuracy": 0.21091639548540114,
      "num_tokens": 100745658.0,
      "step": 54600
    },
    {
      "entropy": 5.503027725219726,
      "epoch": 4.587523629489603,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003072665760086618,
      "loss": 4.9982,
      "mean_token_accuracy": 0.20856858789920807,
      "num_tokens": 100755199.0,
      "step": 54605
    },
    {
      "entropy": 5.464056301116943,
      "epoch": 4.587943709304768,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003072369387073794,
      "loss": 4.8373,
      "mean_token_accuracy": 0.22299748808145523,
      "num_tokens": 100763917.0,
      "step": 54610
    },
    {
      "entropy": 5.454362630844116,
      "epoch": 4.588363789119933,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003072073008350362,
      "loss": 4.8342,
      "mean_token_accuracy": 0.22107626646757125,
      "num_tokens": 100772949.0,
      "step": 54615
    },
    {
      "entropy": 5.4625479221344,
      "epoch": 4.5887838689350975,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030717766239215724,
      "loss": 4.9101,
      "mean_token_accuracy": 0.22146876007318497,
      "num_tokens": 100782671.0,
      "step": 54620
    },
    {
      "entropy": 5.447201299667358,
      "epoch": 4.589203948750263,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030714802337926763,
      "loss": 4.8897,
      "mean_token_accuracy": 0.22136459946632386,
      "num_tokens": 100791758.0,
      "step": 54625
    },
    {
      "entropy": 5.383987188339233,
      "epoch": 4.589624028565427,
      "grad_norm": 1.125,
      "learning_rate": 0.0003071183837968922,
      "loss": 4.8532,
      "mean_token_accuracy": 0.21572301685810089,
      "num_tokens": 100801121.0,
      "step": 54630
    },
    {
      "entropy": 5.46113076210022,
      "epoch": 4.590044108380592,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003070887436455562,
      "loss": 4.9346,
      "mean_token_accuracy": 0.21968727707862853,
      "num_tokens": 100810220.0,
      "step": 54635
    },
    {
      "entropy": 5.494950008392334,
      "epoch": 4.5904641881957575,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030705910292578464,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21077147275209426,
      "num_tokens": 100819835.0,
      "step": 54640
    },
    {
      "entropy": 5.4589190006256105,
      "epoch": 4.590884268010922,
      "grad_norm": 1.125,
      "learning_rate": 0.0003070294616381026,
      "loss": 4.8763,
      "mean_token_accuracy": 0.20980812162160872,
      "num_tokens": 100829316.0,
      "step": 54645
    },
    {
      "entropy": 5.467630815505982,
      "epoch": 4.591304347826087,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030699981978303515,
      "loss": 4.9681,
      "mean_token_accuracy": 0.21658079475164413,
      "num_tokens": 100838255.0,
      "step": 54650
    },
    {
      "entropy": 5.501735496520996,
      "epoch": 4.591724427641251,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030697017736110735,
      "loss": 4.935,
      "mean_token_accuracy": 0.20671902149915694,
      "num_tokens": 100847724.0,
      "step": 54655
    },
    {
      "entropy": 5.486915636062622,
      "epoch": 4.592144507456417,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003069405343728443,
      "loss": 4.8777,
      "mean_token_accuracy": 0.21810507029294968,
      "num_tokens": 100856169.0,
      "step": 54660
    },
    {
      "entropy": 5.4157106399536135,
      "epoch": 4.592564587271582,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030691089081877115,
      "loss": 4.8069,
      "mean_token_accuracy": 0.22809201180934907,
      "num_tokens": 100864837.0,
      "step": 54665
    },
    {
      "entropy": 5.350355529785157,
      "epoch": 4.592984667086746,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000306881246699413,
      "loss": 4.7943,
      "mean_token_accuracy": 0.22141651660203934,
      "num_tokens": 100873231.0,
      "step": 54670
    },
    {
      "entropy": 5.4333607196807865,
      "epoch": 4.5934047469019115,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000306851602015295,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22301054894924163,
      "num_tokens": 100881711.0,
      "step": 54675
    },
    {
      "entropy": 5.426089668273926,
      "epoch": 4.593824826717077,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030682195676694224,
      "loss": 4.8585,
      "mean_token_accuracy": 0.2172284707427025,
      "num_tokens": 100890457.0,
      "step": 54680
    },
    {
      "entropy": 5.3859776020050045,
      "epoch": 4.594244906532241,
      "grad_norm": 1.125,
      "learning_rate": 0.0003067923109548799,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22020359784364701,
      "num_tokens": 100900112.0,
      "step": 54685
    },
    {
      "entropy": 5.445597457885742,
      "epoch": 4.594664986347406,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030676266457963316,
      "loss": 4.8606,
      "mean_token_accuracy": 0.21869742572307588,
      "num_tokens": 100908806.0,
      "step": 54690
    },
    {
      "entropy": 5.4216086864471436,
      "epoch": 4.595085066162571,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003067330176417271,
      "loss": 4.8265,
      "mean_token_accuracy": 0.21967707127332686,
      "num_tokens": 100917957.0,
      "step": 54695
    },
    {
      "entropy": 5.5053431510925295,
      "epoch": 4.595505145977736,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030670337014168707,
      "loss": 4.9883,
      "mean_token_accuracy": 0.21177318543195725,
      "num_tokens": 100927349.0,
      "step": 54700
    },
    {
      "entropy": 5.455557537078858,
      "epoch": 4.5959252257929,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003066737220800381,
      "loss": 4.909,
      "mean_token_accuracy": 0.21630998998880385,
      "num_tokens": 100936829.0,
      "step": 54705
    },
    {
      "entropy": 5.485425853729248,
      "epoch": 4.596345305608065,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003066440734573055,
      "loss": 4.8547,
      "mean_token_accuracy": 0.21635275185108185,
      "num_tokens": 100945936.0,
      "step": 54710
    },
    {
      "entropy": 5.413703441619873,
      "epoch": 4.596765385423231,
      "grad_norm": 1.125,
      "learning_rate": 0.00030661442427401444,
      "loss": 4.915,
      "mean_token_accuracy": 0.2141817942261696,
      "num_tokens": 100954967.0,
      "step": 54715
    },
    {
      "entropy": 5.464714050292969,
      "epoch": 4.597185465238395,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003065847745306901,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22403185665607453,
      "num_tokens": 100964178.0,
      "step": 54720
    },
    {
      "entropy": 5.476903295516967,
      "epoch": 4.59760554505356,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003065551242278578,
      "loss": 4.9176,
      "mean_token_accuracy": 0.21874587684869767,
      "num_tokens": 100974164.0,
      "step": 54725
    },
    {
      "entropy": 5.4528216361999515,
      "epoch": 4.5980256248687255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030652547336604264,
      "loss": 4.9247,
      "mean_token_accuracy": 0.2163985028862953,
      "num_tokens": 100983264.0,
      "step": 54730
    },
    {
      "entropy": 5.364641332626343,
      "epoch": 4.59844570468389,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030649582194576996,
      "loss": 4.8183,
      "mean_token_accuracy": 0.21919133365154267,
      "num_tokens": 100991481.0,
      "step": 54735
    },
    {
      "entropy": 5.4136570453643795,
      "epoch": 4.598865784499055,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003064661699675651,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2112277552485466,
      "num_tokens": 101001143.0,
      "step": 54740
    },
    {
      "entropy": 5.468876552581787,
      "epoch": 4.599285864314219,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030643651743195323,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2138113111257553,
      "num_tokens": 101009792.0,
      "step": 54745
    },
    {
      "entropy": 5.510057353973389,
      "epoch": 4.599705944129385,
      "grad_norm": 1.125,
      "learning_rate": 0.00030640686433945955,
      "loss": 4.923,
      "mean_token_accuracy": 0.21504975408315657,
      "num_tokens": 101019046.0,
      "step": 54750
    },
    {
      "entropy": 5.490234375,
      "epoch": 4.60012602394455,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003063772106906094,
      "loss": 4.9673,
      "mean_token_accuracy": 0.21394299119710922,
      "num_tokens": 101028203.0,
      "step": 54755
    },
    {
      "entropy": 5.511325216293335,
      "epoch": 4.600546103759714,
      "grad_norm": 1.125,
      "learning_rate": 0.0003063475564859282,
      "loss": 4.9406,
      "mean_token_accuracy": 0.2145098328590393,
      "num_tokens": 101037379.0,
      "step": 54760
    },
    {
      "entropy": 5.510268211364746,
      "epoch": 4.600966183574879,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030631790172594127,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21753198206424712,
      "num_tokens": 101046751.0,
      "step": 54765
    },
    {
      "entropy": 5.394745254516602,
      "epoch": 4.601386263390044,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003062882464111737,
      "loss": 4.7761,
      "mean_token_accuracy": 0.23070513755083083,
      "num_tokens": 101055287.0,
      "step": 54770
    },
    {
      "entropy": 5.451254749298096,
      "epoch": 4.601806343205209,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030625859054215105,
      "loss": 4.9921,
      "mean_token_accuracy": 0.20472314953804016,
      "num_tokens": 101064053.0,
      "step": 54775
    },
    {
      "entropy": 5.402079391479492,
      "epoch": 4.602226423020374,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003062289341193985,
      "loss": 4.7743,
      "mean_token_accuracy": 0.23107171356678008,
      "num_tokens": 101072583.0,
      "step": 54780
    },
    {
      "entropy": 5.479077863693237,
      "epoch": 4.6026465028355386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003061992771434415,
      "loss": 4.8914,
      "mean_token_accuracy": 0.22611574232578277,
      "num_tokens": 101081644.0,
      "step": 54785
    },
    {
      "entropy": 5.509877157211304,
      "epoch": 4.603066582650704,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030616961961480533,
      "loss": 4.9724,
      "mean_token_accuracy": 0.2105510875582695,
      "num_tokens": 101090905.0,
      "step": 54790
    },
    {
      "entropy": 5.460488176345825,
      "epoch": 4.603486662465868,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003061399615340154,
      "loss": 4.8776,
      "mean_token_accuracy": 0.21477873474359513,
      "num_tokens": 101100914.0,
      "step": 54795
    },
    {
      "entropy": 5.483600664138794,
      "epoch": 4.603906742281033,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030611030290159715,
      "loss": 4.911,
      "mean_token_accuracy": 0.21211744993925094,
      "num_tokens": 101109988.0,
      "step": 54800
    },
    {
      "entropy": 5.455931520462036,
      "epoch": 4.604326822096199,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030608064371807583,
      "loss": 4.8716,
      "mean_token_accuracy": 0.2171454668045044,
      "num_tokens": 101118906.0,
      "step": 54805
    },
    {
      "entropy": 5.465054845809936,
      "epoch": 4.604746901911363,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003060509839839769,
      "loss": 4.8279,
      "mean_token_accuracy": 0.22091419547796248,
      "num_tokens": 101127461.0,
      "step": 54810
    },
    {
      "entropy": 5.483999013900757,
      "epoch": 4.605166981726528,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003060213236998258,
      "loss": 4.9492,
      "mean_token_accuracy": 0.2150166928768158,
      "num_tokens": 101137264.0,
      "step": 54815
    },
    {
      "entropy": 5.481100034713745,
      "epoch": 4.6055870615416925,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000305991662866148,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21691437512636186,
      "num_tokens": 101145695.0,
      "step": 54820
    },
    {
      "entropy": 5.481760931015015,
      "epoch": 4.606007141356858,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030596200148346874,
      "loss": 4.8959,
      "mean_token_accuracy": 0.2154368758201599,
      "num_tokens": 101154851.0,
      "step": 54825
    },
    {
      "entropy": 5.416166639328003,
      "epoch": 4.606427221172023,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003059323395523137,
      "loss": 4.8571,
      "mean_token_accuracy": 0.2187826916575432,
      "num_tokens": 101163881.0,
      "step": 54830
    },
    {
      "entropy": 5.46351957321167,
      "epoch": 4.606847300987187,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003059026770732081,
      "loss": 4.901,
      "mean_token_accuracy": 0.22097846269607543,
      "num_tokens": 101173930.0,
      "step": 54835
    },
    {
      "entropy": 5.404384899139404,
      "epoch": 4.6072673808023525,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003058730140466775,
      "loss": 4.7879,
      "mean_token_accuracy": 0.22872604578733444,
      "num_tokens": 101182620.0,
      "step": 54840
    },
    {
      "entropy": 5.391747999191284,
      "epoch": 4.607687460617518,
      "grad_norm": 0.984375,
      "learning_rate": 0.00030584335047324735,
      "loss": 4.8026,
      "mean_token_accuracy": 0.2213890701532364,
      "num_tokens": 101192788.0,
      "step": 54845
    },
    {
      "entropy": 5.432500219345092,
      "epoch": 4.608107540432682,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003058136863534431,
      "loss": 4.8985,
      "mean_token_accuracy": 0.2165445491671562,
      "num_tokens": 101202765.0,
      "step": 54850
    },
    {
      "entropy": 5.515111589431763,
      "epoch": 4.608527620247847,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003057840216877903,
      "loss": 4.9703,
      "mean_token_accuracy": 0.2161567375063896,
      "num_tokens": 101211881.0,
      "step": 54855
    },
    {
      "entropy": 5.508906602859497,
      "epoch": 4.608947700063012,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003057543564768144,
      "loss": 4.8918,
      "mean_token_accuracy": 0.21965236514806746,
      "num_tokens": 101220598.0,
      "step": 54860
    },
    {
      "entropy": 5.430937480926514,
      "epoch": 4.609367779878177,
      "grad_norm": 1.125,
      "learning_rate": 0.00030572469072104097,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22257996946573258,
      "num_tokens": 101229287.0,
      "step": 54865
    },
    {
      "entropy": 5.4407305240631105,
      "epoch": 4.609787859693341,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003056950244209953,
      "loss": 4.8921,
      "mean_token_accuracy": 0.21696760803461074,
      "num_tokens": 101239374.0,
      "step": 54870
    },
    {
      "entropy": 5.467802619934082,
      "epoch": 4.6102079395085065,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003056653575772032,
      "loss": 4.8952,
      "mean_token_accuracy": 0.21702210903167723,
      "num_tokens": 101248662.0,
      "step": 54875
    },
    {
      "entropy": 5.4193178653717045,
      "epoch": 4.610628019323672,
      "grad_norm": 1.125,
      "learning_rate": 0.0003056356901901901,
      "loss": 4.8839,
      "mean_token_accuracy": 0.21820851415395737,
      "num_tokens": 101257671.0,
      "step": 54880
    },
    {
      "entropy": 5.410844516754151,
      "epoch": 4.611048099138836,
      "grad_norm": 1.125,
      "learning_rate": 0.0003056060222604815,
      "loss": 4.8801,
      "mean_token_accuracy": 0.216961333155632,
      "num_tokens": 101266890.0,
      "step": 54885
    },
    {
      "entropy": 5.470428562164306,
      "epoch": 4.611468178954001,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030557635378860296,
      "loss": 4.9434,
      "mean_token_accuracy": 0.2164037197828293,
      "num_tokens": 101276806.0,
      "step": 54890
    },
    {
      "entropy": 5.432820081710815,
      "epoch": 4.6118882587691665,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003055466847750801,
      "loss": 4.9119,
      "mean_token_accuracy": 0.21073632091283798,
      "num_tokens": 101285507.0,
      "step": 54895
    },
    {
      "entropy": 5.512731981277466,
      "epoch": 4.612308338584331,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003055170152204384,
      "loss": 4.8696,
      "mean_token_accuracy": 0.21979828774929047,
      "num_tokens": 101294485.0,
      "step": 54900
    },
    {
      "entropy": 5.4986066818237305,
      "epoch": 4.612728418399496,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003054873451252035,
      "loss": 5.0076,
      "mean_token_accuracy": 0.21216352730989457,
      "num_tokens": 101304626.0,
      "step": 54905
    },
    {
      "entropy": 5.521023178100586,
      "epoch": 4.61314849821466,
      "grad_norm": 1.125,
      "learning_rate": 0.00030545767448990097,
      "loss": 4.9623,
      "mean_token_accuracy": 0.21036500334739686,
      "num_tokens": 101313731.0,
      "step": 54910
    },
    {
      "entropy": 5.385000228881836,
      "epoch": 4.613568578029826,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030542800331505646,
      "loss": 4.8024,
      "mean_token_accuracy": 0.22317394018173217,
      "num_tokens": 101323448.0,
      "step": 54915
    },
    {
      "entropy": 5.429712963104248,
      "epoch": 4.613988657844991,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003053983316011955,
      "loss": 4.8304,
      "mean_token_accuracy": 0.21794428676366806,
      "num_tokens": 101332840.0,
      "step": 54920
    },
    {
      "entropy": 5.434830045700073,
      "epoch": 4.614408737660155,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030536865934884373,
      "loss": 4.7834,
      "mean_token_accuracy": 0.2264458566904068,
      "num_tokens": 101342335.0,
      "step": 54925
    },
    {
      "entropy": 5.505319738388062,
      "epoch": 4.6148288174753205,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030533898655852676,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2129766672849655,
      "num_tokens": 101351237.0,
      "step": 54930
    },
    {
      "entropy": 5.4392005443573,
      "epoch": 4.615248897290485,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003053093132307704,
      "loss": 4.8872,
      "mean_token_accuracy": 0.21930857598781586,
      "num_tokens": 101360424.0,
      "step": 54935
    },
    {
      "entropy": 5.4484504699707035,
      "epoch": 4.61566897710565,
      "grad_norm": 1.125,
      "learning_rate": 0.0003052796393661001,
      "loss": 4.915,
      "mean_token_accuracy": 0.21300792098045349,
      "num_tokens": 101369622.0,
      "step": 54940
    },
    {
      "entropy": 5.432931518554687,
      "epoch": 4.616089056920815,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003052499649650415,
      "loss": 4.8657,
      "mean_token_accuracy": 0.2201973393559456,
      "num_tokens": 101378584.0,
      "step": 54945
    },
    {
      "entropy": 5.471130609512329,
      "epoch": 4.61650913673598,
      "grad_norm": 1.125,
      "learning_rate": 0.0003052202900281205,
      "loss": 4.8007,
      "mean_token_accuracy": 0.22595966458320618,
      "num_tokens": 101387621.0,
      "step": 54950
    },
    {
      "entropy": 5.490075159072876,
      "epoch": 4.616929216551145,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003051906145558626,
      "loss": 4.9034,
      "mean_token_accuracy": 0.21955860853195192,
      "num_tokens": 101396733.0,
      "step": 54955
    },
    {
      "entropy": 5.429061651229858,
      "epoch": 4.61734929636631,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003051609385487934,
      "loss": 4.9149,
      "mean_token_accuracy": 0.21115428656339646,
      "num_tokens": 101405778.0,
      "step": 54960
    },
    {
      "entropy": 5.451109600067139,
      "epoch": 4.617769376181474,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030513126200743874,
      "loss": 4.9536,
      "mean_token_accuracy": 0.2107529118657112,
      "num_tokens": 101415384.0,
      "step": 54965
    },
    {
      "entropy": 5.507460641860962,
      "epoch": 4.61818945599664,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003051015849323243,
      "loss": 4.9925,
      "mean_token_accuracy": 0.2116699069738388,
      "num_tokens": 101424860.0,
      "step": 54970
    },
    {
      "entropy": 5.5630778789520265,
      "epoch": 4.618609535811804,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030507190732397585,
      "loss": 4.9566,
      "mean_token_accuracy": 0.21402603685855864,
      "num_tokens": 101434139.0,
      "step": 54975
    },
    {
      "entropy": 5.469910192489624,
      "epoch": 4.619029615626969,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030504222918291895,
      "loss": 4.8719,
      "mean_token_accuracy": 0.21559948176145555,
      "num_tokens": 101443760.0,
      "step": 54980
    },
    {
      "entropy": 5.367072725296021,
      "epoch": 4.619449695442134,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030501255050967947,
      "loss": 4.8337,
      "mean_token_accuracy": 0.21555509865283967,
      "num_tokens": 101452823.0,
      "step": 54985
    },
    {
      "entropy": 5.409564542770386,
      "epoch": 4.619869775257299,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000304982871304783,
      "loss": 4.8178,
      "mean_token_accuracy": 0.22116781920194625,
      "num_tokens": 101461031.0,
      "step": 54990
    },
    {
      "entropy": 5.481792402267456,
      "epoch": 4.620289855072464,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030495319156875555,
      "loss": 5.0,
      "mean_token_accuracy": 0.21146105378866195,
      "num_tokens": 101471602.0,
      "step": 54995
    },
    {
      "entropy": 5.48417820930481,
      "epoch": 4.620709934887628,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030492351130212273,
      "loss": 4.9019,
      "mean_token_accuracy": 0.2136295408010483,
      "num_tokens": 101480324.0,
      "step": 55000
    },
    {
      "entropy": 5.488829898834228,
      "epoch": 4.621130014702794,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003048938305054103,
      "loss": 4.9621,
      "mean_token_accuracy": 0.2121818408370018,
      "num_tokens": 101490121.0,
      "step": 55005
    },
    {
      "entropy": 5.475684928894043,
      "epoch": 4.621550094517959,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003048641491791441,
      "loss": 4.869,
      "mean_token_accuracy": 0.2142273247241974,
      "num_tokens": 101498596.0,
      "step": 55010
    },
    {
      "entropy": 5.484486770629883,
      "epoch": 4.621970174333123,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030483446732384973,
      "loss": 4.9308,
      "mean_token_accuracy": 0.21342191696166993,
      "num_tokens": 101507980.0,
      "step": 55015
    },
    {
      "entropy": 5.4942522048950195,
      "epoch": 4.622390254148288,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003048047849400532,
      "loss": 5.0133,
      "mean_token_accuracy": 0.20708169937133789,
      "num_tokens": 101517072.0,
      "step": 55020
    },
    {
      "entropy": 5.491264772415161,
      "epoch": 4.622810333963453,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030477510202828027,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21548108607530594,
      "num_tokens": 101526204.0,
      "step": 55025
    },
    {
      "entropy": 5.451746988296509,
      "epoch": 4.623230413778618,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030474541858905674,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21264532953500748,
      "num_tokens": 101535304.0,
      "step": 55030
    },
    {
      "entropy": 5.453948163986206,
      "epoch": 4.623650493593782,
      "grad_norm": 1.125,
      "learning_rate": 0.00030471573462290846,
      "loss": 4.888,
      "mean_token_accuracy": 0.22600118964910507,
      "num_tokens": 101545588.0,
      "step": 55035
    },
    {
      "entropy": 5.506641292572022,
      "epoch": 4.624070573408948,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003046860501303612,
      "loss": 4.9013,
      "mean_token_accuracy": 0.21749991178512573,
      "num_tokens": 101554643.0,
      "step": 55040
    },
    {
      "entropy": 5.449556732177735,
      "epoch": 4.624490653224113,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003046563651119409,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2206815481185913,
      "num_tokens": 101564118.0,
      "step": 55045
    },
    {
      "entropy": 5.387221097946167,
      "epoch": 4.624910733039277,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003046266795681733,
      "loss": 4.8108,
      "mean_token_accuracy": 0.22940213084220887,
      "num_tokens": 101573060.0,
      "step": 55050
    },
    {
      "entropy": 5.514853048324585,
      "epoch": 4.625330812854442,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003045969934995844,
      "loss": 4.9772,
      "mean_token_accuracy": 0.21399240791797638,
      "num_tokens": 101582362.0,
      "step": 55055
    },
    {
      "entropy": 5.484149360656739,
      "epoch": 4.625750892669608,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003045673069066999,
      "loss": 4.9304,
      "mean_token_accuracy": 0.21356843113899232,
      "num_tokens": 101590808.0,
      "step": 55060
    },
    {
      "entropy": 5.385883092880249,
      "epoch": 4.626170972484772,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030453761979004585,
      "loss": 4.7567,
      "mean_token_accuracy": 0.22592532336711885,
      "num_tokens": 101600318.0,
      "step": 55065
    },
    {
      "entropy": 5.507168531417847,
      "epoch": 4.626591052299937,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030450793215014806,
      "loss": 5.0166,
      "mean_token_accuracy": 0.21143014430999757,
      "num_tokens": 101610156.0,
      "step": 55070
    },
    {
      "entropy": 5.451091575622558,
      "epoch": 4.6270111321151015,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030447824398753244,
      "loss": 4.8091,
      "mean_token_accuracy": 0.22014387995004653,
      "num_tokens": 101619131.0,
      "step": 55075
    },
    {
      "entropy": 5.418769073486328,
      "epoch": 4.627431211930267,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030444855530272494,
      "loss": 4.8282,
      "mean_token_accuracy": 0.2198289766907692,
      "num_tokens": 101629070.0,
      "step": 55080
    },
    {
      "entropy": 5.4417640209198,
      "epoch": 4.627851291745432,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003044188660962514,
      "loss": 4.8633,
      "mean_token_accuracy": 0.2184139996767044,
      "num_tokens": 101637899.0,
      "step": 55085
    },
    {
      "entropy": 5.392961120605468,
      "epoch": 4.628271371560596,
      "grad_norm": 1.125,
      "learning_rate": 0.0003043891763686378,
      "loss": 4.8189,
      "mean_token_accuracy": 0.21882925927639008,
      "num_tokens": 101646533.0,
      "step": 55090
    },
    {
      "entropy": 5.40579571723938,
      "epoch": 4.6286914513757615,
      "grad_norm": 1.125,
      "learning_rate": 0.0003043594861204101,
      "loss": 4.8713,
      "mean_token_accuracy": 0.21695185750722884,
      "num_tokens": 101656002.0,
      "step": 55095
    },
    {
      "entropy": 5.491503047943115,
      "epoch": 4.629111531190926,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003043297953520942,
      "loss": 4.8974,
      "mean_token_accuracy": 0.2193244218826294,
      "num_tokens": 101665285.0,
      "step": 55100
    },
    {
      "entropy": 5.50735387802124,
      "epoch": 4.629531611006091,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030430010406421614,
      "loss": 4.8604,
      "mean_token_accuracy": 0.2163103386759758,
      "num_tokens": 101673624.0,
      "step": 55105
    },
    {
      "entropy": 5.39127893447876,
      "epoch": 4.629951690821256,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030427041225730174,
      "loss": 4.8827,
      "mean_token_accuracy": 0.21992396265268327,
      "num_tokens": 101682485.0,
      "step": 55110
    },
    {
      "entropy": 5.385276556015015,
      "epoch": 4.630371770636421,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003042407199318771,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22020123451948165,
      "num_tokens": 101691894.0,
      "step": 55115
    },
    {
      "entropy": 5.481391477584839,
      "epoch": 4.630791850451586,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030421102708846817,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21301158368587494,
      "num_tokens": 101700518.0,
      "step": 55120
    },
    {
      "entropy": 5.468016529083252,
      "epoch": 4.631211930266751,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003041813337276009,
      "loss": 4.8822,
      "mean_token_accuracy": 0.23466607183218002,
      "num_tokens": 101710400.0,
      "step": 55125
    },
    {
      "entropy": 5.411021947860718,
      "epoch": 4.6316320100819155,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030415163984980135,
      "loss": 4.8066,
      "mean_token_accuracy": 0.22217070907354355,
      "num_tokens": 101718784.0,
      "step": 55130
    },
    {
      "entropy": 5.438217496871948,
      "epoch": 4.632052089897081,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003041219454555955,
      "loss": 4.9044,
      "mean_token_accuracy": 0.2203995555639267,
      "num_tokens": 101728005.0,
      "step": 55135
    },
    {
      "entropy": 5.383721828460693,
      "epoch": 4.632472169712245,
      "grad_norm": 1.125,
      "learning_rate": 0.0003040922505455094,
      "loss": 4.8026,
      "mean_token_accuracy": 0.2229919910430908,
      "num_tokens": 101737634.0,
      "step": 55140
    },
    {
      "entropy": 5.49257378578186,
      "epoch": 4.63289224952741,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030406255512006907,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20975909531116485,
      "num_tokens": 101747062.0,
      "step": 55145
    },
    {
      "entropy": 5.4704536437988285,
      "epoch": 4.633312329342575,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003040328591798006,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21785276830196382,
      "num_tokens": 101757138.0,
      "step": 55150
    },
    {
      "entropy": 5.457318639755249,
      "epoch": 4.63373240915774,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003040031627252299,
      "loss": 4.7984,
      "mean_token_accuracy": 0.22335630655288696,
      "num_tokens": 101764865.0,
      "step": 55155
    },
    {
      "entropy": 5.443898153305054,
      "epoch": 4.634152488972905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030397346575688306,
      "loss": 4.8848,
      "mean_token_accuracy": 0.2149139955639839,
      "num_tokens": 101773967.0,
      "step": 55160
    },
    {
      "entropy": 5.407776355743408,
      "epoch": 4.6345725687880694,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030394376827528626,
      "loss": 4.8945,
      "mean_token_accuracy": 0.21868054568767548,
      "num_tokens": 101782808.0,
      "step": 55165
    },
    {
      "entropy": 5.4707932472229,
      "epoch": 4.634992648603235,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030391407028096547,
      "loss": 4.9292,
      "mean_token_accuracy": 0.2095673516392708,
      "num_tokens": 101792158.0,
      "step": 55170
    },
    {
      "entropy": 5.473042678833008,
      "epoch": 4.6354127284184,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003038843717744468,
      "loss": 4.9104,
      "mean_token_accuracy": 0.21402146816253662,
      "num_tokens": 101802094.0,
      "step": 55175
    },
    {
      "entropy": 5.419084453582764,
      "epoch": 4.635832808233564,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030385467275625646,
      "loss": 4.8504,
      "mean_token_accuracy": 0.21798128336668016,
      "num_tokens": 101811522.0,
      "step": 55180
    },
    {
      "entropy": 5.491227722167968,
      "epoch": 4.6362528880487295,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030382497322692034,
      "loss": 4.921,
      "mean_token_accuracy": 0.2124365448951721,
      "num_tokens": 101821420.0,
      "step": 55185
    },
    {
      "entropy": 5.4712158203125,
      "epoch": 4.636672967863894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030379527318696474,
      "loss": 4.997,
      "mean_token_accuracy": 0.21217692345380784,
      "num_tokens": 101830925.0,
      "step": 55190
    },
    {
      "entropy": 5.425571155548096,
      "epoch": 4.637093047679059,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030376557263691564,
      "loss": 4.8534,
      "mean_token_accuracy": 0.2258509650826454,
      "num_tokens": 101840240.0,
      "step": 55195
    },
    {
      "entropy": 5.418847751617432,
      "epoch": 4.637513127494224,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003037358715772993,
      "loss": 4.8602,
      "mean_token_accuracy": 0.2172538533806801,
      "num_tokens": 101850495.0,
      "step": 55200
    },
    {
      "entropy": 5.442138910293579,
      "epoch": 4.637933207309389,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003037061700086417,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22542236149311065,
      "num_tokens": 101859592.0,
      "step": 55205
    },
    {
      "entropy": 5.35948600769043,
      "epoch": 4.638353287124554,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003036764679314692,
      "loss": 4.8192,
      "mean_token_accuracy": 0.22629837244749068,
      "num_tokens": 101868611.0,
      "step": 55210
    },
    {
      "entropy": 5.647801351547241,
      "epoch": 4.638773366939718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030364676534630773,
      "loss": 5.1412,
      "mean_token_accuracy": 0.20309196412563324,
      "num_tokens": 101878141.0,
      "step": 55215
    },
    {
      "entropy": 5.496062803268432,
      "epoch": 4.639193446754883,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003036170622536836,
      "loss": 4.8646,
      "mean_token_accuracy": 0.21899560689926148,
      "num_tokens": 101887545.0,
      "step": 55220
    },
    {
      "entropy": 5.368777942657471,
      "epoch": 4.639613526570049,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030358735865412297,
      "loss": 4.8362,
      "mean_token_accuracy": 0.21639158427715302,
      "num_tokens": 101896073.0,
      "step": 55225
    },
    {
      "entropy": 5.424513339996338,
      "epoch": 4.640033606385213,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030355765454815204,
      "loss": 4.9075,
      "mean_token_accuracy": 0.22046383321285248,
      "num_tokens": 101905500.0,
      "step": 55230
    },
    {
      "entropy": 5.526131629943848,
      "epoch": 4.640453686200378,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030352794993629683,
      "loss": 4.876,
      "mean_token_accuracy": 0.21417345255613326,
      "num_tokens": 101914874.0,
      "step": 55235
    },
    {
      "entropy": 5.462632894515991,
      "epoch": 4.640873766015543,
      "grad_norm": 1.125,
      "learning_rate": 0.00030349824481908376,
      "loss": 4.8769,
      "mean_token_accuracy": 0.21547288298606873,
      "num_tokens": 101924819.0,
      "step": 55240
    },
    {
      "entropy": 5.545235729217529,
      "epoch": 4.641293845830708,
      "grad_norm": 1.09375,
      "learning_rate": 0.000303468539197039,
      "loss": 4.9237,
      "mean_token_accuracy": 0.22079810947179795,
      "num_tokens": 101934042.0,
      "step": 55245
    },
    {
      "entropy": 5.421803140640259,
      "epoch": 4.641713925645873,
      "grad_norm": 1.0,
      "learning_rate": 0.00030343883307068866,
      "loss": 4.7795,
      "mean_token_accuracy": 0.2269093096256256,
      "num_tokens": 101943438.0,
      "step": 55250
    },
    {
      "entropy": 5.371182155609131,
      "epoch": 4.642134005461037,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000303409126440559,
      "loss": 4.7865,
      "mean_token_accuracy": 0.22897724360227584,
      "num_tokens": 101952686.0,
      "step": 55255
    },
    {
      "entropy": 5.445937395095825,
      "epoch": 4.642554085276203,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003033794193071764,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2194256976246834,
      "num_tokens": 101961119.0,
      "step": 55260
    },
    {
      "entropy": 5.389370346069336,
      "epoch": 4.642974165091367,
      "grad_norm": 1.125,
      "learning_rate": 0.000303349711671067,
      "loss": 4.8263,
      "mean_token_accuracy": 0.22189923077821733,
      "num_tokens": 101970509.0,
      "step": 55265
    },
    {
      "entropy": 5.422046899795532,
      "epoch": 4.643394244906532,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000303320003532757,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21795699000358582,
      "num_tokens": 101979366.0,
      "step": 55270
    },
    {
      "entropy": 5.414199304580689,
      "epoch": 4.643814324721697,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003032902948927727,
      "loss": 4.778,
      "mean_token_accuracy": 0.23002485781908036,
      "num_tokens": 101988216.0,
      "step": 55275
    },
    {
      "entropy": 5.455317258834839,
      "epoch": 4.644234404536862,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030326058575164046,
      "loss": 4.8636,
      "mean_token_accuracy": 0.22025780826807023,
      "num_tokens": 101998284.0,
      "step": 55280
    },
    {
      "entropy": 5.4990379333496096,
      "epoch": 4.644654484352027,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030323087610988644,
      "loss": 4.9356,
      "mean_token_accuracy": 0.21460287719964982,
      "num_tokens": 102007724.0,
      "step": 55285
    },
    {
      "entropy": 5.36706132888794,
      "epoch": 4.645074564167192,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030320116596803696,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2211686983704567,
      "num_tokens": 102016982.0,
      "step": 55290
    },
    {
      "entropy": 5.498921060562134,
      "epoch": 4.645494643982357,
      "grad_norm": 1.125,
      "learning_rate": 0.00030317145532661845,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21096959114074706,
      "num_tokens": 102026103.0,
      "step": 55295
    },
    {
      "entropy": 5.552407121658325,
      "epoch": 4.645914723797522,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000303141744186157,
      "loss": 5.0019,
      "mean_token_accuracy": 0.208813938498497,
      "num_tokens": 102034517.0,
      "step": 55300
    },
    {
      "entropy": 5.474820613861084,
      "epoch": 4.646334803612686,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003031120325471791,
      "loss": 4.8816,
      "mean_token_accuracy": 0.21437585204839707,
      "num_tokens": 102043135.0,
      "step": 55305
    },
    {
      "entropy": 5.415582513809204,
      "epoch": 4.646754883427851,
      "grad_norm": 1.171875,
      "learning_rate": 0.000303082320410211,
      "loss": 4.8885,
      "mean_token_accuracy": 0.2170054167509079,
      "num_tokens": 102052497.0,
      "step": 55310
    },
    {
      "entropy": 5.395287561416626,
      "epoch": 4.647174963243016,
      "grad_norm": 1.125,
      "learning_rate": 0.0003030526077757791,
      "loss": 4.8451,
      "mean_token_accuracy": 0.21758719384670258,
      "num_tokens": 102062151.0,
      "step": 55315
    },
    {
      "entropy": 5.46266884803772,
      "epoch": 4.647595043058181,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003030228946444096,
      "loss": 4.8491,
      "mean_token_accuracy": 0.21776579469442367,
      "num_tokens": 102071460.0,
      "step": 55320
    },
    {
      "entropy": 5.449364900588989,
      "epoch": 4.648015122873346,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003029931810166291,
      "loss": 4.9005,
      "mean_token_accuracy": 0.2202997773885727,
      "num_tokens": 102081574.0,
      "step": 55325
    },
    {
      "entropy": 5.508600378036499,
      "epoch": 4.6484352026885105,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030296346689296364,
      "loss": 4.9258,
      "mean_token_accuracy": 0.2149333253502846,
      "num_tokens": 102090146.0,
      "step": 55330
    },
    {
      "entropy": 5.499449300765991,
      "epoch": 4.648855282503676,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003029337522739399,
      "loss": 4.9252,
      "mean_token_accuracy": 0.20995070934295654,
      "num_tokens": 102100188.0,
      "step": 55335
    },
    {
      "entropy": 5.392921733856201,
      "epoch": 4.649275362318841,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003029040371600841,
      "loss": 4.8295,
      "mean_token_accuracy": 0.22310174852609635,
      "num_tokens": 102108987.0,
      "step": 55340
    },
    {
      "entropy": 5.43861870765686,
      "epoch": 4.649695442134005,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003028743215519226,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22833293825387954,
      "num_tokens": 102117310.0,
      "step": 55345
    },
    {
      "entropy": 5.387954187393189,
      "epoch": 4.6501155219491705,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030284460544998195,
      "loss": 4.9145,
      "mean_token_accuracy": 0.2167169764637947,
      "num_tokens": 102126132.0,
      "step": 55350
    },
    {
      "entropy": 5.456964206695557,
      "epoch": 4.650535601764335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003028148888547884,
      "loss": 4.8609,
      "mean_token_accuracy": 0.2240661174058914,
      "num_tokens": 102135337.0,
      "step": 55355
    },
    {
      "entropy": 5.466711139678955,
      "epoch": 4.6509556815795,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030278517176686833,
      "loss": 4.9492,
      "mean_token_accuracy": 0.20990706235170364,
      "num_tokens": 102144617.0,
      "step": 55360
    },
    {
      "entropy": 5.505498313903809,
      "epoch": 4.651375761394665,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030275545418674825,
      "loss": 4.9677,
      "mean_token_accuracy": 0.2068073645234108,
      "num_tokens": 102154409.0,
      "step": 55365
    },
    {
      "entropy": 5.480707216262817,
      "epoch": 4.65179584120983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030272573611495474,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2119554579257965,
      "num_tokens": 102162905.0,
      "step": 55370
    },
    {
      "entropy": 5.429377937316895,
      "epoch": 4.652215921024995,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030269601755201396,
      "loss": 4.8036,
      "mean_token_accuracy": 0.22129036635160446,
      "num_tokens": 102172108.0,
      "step": 55375
    },
    {
      "entropy": 5.474956464767456,
      "epoch": 4.652636000840159,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003026662984984525,
      "loss": 4.999,
      "mean_token_accuracy": 0.2117079183459282,
      "num_tokens": 102182580.0,
      "step": 55380
    },
    {
      "entropy": 5.400780725479126,
      "epoch": 4.6530560806553245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030263657895479687,
      "loss": 4.7814,
      "mean_token_accuracy": 0.21775009781122207,
      "num_tokens": 102191291.0,
      "step": 55385
    },
    {
      "entropy": 5.473122930526733,
      "epoch": 4.65347616047049,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030260685892157343,
      "loss": 4.859,
      "mean_token_accuracy": 0.21786966621875764,
      "num_tokens": 102200149.0,
      "step": 55390
    },
    {
      "entropy": 5.412872982025147,
      "epoch": 4.653896240285654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003025771383993087,
      "loss": 4.8337,
      "mean_token_accuracy": 0.21806880682706833,
      "num_tokens": 102209822.0,
      "step": 55395
    },
    {
      "entropy": 5.495566034317017,
      "epoch": 4.654316320100819,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003025474173885292,
      "loss": 5.0006,
      "mean_token_accuracy": 0.21235795468091964,
      "num_tokens": 102219138.0,
      "step": 55400
    },
    {
      "entropy": 5.414843273162842,
      "epoch": 4.6547363999159845,
      "grad_norm": 1.125,
      "learning_rate": 0.0003025176958897614,
      "loss": 4.9011,
      "mean_token_accuracy": 0.21819268465042113,
      "num_tokens": 102228845.0,
      "step": 55405
    },
    {
      "entropy": 5.473319387435913,
      "epoch": 4.655156479731149,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003024879739035317,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22120991349220276,
      "num_tokens": 102237407.0,
      "step": 55410
    },
    {
      "entropy": 5.606364154815674,
      "epoch": 4.655576559546314,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003024582514303668,
      "loss": 5.1553,
      "mean_token_accuracy": 0.20285670906305314,
      "num_tokens": 102247518.0,
      "step": 55415
    },
    {
      "entropy": 5.443685674667359,
      "epoch": 4.6559966393614785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030242852847079304,
      "loss": 4.8592,
      "mean_token_accuracy": 0.21776944249868393,
      "num_tokens": 102256599.0,
      "step": 55420
    },
    {
      "entropy": 5.456258106231689,
      "epoch": 4.656416719176644,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030239880502533703,
      "loss": 4.9367,
      "mean_token_accuracy": 0.21166216284036637,
      "num_tokens": 102265792.0,
      "step": 55425
    },
    {
      "entropy": 5.520225620269775,
      "epoch": 4.656836798991808,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030236908109452533,
      "loss": 4.9291,
      "mean_token_accuracy": 0.217570161819458,
      "num_tokens": 102274550.0,
      "step": 55430
    },
    {
      "entropy": 5.439754152297974,
      "epoch": 4.657256878806973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003023393566788844,
      "loss": 4.8683,
      "mean_token_accuracy": 0.21242049485445022,
      "num_tokens": 102283358.0,
      "step": 55435
    },
    {
      "entropy": 5.430142974853515,
      "epoch": 4.6576769586221385,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030230963177894097,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22570035606622696,
      "num_tokens": 102292271.0,
      "step": 55440
    },
    {
      "entropy": 5.473448085784912,
      "epoch": 4.658097038437303,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030227990639522127,
      "loss": 4.9074,
      "mean_token_accuracy": 0.2137601777911186,
      "num_tokens": 102301731.0,
      "step": 55445
    },
    {
      "entropy": 5.4246901512146,
      "epoch": 4.658517118252468,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003022501805282522,
      "loss": 4.9017,
      "mean_token_accuracy": 0.21957540810108184,
      "num_tokens": 102311055.0,
      "step": 55450
    },
    {
      "entropy": 5.455975580215454,
      "epoch": 4.658937198067633,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030222045417856023,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2109724000096321,
      "num_tokens": 102320002.0,
      "step": 55455
    },
    {
      "entropy": 5.572359943389893,
      "epoch": 4.659357277882798,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003021907273466719,
      "loss": 4.9924,
      "mean_token_accuracy": 0.21250552535057068,
      "num_tokens": 102330155.0,
      "step": 55460
    },
    {
      "entropy": 5.3530045509338375,
      "epoch": 4.659777357697963,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003021610000331139,
      "loss": 4.7956,
      "mean_token_accuracy": 0.22608762830495835,
      "num_tokens": 102339166.0,
      "step": 55465
    },
    {
      "entropy": 5.482978439331054,
      "epoch": 4.660197437513127,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003021312722384126,
      "loss": 4.9809,
      "mean_token_accuracy": 0.20944205224514006,
      "num_tokens": 102349015.0,
      "step": 55470
    },
    {
      "entropy": 5.549770355224609,
      "epoch": 4.660617517328292,
      "grad_norm": 1.09375,
      "learning_rate": 0.000302101543963095,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21047901809215547,
      "num_tokens": 102358426.0,
      "step": 55475
    },
    {
      "entropy": 5.437986612319946,
      "epoch": 4.661037597143457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003020718152076873,
      "loss": 4.8166,
      "mean_token_accuracy": 0.22722112387418747,
      "num_tokens": 102367759.0,
      "step": 55480
    },
    {
      "entropy": 5.27188458442688,
      "epoch": 4.661457676958622,
      "grad_norm": 1.125,
      "learning_rate": 0.0003020420859727165,
      "loss": 4.6895,
      "mean_token_accuracy": 0.2346004769206047,
      "num_tokens": 102376893.0,
      "step": 55485
    },
    {
      "entropy": 5.338548851013184,
      "epoch": 4.661877756773787,
      "grad_norm": 1.078125,
      "learning_rate": 0.000302012356258709,
      "loss": 4.8371,
      "mean_token_accuracy": 0.22617494761943818,
      "num_tokens": 102386066.0,
      "step": 55490
    },
    {
      "entropy": 5.476619720458984,
      "epoch": 4.662297836588952,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030198262606619147,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21891734302043914,
      "num_tokens": 102395100.0,
      "step": 55495
    },
    {
      "entropy": 5.53309907913208,
      "epoch": 4.662717916404117,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003019528953956907,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2151843011379242,
      "num_tokens": 102404275.0,
      "step": 55500
    },
    {
      "entropy": 5.56096305847168,
      "epoch": 4.663137996219282,
      "grad_norm": 1.578125,
      "learning_rate": 0.00030192316424773326,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21510347425937654,
      "num_tokens": 102414771.0,
      "step": 55505
    },
    {
      "entropy": 5.520445823669434,
      "epoch": 4.663558076034446,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003018934326228458,
      "loss": 4.9211,
      "mean_token_accuracy": 0.2117664709687233,
      "num_tokens": 102422986.0,
      "step": 55510
    },
    {
      "entropy": 5.374188661575317,
      "epoch": 4.663978155849612,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003018637005215549,
      "loss": 4.8392,
      "mean_token_accuracy": 0.2315279498696327,
      "num_tokens": 102432033.0,
      "step": 55515
    },
    {
      "entropy": 5.454856014251709,
      "epoch": 4.664398235664776,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030183396794438756,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2244147390127182,
      "num_tokens": 102441061.0,
      "step": 55520
    },
    {
      "entropy": 5.5066869258880615,
      "epoch": 4.664818315479941,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030180423489187025,
      "loss": 4.9229,
      "mean_token_accuracy": 0.21496306508779525,
      "num_tokens": 102450411.0,
      "step": 55525
    },
    {
      "entropy": 5.439063978195191,
      "epoch": 4.665238395295106,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030177450136452974,
      "loss": 4.8617,
      "mean_token_accuracy": 0.2248008668422699,
      "num_tokens": 102459232.0,
      "step": 55530
    },
    {
      "entropy": 5.476440954208374,
      "epoch": 4.665658475110271,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003017447673628927,
      "loss": 4.9274,
      "mean_token_accuracy": 0.2176782950758934,
      "num_tokens": 102468460.0,
      "step": 55535
    },
    {
      "entropy": 5.453627872467041,
      "epoch": 4.666078554925436,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030171503288748585,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21553492397069932,
      "num_tokens": 102477328.0,
      "step": 55540
    },
    {
      "entropy": 5.487293004989624,
      "epoch": 4.6664986347406,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000301685297938836,
      "loss": 4.8965,
      "mean_token_accuracy": 0.22038578391075134,
      "num_tokens": 102487330.0,
      "step": 55545
    },
    {
      "entropy": 5.460587644577027,
      "epoch": 4.666918714555766,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003016555625174698,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2147369608283043,
      "num_tokens": 102496901.0,
      "step": 55550
    },
    {
      "entropy": 5.432525205612182,
      "epoch": 4.667338794370931,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030162582662391404,
      "loss": 4.8002,
      "mean_token_accuracy": 0.23076920658349992,
      "num_tokens": 102506327.0,
      "step": 55555
    },
    {
      "entropy": 5.368944120407105,
      "epoch": 4.667758874186095,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003015960902586955,
      "loss": 4.8239,
      "mean_token_accuracy": 0.22005745023488998,
      "num_tokens": 102515467.0,
      "step": 55560
    },
    {
      "entropy": 5.479564809799195,
      "epoch": 4.66817895400126,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003015663534223409,
      "loss": 4.9194,
      "mean_token_accuracy": 0.21827913969755172,
      "num_tokens": 102524163.0,
      "step": 55565
    },
    {
      "entropy": 5.4442461967468265,
      "epoch": 4.668599033816426,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000301536616115377,
      "loss": 4.8732,
      "mean_token_accuracy": 0.21774342209100722,
      "num_tokens": 102533001.0,
      "step": 55570
    },
    {
      "entropy": 5.418357944488525,
      "epoch": 4.66901911363159,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00030150687833833067,
      "loss": 4.8307,
      "mean_token_accuracy": 0.22661537677049637,
      "num_tokens": 102542479.0,
      "step": 55575
    },
    {
      "entropy": 5.461263847351074,
      "epoch": 4.669439193446755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003014771400917285,
      "loss": 4.9508,
      "mean_token_accuracy": 0.2087525248527527,
      "num_tokens": 102551670.0,
      "step": 55580
    },
    {
      "entropy": 5.537664222717285,
      "epoch": 4.6698592732619195,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030144740137609755,
      "loss": 5.0574,
      "mean_token_accuracy": 0.21195725351572037,
      "num_tokens": 102561641.0,
      "step": 55585
    },
    {
      "entropy": 5.528768110275268,
      "epoch": 4.670279353077085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003014176621919645,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21268968433141708,
      "num_tokens": 102570328.0,
      "step": 55590
    },
    {
      "entropy": 5.428708934783936,
      "epoch": 4.670699432892249,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030138792253985616,
      "loss": 4.803,
      "mean_token_accuracy": 0.2271408975124359,
      "num_tokens": 102579249.0,
      "step": 55595
    },
    {
      "entropy": 5.425085639953613,
      "epoch": 4.671119512707414,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030135818242029935,
      "loss": 4.8867,
      "mean_token_accuracy": 0.2177612617611885,
      "num_tokens": 102588637.0,
      "step": 55600
    },
    {
      "entropy": 5.443942975997925,
      "epoch": 4.6715395925225796,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003013284418338209,
      "loss": 4.8839,
      "mean_token_accuracy": 0.2234516903758049,
      "num_tokens": 102598476.0,
      "step": 55605
    },
    {
      "entropy": 5.4813707828521725,
      "epoch": 4.671959672337744,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003012987007809477,
      "loss": 4.9253,
      "mean_token_accuracy": 0.2135149657726288,
      "num_tokens": 102606798.0,
      "step": 55610
    },
    {
      "entropy": 5.519938135147095,
      "epoch": 4.672379752152909,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003012689592622065,
      "loss": 4.9566,
      "mean_token_accuracy": 0.21268806159496306,
      "num_tokens": 102615809.0,
      "step": 55615
    },
    {
      "entropy": 5.485223293304443,
      "epoch": 4.672799831968074,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003012392172781243,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21072934269905091,
      "num_tokens": 102626122.0,
      "step": 55620
    },
    {
      "entropy": 5.412709331512451,
      "epoch": 4.673219911783239,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003012094748292278,
      "loss": 4.8299,
      "mean_token_accuracy": 0.22005725800991058,
      "num_tokens": 102635235.0,
      "step": 55625
    },
    {
      "entropy": 5.402344846725464,
      "epoch": 4.673639991598404,
      "grad_norm": 1.078125,
      "learning_rate": 0.000301179731916044,
      "loss": 4.8401,
      "mean_token_accuracy": 0.22361618429422378,
      "num_tokens": 102644198.0,
      "step": 55630
    },
    {
      "entropy": 5.4147923469543455,
      "epoch": 4.674060071413568,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030114998853909966,
      "loss": 4.8572,
      "mean_token_accuracy": 0.22633204907178878,
      "num_tokens": 102654094.0,
      "step": 55635
    },
    {
      "entropy": 5.573966646194458,
      "epoch": 4.6744801512287335,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003011202446989218,
      "loss": 5.0846,
      "mean_token_accuracy": 0.20300621837377547,
      "num_tokens": 102663994.0,
      "step": 55640
    },
    {
      "entropy": 5.58782548904419,
      "epoch": 4.674900231043898,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030109050039603717,
      "loss": 4.9984,
      "mean_token_accuracy": 0.20582996904850007,
      "num_tokens": 102674453.0,
      "step": 55645
    },
    {
      "entropy": 5.457020568847656,
      "epoch": 4.675320310859063,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003010607556309728,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2133360520005226,
      "num_tokens": 102683538.0,
      "step": 55650
    },
    {
      "entropy": 5.469255161285401,
      "epoch": 4.675740390674228,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003010310104042556,
      "loss": 4.8459,
      "mean_token_accuracy": 0.21715502738952636,
      "num_tokens": 102692307.0,
      "step": 55655
    },
    {
      "entropy": 5.381317901611328,
      "epoch": 4.676160470489393,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030100126471641243,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21304733753204347,
      "num_tokens": 102701379.0,
      "step": 55660
    },
    {
      "entropy": 5.393168973922729,
      "epoch": 4.676580550304558,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003009715185679702,
      "loss": 4.8262,
      "mean_token_accuracy": 0.21472048163414,
      "num_tokens": 102711440.0,
      "step": 55665
    },
    {
      "entropy": 5.450568723678589,
      "epoch": 4.677000630119723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000300941771959456,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2184698224067688,
      "num_tokens": 102720390.0,
      "step": 55670
    },
    {
      "entropy": 5.514615106582641,
      "epoch": 4.6774207099348875,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030091202489139664,
      "loss": 4.8633,
      "mean_token_accuracy": 0.21308188885450363,
      "num_tokens": 102729065.0,
      "step": 55675
    },
    {
      "entropy": 5.472133016586303,
      "epoch": 4.677840789750053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030088227736431905,
      "loss": 4.8929,
      "mean_token_accuracy": 0.22075984328985215,
      "num_tokens": 102738948.0,
      "step": 55680
    },
    {
      "entropy": 5.4009809494018555,
      "epoch": 4.678260869565217,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003008525293787503,
      "loss": 4.8772,
      "mean_token_accuracy": 0.21993754506111146,
      "num_tokens": 102748859.0,
      "step": 55685
    },
    {
      "entropy": 5.447149038314819,
      "epoch": 4.678680949380382,
      "grad_norm": 1.125,
      "learning_rate": 0.0003008227809352173,
      "loss": 4.8517,
      "mean_token_accuracy": 0.2136866569519043,
      "num_tokens": 102757689.0,
      "step": 55690
    },
    {
      "entropy": 5.4570684909820555,
      "epoch": 4.6791010291955475,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030079303203424694,
      "loss": 4.8352,
      "mean_token_accuracy": 0.21597931236028672,
      "num_tokens": 102766074.0,
      "step": 55695
    },
    {
      "entropy": 5.534800481796265,
      "epoch": 4.679521109010712,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030076328267636637,
      "loss": 4.9514,
      "mean_token_accuracy": 0.21190141439437865,
      "num_tokens": 102775647.0,
      "step": 55700
    },
    {
      "entropy": 5.465570735931396,
      "epoch": 4.679941188825877,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030073353286210256,
      "loss": 4.8579,
      "mean_token_accuracy": 0.21398387104272842,
      "num_tokens": 102785406.0,
      "step": 55705
    },
    {
      "entropy": 5.436438322067261,
      "epoch": 4.680361268641041,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003007037825919824,
      "loss": 4.8778,
      "mean_token_accuracy": 0.2239184111356735,
      "num_tokens": 102794642.0,
      "step": 55710
    },
    {
      "entropy": 5.477288293838501,
      "epoch": 4.680781348456207,
      "grad_norm": 1.125,
      "learning_rate": 0.0003006740318665329,
      "loss": 4.9149,
      "mean_token_accuracy": 0.21627793610095977,
      "num_tokens": 102802954.0,
      "step": 55715
    },
    {
      "entropy": 5.466289806365967,
      "epoch": 4.681201428271372,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030064428068628127,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22141362428665162,
      "num_tokens": 102812289.0,
      "step": 55720
    },
    {
      "entropy": 5.484282827377319,
      "epoch": 4.681621508086536,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030061452905175433,
      "loss": 4.8881,
      "mean_token_accuracy": 0.21399415880441666,
      "num_tokens": 102821713.0,
      "step": 55725
    },
    {
      "entropy": 5.484799242019653,
      "epoch": 4.682041587901701,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030058477696347924,
      "loss": 4.8676,
      "mean_token_accuracy": 0.21962660998106004,
      "num_tokens": 102830917.0,
      "step": 55730
    },
    {
      "entropy": 5.457583093643189,
      "epoch": 4.682461667716867,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030055502442198303,
      "loss": 4.907,
      "mean_token_accuracy": 0.2113744780421257,
      "num_tokens": 102840259.0,
      "step": 55735
    },
    {
      "entropy": 5.458172082901001,
      "epoch": 4.682881747532031,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003005252714277927,
      "loss": 4.9429,
      "mean_token_accuracy": 0.2158936396241188,
      "num_tokens": 102849331.0,
      "step": 55740
    },
    {
      "entropy": 5.421345376968384,
      "epoch": 4.683301827347196,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003004955179814353,
      "loss": 4.861,
      "mean_token_accuracy": 0.21980981528759003,
      "num_tokens": 102858085.0,
      "step": 55745
    },
    {
      "entropy": 5.473105621337891,
      "epoch": 4.683721907162361,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030046576408343794,
      "loss": 4.8796,
      "mean_token_accuracy": 0.2181428223848343,
      "num_tokens": 102868226.0,
      "step": 55750
    },
    {
      "entropy": 5.431120300292969,
      "epoch": 4.684141986977526,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003004360097343277,
      "loss": 4.9072,
      "mean_token_accuracy": 0.2160663992166519,
      "num_tokens": 102877513.0,
      "step": 55755
    },
    {
      "entropy": 5.422254419326782,
      "epoch": 4.68456206679269,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030040625493463164,
      "loss": 4.929,
      "mean_token_accuracy": 0.22179352045059203,
      "num_tokens": 102887487.0,
      "step": 55760
    },
    {
      "entropy": 5.4755775928497314,
      "epoch": 4.684982146607855,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030037649968487685,
      "loss": 4.9329,
      "mean_token_accuracy": 0.2194121703505516,
      "num_tokens": 102896238.0,
      "step": 55765
    },
    {
      "entropy": 5.524841213226319,
      "epoch": 4.685402226423021,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003003467439855904,
      "loss": 4.9421,
      "mean_token_accuracy": 0.21571791023015977,
      "num_tokens": 102905635.0,
      "step": 55770
    },
    {
      "entropy": 5.442757081985474,
      "epoch": 4.685822306238185,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003003169878372995,
      "loss": 4.8763,
      "mean_token_accuracy": 0.21246642619371414,
      "num_tokens": 102914909.0,
      "step": 55775
    },
    {
      "entropy": 5.391583442687988,
      "epoch": 4.68624238605335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030028723124053117,
      "loss": 4.7832,
      "mean_token_accuracy": 0.22397435903549195,
      "num_tokens": 102924404.0,
      "step": 55780
    },
    {
      "entropy": 5.4243492603302,
      "epoch": 4.686662465868515,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003002574741958125,
      "loss": 4.8885,
      "mean_token_accuracy": 0.21954572349786758,
      "num_tokens": 102933417.0,
      "step": 55785
    },
    {
      "entropy": 5.405896997451782,
      "epoch": 4.68708254568368,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003002277167036707,
      "loss": 4.8712,
      "mean_token_accuracy": 0.21509913951158524,
      "num_tokens": 102942739.0,
      "step": 55790
    },
    {
      "entropy": 5.596616220474243,
      "epoch": 4.687502625498845,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030019795876463294,
      "loss": 5.0448,
      "mean_token_accuracy": 0.20414612889289857,
      "num_tokens": 102953675.0,
      "step": 55795
    },
    {
      "entropy": 5.513381052017212,
      "epoch": 4.687922705314009,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003001682003792263,
      "loss": 4.9043,
      "mean_token_accuracy": 0.21848333477973939,
      "num_tokens": 102963168.0,
      "step": 55800
    },
    {
      "entropy": 5.5240837097167965,
      "epoch": 4.688342785129175,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003001384415479779,
      "loss": 5.0067,
      "mean_token_accuracy": 0.21059591621160506,
      "num_tokens": 102972472.0,
      "step": 55805
    },
    {
      "entropy": 5.4070281982421875,
      "epoch": 4.68876286494434,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003001086822714151,
      "loss": 4.8428,
      "mean_token_accuracy": 0.22122933864593505,
      "num_tokens": 102981776.0,
      "step": 55810
    },
    {
      "entropy": 5.413099241256714,
      "epoch": 4.689182944759504,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003000789225500647,
      "loss": 4.867,
      "mean_token_accuracy": 0.2213425189256668,
      "num_tokens": 102991249.0,
      "step": 55815
    },
    {
      "entropy": 5.4903069019317625,
      "epoch": 4.689603024574669,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030004916238445434,
      "loss": 4.9027,
      "mean_token_accuracy": 0.2146654784679413,
      "num_tokens": 103000448.0,
      "step": 55820
    },
    {
      "entropy": 5.451664686203003,
      "epoch": 4.690023104389834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030001940177511085,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22048920392990112,
      "num_tokens": 103009776.0,
      "step": 55825
    },
    {
      "entropy": 5.480648231506348,
      "epoch": 4.690443184204999,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002999896407225615,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22260497212409974,
      "num_tokens": 103019891.0,
      "step": 55830
    },
    {
      "entropy": 5.4129022598266605,
      "epoch": 4.690863264020164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002999598792273336,
      "loss": 4.88,
      "mean_token_accuracy": 0.22352721095085143,
      "num_tokens": 103028815.0,
      "step": 55835
    },
    {
      "entropy": 5.513719177246093,
      "epoch": 4.6912833438353285,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002999301172899543,
      "loss": 4.9492,
      "mean_token_accuracy": 0.2153528153896332,
      "num_tokens": 103037424.0,
      "step": 55840
    },
    {
      "entropy": 5.4562324523925785,
      "epoch": 4.691703423650494,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029990035491095075,
      "loss": 4.8785,
      "mean_token_accuracy": 0.22141659557819365,
      "num_tokens": 103045877.0,
      "step": 55845
    },
    {
      "entropy": 5.410989999771118,
      "epoch": 4.692123503465658,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029987059209085024,
      "loss": 4.8049,
      "mean_token_accuracy": 0.22102713882923125,
      "num_tokens": 103055237.0,
      "step": 55850
    },
    {
      "entropy": 5.4103498458862305,
      "epoch": 4.692543583280823,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029984082883018,
      "loss": 4.8778,
      "mean_token_accuracy": 0.22038723677396774,
      "num_tokens": 103064439.0,
      "step": 55855
    },
    {
      "entropy": 5.432931995391845,
      "epoch": 4.692963663095989,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002998110651294673,
      "loss": 4.8204,
      "mean_token_accuracy": 0.21928150653839112,
      "num_tokens": 103072961.0,
      "step": 55860
    },
    {
      "entropy": 5.430190515518189,
      "epoch": 4.693383742911153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029978130098923933,
      "loss": 4.8971,
      "mean_token_accuracy": 0.22074990272521972,
      "num_tokens": 103081749.0,
      "step": 55865
    },
    {
      "entropy": 5.446917009353638,
      "epoch": 4.693803822726318,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029975153641002334,
      "loss": 4.8734,
      "mean_token_accuracy": 0.2216469094157219,
      "num_tokens": 103090858.0,
      "step": 55870
    },
    {
      "entropy": 5.4708692073822025,
      "epoch": 4.6942239025414825,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029972177139234667,
      "loss": 4.9654,
      "mean_token_accuracy": 0.21651619225740432,
      "num_tokens": 103100773.0,
      "step": 55875
    },
    {
      "entropy": 5.483147525787354,
      "epoch": 4.694643982356648,
      "grad_norm": 1.125,
      "learning_rate": 0.0002996920059367366,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21398291289806365,
      "num_tokens": 103110275.0,
      "step": 55880
    },
    {
      "entropy": 5.435331392288208,
      "epoch": 4.695064062171813,
      "grad_norm": 1.125,
      "learning_rate": 0.0002996622400437203,
      "loss": 4.7368,
      "mean_token_accuracy": 0.23568743616342544,
      "num_tokens": 103119953.0,
      "step": 55885
    },
    {
      "entropy": 5.453785133361817,
      "epoch": 4.695484141986977,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029963247371382515,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21288827508687974,
      "num_tokens": 103129279.0,
      "step": 55890
    },
    {
      "entropy": 5.445659589767456,
      "epoch": 4.6959042218021425,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002996027069475784,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2188740149140358,
      "num_tokens": 103138319.0,
      "step": 55895
    },
    {
      "entropy": 5.40391583442688,
      "epoch": 4.696324301617308,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029957293974550733,
      "loss": 4.8908,
      "mean_token_accuracy": 0.21694615483283997,
      "num_tokens": 103146694.0,
      "step": 55900
    },
    {
      "entropy": 5.463732147216797,
      "epoch": 4.696744381432472,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002995431721081393,
      "loss": 4.9047,
      "mean_token_accuracy": 0.21284673362970352,
      "num_tokens": 103156748.0,
      "step": 55905
    },
    {
      "entropy": 5.426425933837891,
      "epoch": 4.697164461247637,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029951340403600165,
      "loss": 4.8185,
      "mean_token_accuracy": 0.22171878665685654,
      "num_tokens": 103166040.0,
      "step": 55910
    },
    {
      "entropy": 5.418071842193603,
      "epoch": 4.697584541062802,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002994836355296216,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22219040542840957,
      "num_tokens": 103175840.0,
      "step": 55915
    },
    {
      "entropy": 5.397780704498291,
      "epoch": 4.698004620877967,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029945386658952657,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22997195273637772,
      "num_tokens": 103185242.0,
      "step": 55920
    },
    {
      "entropy": 5.542292070388794,
      "epoch": 4.698424700693131,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002994240972162439,
      "loss": 4.9686,
      "mean_token_accuracy": 0.2081766039133072,
      "num_tokens": 103194443.0,
      "step": 55925
    },
    {
      "entropy": 5.47090368270874,
      "epoch": 4.6988447805082965,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000299394327410301,
      "loss": 4.8781,
      "mean_token_accuracy": 0.21910077035427095,
      "num_tokens": 103202750.0,
      "step": 55930
    },
    {
      "entropy": 5.3877684593200685,
      "epoch": 4.699264860323462,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029936455717222507,
      "loss": 4.8336,
      "mean_token_accuracy": 0.21609449088573457,
      "num_tokens": 103211110.0,
      "step": 55935
    },
    {
      "entropy": 5.381369161605835,
      "epoch": 4.699684940138626,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029933478650254354,
      "loss": 4.8405,
      "mean_token_accuracy": 0.22385285049676895,
      "num_tokens": 103220233.0,
      "step": 55940
    },
    {
      "entropy": 5.4998030185699465,
      "epoch": 4.700105019953791,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029930501540178383,
      "loss": 4.8905,
      "mean_token_accuracy": 0.20936246663331987,
      "num_tokens": 103230254.0,
      "step": 55945
    },
    {
      "entropy": 5.504878187179566,
      "epoch": 4.7005250997689565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029927524387047335,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21607645899057387,
      "num_tokens": 103239644.0,
      "step": 55950
    },
    {
      "entropy": 5.436278009414673,
      "epoch": 4.700945179584121,
      "grad_norm": 1.125,
      "learning_rate": 0.0002992454719091394,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21939898133277894,
      "num_tokens": 103249033.0,
      "step": 55955
    },
    {
      "entropy": 5.375126552581787,
      "epoch": 4.701365259399286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002992156995183093,
      "loss": 4.8062,
      "mean_token_accuracy": 0.225173619389534,
      "num_tokens": 103257872.0,
      "step": 55960
    },
    {
      "entropy": 5.435326623916626,
      "epoch": 4.70178533921445,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002991859266985106,
      "loss": 4.9475,
      "mean_token_accuracy": 0.21103848665952682,
      "num_tokens": 103266920.0,
      "step": 55965
    },
    {
      "entropy": 5.553891658782959,
      "epoch": 4.702205419029616,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002991561534502707,
      "loss": 4.9797,
      "mean_token_accuracy": 0.21024443060159684,
      "num_tokens": 103276006.0,
      "step": 55970
    },
    {
      "entropy": 5.433764696121216,
      "epoch": 4.702625498844781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002991263797741169,
      "loss": 4.793,
      "mean_token_accuracy": 0.22027264386415482,
      "num_tokens": 103284757.0,
      "step": 55975
    },
    {
      "entropy": 5.483930063247681,
      "epoch": 4.703045578659945,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029909660567057684,
      "loss": 5.0171,
      "mean_token_accuracy": 0.20323722809553146,
      "num_tokens": 103295163.0,
      "step": 55980
    },
    {
      "entropy": 5.497919797897339,
      "epoch": 4.7034656584751104,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002990668311401777,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2115251377224922,
      "num_tokens": 103304150.0,
      "step": 55985
    },
    {
      "entropy": 5.429057455062866,
      "epoch": 4.703885738290275,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029903705618344704,
      "loss": 4.8092,
      "mean_token_accuracy": 0.22700829356908797,
      "num_tokens": 103312731.0,
      "step": 55990
    },
    {
      "entropy": 5.422439432144165,
      "epoch": 4.70430581810544,
      "grad_norm": 1.125,
      "learning_rate": 0.00029900728080091237,
      "loss": 4.9066,
      "mean_token_accuracy": 0.21544847935438155,
      "num_tokens": 103322253.0,
      "step": 55995
    },
    {
      "entropy": 5.5302191257476805,
      "epoch": 4.704725897920605,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029897750499310107,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21142754703760147,
      "num_tokens": 103331151.0,
      "step": 56000
    },
    {
      "entropy": 5.516296148300171,
      "epoch": 4.70514597773577,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029894772876054067,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21671930104494094,
      "num_tokens": 103339996.0,
      "step": 56005
    },
    {
      "entropy": 5.480396461486817,
      "epoch": 4.705566057550935,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029891795210375846,
      "loss": 4.8726,
      "mean_token_accuracy": 0.2243341714143753,
      "num_tokens": 103349160.0,
      "step": 56010
    },
    {
      "entropy": 5.483700609207153,
      "epoch": 4.7059861373661,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002988881750232821,
      "loss": 4.919,
      "mean_token_accuracy": 0.22270269244909285,
      "num_tokens": 103358325.0,
      "step": 56015
    },
    {
      "entropy": 5.51353235244751,
      "epoch": 4.706406217181264,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029885839751963905,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22284756898880004,
      "num_tokens": 103366373.0,
      "step": 56020
    },
    {
      "entropy": 5.441526937484741,
      "epoch": 4.70682629699643,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029882861959335677,
      "loss": 4.896,
      "mean_token_accuracy": 0.22096028476953505,
      "num_tokens": 103375849.0,
      "step": 56025
    },
    {
      "entropy": 5.485047054290772,
      "epoch": 4.707246376811594,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029879884124496276,
      "loss": 4.9896,
      "mean_token_accuracy": 0.20524532049894334,
      "num_tokens": 103385424.0,
      "step": 56030
    },
    {
      "entropy": 5.511054182052613,
      "epoch": 4.707666456626759,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002987690624749845,
      "loss": 4.9177,
      "mean_token_accuracy": 0.21931935846805573,
      "num_tokens": 103395061.0,
      "step": 56035
    },
    {
      "entropy": 5.510229206085205,
      "epoch": 4.7080865364419235,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029873928328394953,
      "loss": 4.8846,
      "mean_token_accuracy": 0.21628629565238952,
      "num_tokens": 103403791.0,
      "step": 56040
    },
    {
      "entropy": 5.460731506347656,
      "epoch": 4.708506616257089,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002987095036723854,
      "loss": 4.8874,
      "mean_token_accuracy": 0.21883353739976882,
      "num_tokens": 103412882.0,
      "step": 56045
    },
    {
      "entropy": 5.40406289100647,
      "epoch": 4.708926696072254,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002986797236408196,
      "loss": 4.8518,
      "mean_token_accuracy": 0.21544483602046965,
      "num_tokens": 103422155.0,
      "step": 56050
    },
    {
      "entropy": 5.5303527355194095,
      "epoch": 4.709346775887418,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00029864994318977965,
      "loss": 4.9357,
      "mean_token_accuracy": 0.2211714953184128,
      "num_tokens": 103431471.0,
      "step": 56055
    },
    {
      "entropy": 5.478689861297608,
      "epoch": 4.709766855702584,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002986201623197933,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22823977917432786,
      "num_tokens": 103441220.0,
      "step": 56060
    },
    {
      "entropy": 5.499627256393433,
      "epoch": 4.710186935517749,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002985903810313878,
      "loss": 4.935,
      "mean_token_accuracy": 0.2115200713276863,
      "num_tokens": 103450798.0,
      "step": 56065
    },
    {
      "entropy": 5.385040950775147,
      "epoch": 4.710607015332913,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029856059932509087,
      "loss": 4.7948,
      "mean_token_accuracy": 0.22436854094266892,
      "num_tokens": 103459227.0,
      "step": 56070
    },
    {
      "entropy": 5.462242984771729,
      "epoch": 4.711027095148078,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029853081720143004,
      "loss": 4.9255,
      "mean_token_accuracy": 0.21017962992191314,
      "num_tokens": 103468424.0,
      "step": 56075
    },
    {
      "entropy": 5.473430347442627,
      "epoch": 4.711447174963243,
      "grad_norm": 1.15625,
      "learning_rate": 0.000298501034660933,
      "loss": 4.9647,
      "mean_token_accuracy": 0.21260615885257722,
      "num_tokens": 103478239.0,
      "step": 56080
    },
    {
      "entropy": 5.484047555923462,
      "epoch": 4.711867254778408,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029847125170412713,
      "loss": 4.8622,
      "mean_token_accuracy": 0.21933605372905732,
      "num_tokens": 103487420.0,
      "step": 56085
    },
    {
      "entropy": 5.425965881347656,
      "epoch": 4.712287334593572,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002984414683315402,
      "loss": 4.8312,
      "mean_token_accuracy": 0.22353503853082657,
      "num_tokens": 103496512.0,
      "step": 56090
    },
    {
      "entropy": 5.429271221160889,
      "epoch": 4.7127074144087375,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029841168454369967,
      "loss": 4.8272,
      "mean_token_accuracy": 0.22506688237190248,
      "num_tokens": 103504395.0,
      "step": 56095
    },
    {
      "entropy": 5.399818181991577,
      "epoch": 4.713127494223903,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002983819003411332,
      "loss": 4.9382,
      "mean_token_accuracy": 0.2176978901028633,
      "num_tokens": 103513830.0,
      "step": 56100
    },
    {
      "entropy": 5.507645511627198,
      "epoch": 4.713547574039067,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029835211572436845,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2109343335032463,
      "num_tokens": 103521685.0,
      "step": 56105
    },
    {
      "entropy": 5.567578744888306,
      "epoch": 4.713967653854232,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002983223306939329,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21325420141220092,
      "num_tokens": 103530447.0,
      "step": 56110
    },
    {
      "entropy": 5.501605987548828,
      "epoch": 4.714387733669398,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029829254525035433,
      "loss": 4.942,
      "mean_token_accuracy": 0.21747890561819078,
      "num_tokens": 103538961.0,
      "step": 56115
    },
    {
      "entropy": 5.426162624359131,
      "epoch": 4.714807813484562,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029826275939416033,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2124241530895233,
      "num_tokens": 103548241.0,
      "step": 56120
    },
    {
      "entropy": 5.473891973495483,
      "epoch": 4.715227893299727,
      "grad_norm": 1.125,
      "learning_rate": 0.0002982329731258785,
      "loss": 4.9809,
      "mean_token_accuracy": 0.21502641439437867,
      "num_tokens": 103557229.0,
      "step": 56125
    },
    {
      "entropy": 5.470236158370971,
      "epoch": 4.7156479731148915,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029820318644603654,
      "loss": 4.8914,
      "mean_token_accuracy": 0.2153386041522026,
      "num_tokens": 103567545.0,
      "step": 56130
    },
    {
      "entropy": 5.486643648147583,
      "epoch": 4.716068052930057,
      "grad_norm": 1.15625,
      "learning_rate": 0.000298173399355162,
      "loss": 4.9183,
      "mean_token_accuracy": 0.2208416536450386,
      "num_tokens": 103576060.0,
      "step": 56135
    },
    {
      "entropy": 5.401233053207397,
      "epoch": 4.716488132745222,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029814361185378276,
      "loss": 4.9111,
      "mean_token_accuracy": 0.2202914610505104,
      "num_tokens": 103586191.0,
      "step": 56140
    },
    {
      "entropy": 5.474684953689575,
      "epoch": 4.716908212560386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002981138239424263,
      "loss": 4.8671,
      "mean_token_accuracy": 0.2185742199420929,
      "num_tokens": 103595447.0,
      "step": 56145
    },
    {
      "entropy": 5.478495693206787,
      "epoch": 4.7173282923755515,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029808403562162035,
      "loss": 4.8653,
      "mean_token_accuracy": 0.22299010157585145,
      "num_tokens": 103605222.0,
      "step": 56150
    },
    {
      "entropy": 5.442798471450805,
      "epoch": 4.717748372190716,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002980542468918925,
      "loss": 4.905,
      "mean_token_accuracy": 0.2225141704082489,
      "num_tokens": 103613741.0,
      "step": 56155
    },
    {
      "entropy": 5.414291477203369,
      "epoch": 4.718168452005881,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002980244577537707,
      "loss": 4.8377,
      "mean_token_accuracy": 0.2250019446015358,
      "num_tokens": 103623399.0,
      "step": 56160
    },
    {
      "entropy": 5.486568737030029,
      "epoch": 4.718588531821046,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002979946682077824,
      "loss": 4.8739,
      "mean_token_accuracy": 0.22056620717048644,
      "num_tokens": 103632736.0,
      "step": 56165
    },
    {
      "entropy": 5.44663438796997,
      "epoch": 4.719008611636211,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002979648782544554,
      "loss": 4.8601,
      "mean_token_accuracy": 0.21406541168689727,
      "num_tokens": 103641017.0,
      "step": 56170
    },
    {
      "entropy": 5.488203048706055,
      "epoch": 4.719428691451376,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002979350878943174,
      "loss": 4.9104,
      "mean_token_accuracy": 0.21656567901372908,
      "num_tokens": 103649514.0,
      "step": 56175
    },
    {
      "entropy": 5.434928131103516,
      "epoch": 4.719848771266541,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029790529712789613,
      "loss": 4.9149,
      "mean_token_accuracy": 0.21489461064338683,
      "num_tokens": 103659444.0,
      "step": 56180
    },
    {
      "entropy": 5.4691181659698485,
      "epoch": 4.7202688510817055,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029787550595571934,
      "loss": 4.878,
      "mean_token_accuracy": 0.21889561861753465,
      "num_tokens": 103667505.0,
      "step": 56185
    },
    {
      "entropy": 5.437359952926636,
      "epoch": 4.720688930896871,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002978457143783147,
      "loss": 4.9124,
      "mean_token_accuracy": 0.2183072790503502,
      "num_tokens": 103676112.0,
      "step": 56190
    },
    {
      "entropy": 5.554619932174683,
      "epoch": 4.721109010712035,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029781592239620997,
      "loss": 4.9962,
      "mean_token_accuracy": 0.2153442919254303,
      "num_tokens": 103686564.0,
      "step": 56195
    },
    {
      "entropy": 5.51021637916565,
      "epoch": 4.7215290905272,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029778613000993297,
      "loss": 4.8775,
      "mean_token_accuracy": 0.21635417640209198,
      "num_tokens": 103695091.0,
      "step": 56200
    },
    {
      "entropy": 5.490937995910644,
      "epoch": 4.721949170342365,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029775633722001145,
      "loss": 4.9381,
      "mean_token_accuracy": 0.21071213334798813,
      "num_tokens": 103704820.0,
      "step": 56205
    },
    {
      "entropy": 5.395477676391602,
      "epoch": 4.72236925015753,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002977265440269731,
      "loss": 4.8294,
      "mean_token_accuracy": 0.2243237867951393,
      "num_tokens": 103714631.0,
      "step": 56210
    },
    {
      "entropy": 5.451538515090943,
      "epoch": 4.722789329972695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002976967504313457,
      "loss": 4.901,
      "mean_token_accuracy": 0.21876534223556518,
      "num_tokens": 103724156.0,
      "step": 56215
    },
    {
      "entropy": 5.529795837402344,
      "epoch": 4.723209409787859,
      "grad_norm": 1.0625,
      "learning_rate": 0.000297666956433657,
      "loss": 5.0134,
      "mean_token_accuracy": 0.21129706650972366,
      "num_tokens": 103734309.0,
      "step": 56220
    },
    {
      "entropy": 5.483452939987183,
      "epoch": 4.723629489603025,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000297637162034435,
      "loss": 4.876,
      "mean_token_accuracy": 0.21757994443178177,
      "num_tokens": 103743538.0,
      "step": 56225
    },
    {
      "entropy": 5.366693687438965,
      "epoch": 4.72404956941819,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002976073672342072,
      "loss": 4.7576,
      "mean_token_accuracy": 0.22859796434640883,
      "num_tokens": 103753095.0,
      "step": 56230
    },
    {
      "entropy": 5.474149513244629,
      "epoch": 4.724469649233354,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002975775720335016,
      "loss": 4.9665,
      "mean_token_accuracy": 0.21039971262216567,
      "num_tokens": 103762238.0,
      "step": 56235
    },
    {
      "entropy": 5.541218566894531,
      "epoch": 4.7248897290485194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002975477764328459,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21219355016946792,
      "num_tokens": 103771640.0,
      "step": 56240
    },
    {
      "entropy": 5.467045164108276,
      "epoch": 4.725309808863684,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000297517980432768,
      "loss": 4.8656,
      "mean_token_accuracy": 0.2164624661207199,
      "num_tokens": 103779877.0,
      "step": 56245
    },
    {
      "entropy": 5.496429061889648,
      "epoch": 4.725729888678849,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002974881840337956,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22269217222929,
      "num_tokens": 103789439.0,
      "step": 56250
    },
    {
      "entropy": 5.447111749649048,
      "epoch": 4.726149968494013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002974583872364566,
      "loss": 4.9183,
      "mean_token_accuracy": 0.21515820324420928,
      "num_tokens": 103798118.0,
      "step": 56255
    },
    {
      "entropy": 5.472967576980591,
      "epoch": 4.726570048309179,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002974285900412789,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21165023297071456,
      "num_tokens": 103807573.0,
      "step": 56260
    },
    {
      "entropy": 5.527084875106811,
      "epoch": 4.726990128124344,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029739879244879017,
      "loss": 4.9557,
      "mean_token_accuracy": 0.21340622305870055,
      "num_tokens": 103816219.0,
      "step": 56265
    },
    {
      "entropy": 5.498032093048096,
      "epoch": 4.727410207939508,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002973689944595184,
      "loss": 4.8761,
      "mean_token_accuracy": 0.21915532648563385,
      "num_tokens": 103825482.0,
      "step": 56270
    },
    {
      "entropy": 5.537618732452392,
      "epoch": 4.727830287754673,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029733919607399144,
      "loss": 5.0218,
      "mean_token_accuracy": 0.2034021109342575,
      "num_tokens": 103834778.0,
      "step": 56275
    },
    {
      "entropy": 5.445546007156372,
      "epoch": 4.728250367569839,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002973093972927372,
      "loss": 4.9106,
      "mean_token_accuracy": 0.21334071606397628,
      "num_tokens": 103844009.0,
      "step": 56280
    },
    {
      "entropy": 5.466146087646484,
      "epoch": 4.728670447385003,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002972795981162834,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22188461869955062,
      "num_tokens": 103852759.0,
      "step": 56285
    },
    {
      "entropy": 5.517182350158691,
      "epoch": 4.729090527200168,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000297249798545158,
      "loss": 4.9518,
      "mean_token_accuracy": 0.22119184136390685,
      "num_tokens": 103862826.0,
      "step": 56290
    },
    {
      "entropy": 5.497540235519409,
      "epoch": 4.7295106070153325,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029721999857988885,
      "loss": 4.9033,
      "mean_token_accuracy": 0.21874417513608932,
      "num_tokens": 103872484.0,
      "step": 56295
    },
    {
      "entropy": 5.5538677215576175,
      "epoch": 4.729930686830498,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002971901982210039,
      "loss": 4.972,
      "mean_token_accuracy": 0.21160714626312255,
      "num_tokens": 103881539.0,
      "step": 56300
    },
    {
      "entropy": 5.407467460632324,
      "epoch": 4.730350766645663,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002971603974690311,
      "loss": 4.8612,
      "mean_token_accuracy": 0.22160327434539795,
      "num_tokens": 103890578.0,
      "step": 56305
    },
    {
      "entropy": 5.433996677398682,
      "epoch": 4.730770846460827,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002971305963244981,
      "loss": 4.903,
      "mean_token_accuracy": 0.21109994649887084,
      "num_tokens": 103899852.0,
      "step": 56310
    },
    {
      "entropy": 5.43271279335022,
      "epoch": 4.731190926275993,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029710079478793307,
      "loss": 4.8458,
      "mean_token_accuracy": 0.22380639463663102,
      "num_tokens": 103909470.0,
      "step": 56315
    },
    {
      "entropy": 5.4549308776855465,
      "epoch": 4.731611006091157,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029707099285986385,
      "loss": 4.8264,
      "mean_token_accuracy": 0.22701101750135422,
      "num_tokens": 103918850.0,
      "step": 56320
    },
    {
      "entropy": 5.510130310058594,
      "epoch": 4.732031085906322,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029704119054081826,
      "loss": 4.9583,
      "mean_token_accuracy": 0.2168423816561699,
      "num_tokens": 103928948.0,
      "step": 56325
    },
    {
      "entropy": 5.494188833236694,
      "epoch": 4.732451165721487,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029701138783132437,
      "loss": 4.8828,
      "mean_token_accuracy": 0.2172333374619484,
      "num_tokens": 103938247.0,
      "step": 56330
    },
    {
      "entropy": 5.441119241714477,
      "epoch": 4.732871245536652,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002969815847319101,
      "loss": 4.8438,
      "mean_token_accuracy": 0.2256275534629822,
      "num_tokens": 103947668.0,
      "step": 56335
    },
    {
      "entropy": 5.442692756652832,
      "epoch": 4.733291325351817,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029695178124310346,
      "loss": 4.8932,
      "mean_token_accuracy": 0.2128991961479187,
      "num_tokens": 103956678.0,
      "step": 56340
    },
    {
      "entropy": 5.449551391601562,
      "epoch": 4.733711405166982,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029692197736543225,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21563103049993515,
      "num_tokens": 103966538.0,
      "step": 56345
    },
    {
      "entropy": 5.518190813064575,
      "epoch": 4.7341314849821465,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029689217309942443,
      "loss": 4.8875,
      "mean_token_accuracy": 0.21544890850782394,
      "num_tokens": 103975533.0,
      "step": 56350
    },
    {
      "entropy": 5.360585355758667,
      "epoch": 4.734551564797312,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002968623684456081,
      "loss": 4.7031,
      "mean_token_accuracy": 0.2313171461224556,
      "num_tokens": 103984331.0,
      "step": 56355
    },
    {
      "entropy": 5.337446308135986,
      "epoch": 4.734971644612476,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002968325634045112,
      "loss": 4.7952,
      "mean_token_accuracy": 0.22843220680952073,
      "num_tokens": 103993781.0,
      "step": 56360
    },
    {
      "entropy": 5.430945920944214,
      "epoch": 4.735391724427641,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002968027579766616,
      "loss": 4.8946,
      "mean_token_accuracy": 0.22498941719532012,
      "num_tokens": 104002769.0,
      "step": 56365
    },
    {
      "entropy": 5.500844097137451,
      "epoch": 4.735811804242806,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002967729521625875,
      "loss": 4.9763,
      "mean_token_accuracy": 0.20863694995641707,
      "num_tokens": 104011910.0,
      "step": 56370
    },
    {
      "entropy": 5.49162278175354,
      "epoch": 4.736231884057971,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002967431459628167,
      "loss": 4.885,
      "mean_token_accuracy": 0.21457546800374985,
      "num_tokens": 104020110.0,
      "step": 56375
    },
    {
      "entropy": 5.494115257263184,
      "epoch": 4.736651963873136,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002967133393778772,
      "loss": 4.9166,
      "mean_token_accuracy": 0.21335486620664595,
      "num_tokens": 104029748.0,
      "step": 56380
    },
    {
      "entropy": 5.541958856582641,
      "epoch": 4.7370720436883005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029668353240829714,
      "loss": 4.9132,
      "mean_token_accuracy": 0.21462339758872986,
      "num_tokens": 104038980.0,
      "step": 56385
    },
    {
      "entropy": 5.486770296096802,
      "epoch": 4.737492123503466,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002966537250546045,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2156643584370613,
      "num_tokens": 104048486.0,
      "step": 56390
    },
    {
      "entropy": 5.388428354263306,
      "epoch": 4.737912203318631,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002966239173173271,
      "loss": 4.7977,
      "mean_token_accuracy": 0.22784204483032228,
      "num_tokens": 104056312.0,
      "step": 56395
    },
    {
      "entropy": 5.473369216918945,
      "epoch": 4.738332283133795,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029659410919699337,
      "loss": 4.9055,
      "mean_token_accuracy": 0.20931552350521088,
      "num_tokens": 104065560.0,
      "step": 56400
    },
    {
      "entropy": 5.4539039611816404,
      "epoch": 4.7387523629489605,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002965643006941309,
      "loss": 4.9322,
      "mean_token_accuracy": 0.22245126515626906,
      "num_tokens": 104074499.0,
      "step": 56405
    },
    {
      "entropy": 5.4922667980194095,
      "epoch": 4.739172442764125,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002965344918092681,
      "loss": 4.9088,
      "mean_token_accuracy": 0.20934925228357315,
      "num_tokens": 104083654.0,
      "step": 56410
    },
    {
      "entropy": 5.492313098907471,
      "epoch": 4.73959252257929,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002965046825429328,
      "loss": 4.938,
      "mean_token_accuracy": 0.20951227992773055,
      "num_tokens": 104094070.0,
      "step": 56415
    },
    {
      "entropy": 5.495210313796997,
      "epoch": 4.740012602394455,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002964748728956532,
      "loss": 4.8471,
      "mean_token_accuracy": 0.22354421019554138,
      "num_tokens": 104101856.0,
      "step": 56420
    },
    {
      "entropy": 5.52791690826416,
      "epoch": 4.74043268220962,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002964450628679572,
      "loss": 5.0127,
      "mean_token_accuracy": 0.20413588285446166,
      "num_tokens": 104111539.0,
      "step": 56425
    },
    {
      "entropy": 5.419832515716553,
      "epoch": 4.740852762024785,
      "grad_norm": 1.015625,
      "learning_rate": 0.000296415252460373,
      "loss": 4.8034,
      "mean_token_accuracy": 0.22298436909914016,
      "num_tokens": 104121013.0,
      "step": 56430
    },
    {
      "entropy": 5.316411638259888,
      "epoch": 4.741272841839949,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002963854416734286,
      "loss": 4.7584,
      "mean_token_accuracy": 0.226460599899292,
      "num_tokens": 104129503.0,
      "step": 56435
    },
    {
      "entropy": 5.4258692264556885,
      "epoch": 4.7416929216551145,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029635563050765216,
      "loss": 4.73,
      "mean_token_accuracy": 0.22894023060798646,
      "num_tokens": 104138188.0,
      "step": 56440
    },
    {
      "entropy": 5.49761323928833,
      "epoch": 4.74211300147028,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002963258189635717,
      "loss": 5.0029,
      "mean_token_accuracy": 0.2093028485774994,
      "num_tokens": 104147471.0,
      "step": 56445
    },
    {
      "entropy": 5.494932222366333,
      "epoch": 4.742533081285444,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002962960070417153,
      "loss": 4.9138,
      "mean_token_accuracy": 0.22011836916208266,
      "num_tokens": 104155392.0,
      "step": 56450
    },
    {
      "entropy": 5.477214193344116,
      "epoch": 4.742953161100609,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029626619474261114,
      "loss": 4.906,
      "mean_token_accuracy": 0.21914318054914475,
      "num_tokens": 104164975.0,
      "step": 56455
    },
    {
      "entropy": 5.4173863410949705,
      "epoch": 4.743373240915774,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029623638206678724,
      "loss": 4.8626,
      "mean_token_accuracy": 0.2216971203684807,
      "num_tokens": 104173389.0,
      "step": 56460
    },
    {
      "entropy": 5.523058700561523,
      "epoch": 4.743793320730939,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002962065690147718,
      "loss": 4.9865,
      "mean_token_accuracy": 0.21008439660072326,
      "num_tokens": 104182639.0,
      "step": 56465
    },
    {
      "entropy": 5.395482873916626,
      "epoch": 4.744213400546104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002961767555870929,
      "loss": 4.8432,
      "mean_token_accuracy": 0.22165460884571075,
      "num_tokens": 104191150.0,
      "step": 56470
    },
    {
      "entropy": 5.493092393875122,
      "epoch": 4.744633480361268,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029614694178427866,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21574110239744188,
      "num_tokens": 104201446.0,
      "step": 56475
    },
    {
      "entropy": 5.49014458656311,
      "epoch": 4.745053560176434,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029611712760685725,
      "loss": 4.9435,
      "mean_token_accuracy": 0.22039750218391418,
      "num_tokens": 104212128.0,
      "step": 56480
    },
    {
      "entropy": 5.491711139678955,
      "epoch": 4.745473639991598,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029608731305535686,
      "loss": 4.9481,
      "mean_token_accuracy": 0.21121934056282043,
      "num_tokens": 104221373.0,
      "step": 56485
    },
    {
      "entropy": 5.5080146312713625,
      "epoch": 4.745893719806763,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002960574981303055,
      "loss": 4.8863,
      "mean_token_accuracy": 0.21808470338582991,
      "num_tokens": 104230138.0,
      "step": 56490
    },
    {
      "entropy": 5.451275730133057,
      "epoch": 4.7463137996219285,
      "grad_norm": 1.125,
      "learning_rate": 0.0002960276828322315,
      "loss": 4.9174,
      "mean_token_accuracy": 0.21675361841917037,
      "num_tokens": 104239557.0,
      "step": 56495
    },
    {
      "entropy": 5.439977359771729,
      "epoch": 4.746733879437093,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002959978671616629,
      "loss": 4.8556,
      "mean_token_accuracy": 0.2248342901468277,
      "num_tokens": 104248538.0,
      "step": 56500
    },
    {
      "entropy": 5.456973266601563,
      "epoch": 4.747153959252258,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029596805111912783,
      "loss": 4.8258,
      "mean_token_accuracy": 0.21844566613435745,
      "num_tokens": 104258163.0,
      "step": 56505
    },
    {
      "entropy": 5.476148509979248,
      "epoch": 4.747574039067423,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029593823470515455,
      "loss": 4.8778,
      "mean_token_accuracy": 0.2194958060979843,
      "num_tokens": 104266673.0,
      "step": 56510
    },
    {
      "entropy": 5.3906348705291744,
      "epoch": 4.747994118882588,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029590841792027125,
      "loss": 4.8519,
      "mean_token_accuracy": 0.22002287358045577,
      "num_tokens": 104277100.0,
      "step": 56515
    },
    {
      "entropy": 5.442325639724731,
      "epoch": 4.748414198697753,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029587860076500606,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21787989288568496,
      "num_tokens": 104286268.0,
      "step": 56520
    },
    {
      "entropy": 5.425271511077881,
      "epoch": 4.748834278512917,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029584878323988715,
      "loss": 4.8754,
      "mean_token_accuracy": 0.21913433223962783,
      "num_tokens": 104295047.0,
      "step": 56525
    },
    {
      "entropy": 5.438084459304809,
      "epoch": 4.749254358328082,
      "grad_norm": 1.125,
      "learning_rate": 0.0002958189653454429,
      "loss": 4.8819,
      "mean_token_accuracy": 0.21415484547615052,
      "num_tokens": 104303481.0,
      "step": 56530
    },
    {
      "entropy": 5.4951910972595215,
      "epoch": 4.749674438143247,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002957891470822013,
      "loss": 4.8251,
      "mean_token_accuracy": 0.22051169276237487,
      "num_tokens": 104311700.0,
      "step": 56535
    },
    {
      "entropy": 5.507949495315552,
      "epoch": 4.750094517958412,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029575932845069073,
      "loss": 4.8959,
      "mean_token_accuracy": 0.2195934534072876,
      "num_tokens": 104320218.0,
      "step": 56540
    },
    {
      "entropy": 5.539477205276489,
      "epoch": 4.750514597773577,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002957295094514393,
      "loss": 4.9905,
      "mean_token_accuracy": 0.2095385655760765,
      "num_tokens": 104329362.0,
      "step": 56545
    },
    {
      "entropy": 5.395453453063965,
      "epoch": 4.7509346775887416,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002956996900849754,
      "loss": 4.8354,
      "mean_token_accuracy": 0.2196988508105278,
      "num_tokens": 104337933.0,
      "step": 56550
    },
    {
      "entropy": 5.416302347183228,
      "epoch": 4.751354757403907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029566987035182704,
      "loss": 4.7994,
      "mean_token_accuracy": 0.22797638475894927,
      "num_tokens": 104347418.0,
      "step": 56555
    },
    {
      "entropy": 5.474671459197998,
      "epoch": 4.751774837219072,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002956400502525226,
      "loss": 4.8955,
      "mean_token_accuracy": 0.22030114382505417,
      "num_tokens": 104355852.0,
      "step": 56560
    },
    {
      "entropy": 5.471318149566651,
      "epoch": 4.752194917034236,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002956102297875903,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2168654903769493,
      "num_tokens": 104365894.0,
      "step": 56565
    },
    {
      "entropy": 5.4723265171051025,
      "epoch": 4.752614996849402,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002955804089575584,
      "loss": 4.9182,
      "mean_token_accuracy": 0.2202944815158844,
      "num_tokens": 104374953.0,
      "step": 56570
    },
    {
      "entropy": 5.453349637985229,
      "epoch": 4.753035076664566,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029555058776295517,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22221145927906036,
      "num_tokens": 104384119.0,
      "step": 56575
    },
    {
      "entropy": 5.435631370544433,
      "epoch": 4.753455156479731,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002955207662043088,
      "loss": 4.8707,
      "mean_token_accuracy": 0.20996631681919098,
      "num_tokens": 104393365.0,
      "step": 56580
    },
    {
      "entropy": 5.4440757751464846,
      "epoch": 4.753875236294896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002954909442821477,
      "loss": 4.7647,
      "mean_token_accuracy": 0.22690746039152146,
      "num_tokens": 104402147.0,
      "step": 56585
    },
    {
      "entropy": 5.460948657989502,
      "epoch": 4.754295316110061,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029546112199699997,
      "loss": 4.8879,
      "mean_token_accuracy": 0.22037551552057266,
      "num_tokens": 104410478.0,
      "step": 56590
    },
    {
      "entropy": 5.313336515426636,
      "epoch": 4.754715395925226,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029543129934939404,
      "loss": 4.7762,
      "mean_token_accuracy": 0.23173078000545502,
      "num_tokens": 104419208.0,
      "step": 56595
    },
    {
      "entropy": 5.497761392593384,
      "epoch": 4.75513547574039,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002954014763398582,
      "loss": 4.8922,
      "mean_token_accuracy": 0.21942654699087144,
      "num_tokens": 104427500.0,
      "step": 56600
    },
    {
      "entropy": 5.49100775718689,
      "epoch": 4.7555555555555555,
      "grad_norm": 1.0,
      "learning_rate": 0.00029537165296892063,
      "loss": 4.9305,
      "mean_token_accuracy": 0.20695115774869918,
      "num_tokens": 104437833.0,
      "step": 56605
    },
    {
      "entropy": 5.494384717941284,
      "epoch": 4.755975635370721,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029534182923710977,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21783762872219087,
      "num_tokens": 104446832.0,
      "step": 56610
    },
    {
      "entropy": 5.53623514175415,
      "epoch": 4.756395715185885,
      "grad_norm": 1.125,
      "learning_rate": 0.00029531200514495396,
      "loss": 4.9504,
      "mean_token_accuracy": 0.21487138718366622,
      "num_tokens": 104456337.0,
      "step": 56615
    },
    {
      "entropy": 5.522937870025634,
      "epoch": 4.75681579500105,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029528218069298137,
      "loss": 4.9048,
      "mean_token_accuracy": 0.22131994664669036,
      "num_tokens": 104464990.0,
      "step": 56620
    },
    {
      "entropy": 5.411896133422852,
      "epoch": 4.757235874816216,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002952523558817204,
      "loss": 4.876,
      "mean_token_accuracy": 0.22297078222036362,
      "num_tokens": 104474159.0,
      "step": 56625
    },
    {
      "entropy": 5.494415426254273,
      "epoch": 4.75765595463138,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029522253071169935,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22074614614248275,
      "num_tokens": 104482728.0,
      "step": 56630
    },
    {
      "entropy": 5.4702574729919435,
      "epoch": 4.758076034446545,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002951927051834466,
      "loss": 4.908,
      "mean_token_accuracy": 0.21923848241567612,
      "num_tokens": 104492163.0,
      "step": 56635
    },
    {
      "entropy": 5.540699911117554,
      "epoch": 4.7584961142617095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002951628792974904,
      "loss": 4.9709,
      "mean_token_accuracy": 0.21344014108181,
      "num_tokens": 104501053.0,
      "step": 56640
    },
    {
      "entropy": 5.4764257907867435,
      "epoch": 4.758916194076875,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029513305305435935,
      "loss": 4.9147,
      "mean_token_accuracy": 0.21930018663406373,
      "num_tokens": 104509825.0,
      "step": 56645
    },
    {
      "entropy": 5.469141101837158,
      "epoch": 4.759336273892039,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029510322645458144,
      "loss": 4.9232,
      "mean_token_accuracy": 0.209531731903553,
      "num_tokens": 104519332.0,
      "step": 56650
    },
    {
      "entropy": 5.485330867767334,
      "epoch": 4.759756353707204,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029507339949868535,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22185132205486296,
      "num_tokens": 104527981.0,
      "step": 56655
    },
    {
      "entropy": 5.433754587173462,
      "epoch": 4.7601764335223695,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002950435721871993,
      "loss": 4.8215,
      "mean_token_accuracy": 0.2282651588320732,
      "num_tokens": 104536700.0,
      "step": 56660
    },
    {
      "entropy": 5.409898853302002,
      "epoch": 4.760596513337534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002950137445206516,
      "loss": 4.8295,
      "mean_token_accuracy": 0.21552229374647142,
      "num_tokens": 104545458.0,
      "step": 56665
    },
    {
      "entropy": 5.421813726425171,
      "epoch": 4.761016593152699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002949839164995708,
      "loss": 4.8567,
      "mean_token_accuracy": 0.22348371148109436,
      "num_tokens": 104554514.0,
      "step": 56670
    },
    {
      "entropy": 5.387321329116821,
      "epoch": 4.761436672967864,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029495408812448524,
      "loss": 4.8822,
      "mean_token_accuracy": 0.21395178735256196,
      "num_tokens": 104562805.0,
      "step": 56675
    },
    {
      "entropy": 5.448543500900269,
      "epoch": 4.761856752783029,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002949242593959232,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2121448203921318,
      "num_tokens": 104571507.0,
      "step": 56680
    },
    {
      "entropy": 5.466728687286377,
      "epoch": 4.762276832598194,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029489443031441325,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2230543851852417,
      "num_tokens": 104581684.0,
      "step": 56685
    },
    {
      "entropy": 5.461110067367554,
      "epoch": 4.762696912413358,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029486460088048365,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22135195583105088,
      "num_tokens": 104590436.0,
      "step": 56690
    },
    {
      "entropy": 5.566982412338257,
      "epoch": 4.7631169922285235,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029483477109466295,
      "loss": 4.9804,
      "mean_token_accuracy": 0.21049199402332305,
      "num_tokens": 104600703.0,
      "step": 56695
    },
    {
      "entropy": 5.443513059616089,
      "epoch": 4.763537072043688,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029480494095747943,
      "loss": 4.8349,
      "mean_token_accuracy": 0.21570807695388794,
      "num_tokens": 104609583.0,
      "step": 56700
    },
    {
      "entropy": 5.36478443145752,
      "epoch": 4.763957151858853,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002947751104694616,
      "loss": 4.8032,
      "mean_token_accuracy": 0.22859407365322112,
      "num_tokens": 104618676.0,
      "step": 56705
    },
    {
      "entropy": 5.358781337738037,
      "epoch": 4.764377231674018,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029474527963113786,
      "loss": 4.8387,
      "mean_token_accuracy": 0.2203911602497101,
      "num_tokens": 104627837.0,
      "step": 56710
    },
    {
      "entropy": 5.4530134201049805,
      "epoch": 4.764797311489183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029471544844303664,
      "loss": 4.9004,
      "mean_token_accuracy": 0.21332940012216567,
      "num_tokens": 104637058.0,
      "step": 56715
    },
    {
      "entropy": 5.457220411300659,
      "epoch": 4.765217391304348,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029468561690568646,
      "loss": 4.8806,
      "mean_token_accuracy": 0.21517267674207688,
      "num_tokens": 104646189.0,
      "step": 56720
    },
    {
      "entropy": 5.455999279022217,
      "epoch": 4.765637471119513,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002946557850196157,
      "loss": 4.909,
      "mean_token_accuracy": 0.21815191507339476,
      "num_tokens": 104655823.0,
      "step": 56725
    },
    {
      "entropy": 5.487800121307373,
      "epoch": 4.766057550934677,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029462595278535275,
      "loss": 4.9187,
      "mean_token_accuracy": 0.2079997643828392,
      "num_tokens": 104665929.0,
      "step": 56730
    },
    {
      "entropy": 5.441781377792358,
      "epoch": 4.766477630749843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002945961202034262,
      "loss": 4.8452,
      "mean_token_accuracy": 0.2172746852040291,
      "num_tokens": 104674574.0,
      "step": 56735
    },
    {
      "entropy": 5.452502727508545,
      "epoch": 4.766897710565007,
      "grad_norm": 1.125,
      "learning_rate": 0.0002945662872743645,
      "loss": 4.9547,
      "mean_token_accuracy": 0.20947134047746657,
      "num_tokens": 104684185.0,
      "step": 56740
    },
    {
      "entropy": 5.575027561187744,
      "epoch": 4.767317790380172,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029453645399869603,
      "loss": 4.9526,
      "mean_token_accuracy": 0.2133770078420639,
      "num_tokens": 104694681.0,
      "step": 56745
    },
    {
      "entropy": 5.461897134780884,
      "epoch": 4.7677378701953375,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002945066203769494,
      "loss": 4.914,
      "mean_token_accuracy": 0.2120026469230652,
      "num_tokens": 104703979.0,
      "step": 56750
    },
    {
      "entropy": 5.381806707382202,
      "epoch": 4.768157950010502,
      "grad_norm": 1.1875,
      "learning_rate": 0.000294476786409653,
      "loss": 4.7786,
      "mean_token_accuracy": 0.23407235741615295,
      "num_tokens": 104713434.0,
      "step": 56755
    },
    {
      "entropy": 5.403412485122681,
      "epoch": 4.768578029825667,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002944469520973354,
      "loss": 4.8719,
      "mean_token_accuracy": 0.22087038457393646,
      "num_tokens": 104722310.0,
      "step": 56760
    },
    {
      "entropy": 5.456198024749756,
      "epoch": 4.768998109640831,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000294417117440525,
      "loss": 4.8635,
      "mean_token_accuracy": 0.21794142723083496,
      "num_tokens": 104731169.0,
      "step": 56765
    },
    {
      "entropy": 5.460400867462158,
      "epoch": 4.769418189455997,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002943872824397504,
      "loss": 4.8803,
      "mean_token_accuracy": 0.2237315744161606,
      "num_tokens": 104741172.0,
      "step": 56770
    },
    {
      "entropy": 5.360224294662475,
      "epoch": 4.769838269271162,
      "grad_norm": 1.125,
      "learning_rate": 0.00029435744709554014,
      "loss": 4.8391,
      "mean_token_accuracy": 0.21442675590515137,
      "num_tokens": 104750945.0,
      "step": 56775
    },
    {
      "entropy": 5.3944295883178714,
      "epoch": 4.770258349086326,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002943276114084226,
      "loss": 4.8167,
      "mean_token_accuracy": 0.2232194200158119,
      "num_tokens": 104759987.0,
      "step": 56780
    },
    {
      "entropy": 5.4748646259307865,
      "epoch": 4.770678428901491,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002942977753789263,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21761527806520461,
      "num_tokens": 104768413.0,
      "step": 56785
    },
    {
      "entropy": 5.339035749435425,
      "epoch": 4.771098508716657,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029426793900758,
      "loss": 4.7163,
      "mean_token_accuracy": 0.22824448496103286,
      "num_tokens": 104777146.0,
      "step": 56790
    },
    {
      "entropy": 5.482869291305542,
      "epoch": 4.771518588531821,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000294238102294912,
      "loss": 4.9191,
      "mean_token_accuracy": 0.20865580439567566,
      "num_tokens": 104787157.0,
      "step": 56795
    },
    {
      "entropy": 5.390078496932984,
      "epoch": 4.771938668346986,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002942082652414509,
      "loss": 4.8332,
      "mean_token_accuracy": 0.2167299821972847,
      "num_tokens": 104795977.0,
      "step": 56800
    },
    {
      "entropy": 5.505529069900513,
      "epoch": 4.7723587481621506,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002941784278477253,
      "loss": 4.9008,
      "mean_token_accuracy": 0.2202059283852577,
      "num_tokens": 104806662.0,
      "step": 56805
    },
    {
      "entropy": 5.513739538192749,
      "epoch": 4.772778827977316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002941485901142637,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2142232894897461,
      "num_tokens": 104815505.0,
      "step": 56810
    },
    {
      "entropy": 5.431328916549683,
      "epoch": 4.77319890779248,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029411875204159475,
      "loss": 4.882,
      "mean_token_accuracy": 0.2203554853796959,
      "num_tokens": 104825414.0,
      "step": 56815
    },
    {
      "entropy": 5.47949914932251,
      "epoch": 4.773618987607645,
      "grad_norm": 1.125,
      "learning_rate": 0.00029408891363024694,
      "loss": 4.9159,
      "mean_token_accuracy": 0.22305352836847306,
      "num_tokens": 104834451.0,
      "step": 56820
    },
    {
      "entropy": 5.475981903076172,
      "epoch": 4.774039067422811,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029405907488074884,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21090639978647233,
      "num_tokens": 104842762.0,
      "step": 56825
    },
    {
      "entropy": 5.53534951210022,
      "epoch": 4.774459147237975,
      "grad_norm": 1.25,
      "learning_rate": 0.000294029235793629,
      "loss": 4.9705,
      "mean_token_accuracy": 0.21524689644575118,
      "num_tokens": 104851497.0,
      "step": 56830
    },
    {
      "entropy": 5.3614531517028805,
      "epoch": 4.77487922705314,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002939993963694161,
      "loss": 4.7957,
      "mean_token_accuracy": 0.22757126539945602,
      "num_tokens": 104859882.0,
      "step": 56835
    },
    {
      "entropy": 5.424849224090576,
      "epoch": 4.775299306868305,
      "grad_norm": 1.125,
      "learning_rate": 0.0002939695566086387,
      "loss": 4.819,
      "mean_token_accuracy": 0.22422731965780257,
      "num_tokens": 104868761.0,
      "step": 56840
    },
    {
      "entropy": 5.46895432472229,
      "epoch": 4.77571938668347,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029393971651182525,
      "loss": 4.8828,
      "mean_token_accuracy": 0.21189365833997725,
      "num_tokens": 104878154.0,
      "step": 56845
    },
    {
      "entropy": 5.514490175247192,
      "epoch": 4.776139466498635,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002939098760795046,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22494169920682908,
      "num_tokens": 104886928.0,
      "step": 56850
    },
    {
      "entropy": 5.489462280273438,
      "epoch": 4.776559546313799,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029388003531220514,
      "loss": 4.9743,
      "mean_token_accuracy": 0.2048697903752327,
      "num_tokens": 104895272.0,
      "step": 56855
    },
    {
      "entropy": 5.370488452911377,
      "epoch": 4.7769796261289645,
      "grad_norm": 1.25,
      "learning_rate": 0.0002938501942104557,
      "loss": 4.8488,
      "mean_token_accuracy": 0.21774516701698304,
      "num_tokens": 104905611.0,
      "step": 56860
    },
    {
      "entropy": 5.4079108238220215,
      "epoch": 4.77739970594413,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029382035277478457,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22639665603637696,
      "num_tokens": 104914650.0,
      "step": 56865
    },
    {
      "entropy": 5.442041254043579,
      "epoch": 4.777819785759294,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029379051100572075,
      "loss": 4.8033,
      "mean_token_accuracy": 0.22163841724395753,
      "num_tokens": 104923090.0,
      "step": 56870
    },
    {
      "entropy": 5.453952789306641,
      "epoch": 4.778239865574459,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029376066890379266,
      "loss": 4.8479,
      "mean_token_accuracy": 0.21493091732263564,
      "num_tokens": 104932328.0,
      "step": 56875
    },
    {
      "entropy": 5.407171821594238,
      "epoch": 4.778659945389624,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029373082646952896,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21095310300588607,
      "num_tokens": 104942832.0,
      "step": 56880
    },
    {
      "entropy": 5.508794164657592,
      "epoch": 4.779080025204789,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029370098370345835,
      "loss": 4.8664,
      "mean_token_accuracy": 0.214994153380394,
      "num_tokens": 104952309.0,
      "step": 56885
    },
    {
      "entropy": 5.492534017562866,
      "epoch": 4.779500105019954,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002936711406061095,
      "loss": 4.921,
      "mean_token_accuracy": 0.2211290568113327,
      "num_tokens": 104961599.0,
      "step": 56890
    },
    {
      "entropy": 5.454746961593628,
      "epoch": 4.7799201848351185,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029364129717801085,
      "loss": 4.8964,
      "mean_token_accuracy": 0.213406041264534,
      "num_tokens": 104970605.0,
      "step": 56895
    },
    {
      "entropy": 5.470461368560791,
      "epoch": 4.780340264650284,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002936114534196914,
      "loss": 4.9413,
      "mean_token_accuracy": 0.2122248411178589,
      "num_tokens": 104980149.0,
      "step": 56900
    },
    {
      "entropy": 5.426455068588257,
      "epoch": 4.780760344465448,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029358160933167947,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2214935764670372,
      "num_tokens": 104989382.0,
      "step": 56905
    },
    {
      "entropy": 5.481209993362427,
      "epoch": 4.781180424280613,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029355176491450405,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21397133469581603,
      "num_tokens": 104998644.0,
      "step": 56910
    },
    {
      "entropy": 5.541811323165893,
      "epoch": 4.7816005040957785,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002935219201686936,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21919865161180496,
      "num_tokens": 105007847.0,
      "step": 56915
    },
    {
      "entropy": 5.415777206420898,
      "epoch": 4.782020583910943,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002934920750947769,
      "loss": 4.8164,
      "mean_token_accuracy": 0.21910788267850875,
      "num_tokens": 105016981.0,
      "step": 56920
    },
    {
      "entropy": 5.485897970199585,
      "epoch": 4.782440663726108,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002934622296932825,
      "loss": 4.8913,
      "mean_token_accuracy": 0.21033385992050171,
      "num_tokens": 105025467.0,
      "step": 56925
    },
    {
      "entropy": 5.456202268600464,
      "epoch": 4.782860743541272,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002934323839647393,
      "loss": 4.8707,
      "mean_token_accuracy": 0.21627719402313234,
      "num_tokens": 105033727.0,
      "step": 56930
    },
    {
      "entropy": 5.4560301303863525,
      "epoch": 4.783280823356438,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002934025379096758,
      "loss": 4.8115,
      "mean_token_accuracy": 0.2237611398100853,
      "num_tokens": 105041928.0,
      "step": 56935
    },
    {
      "entropy": 5.48328275680542,
      "epoch": 4.783700903171603,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029337269152862096,
      "loss": 4.8856,
      "mean_token_accuracy": 0.21428589522838593,
      "num_tokens": 105050905.0,
      "step": 56940
    },
    {
      "entropy": 5.464204597473144,
      "epoch": 4.784120982986767,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029334284482210326,
      "loss": 4.9058,
      "mean_token_accuracy": 0.21154846400022506,
      "num_tokens": 105060331.0,
      "step": 56945
    },
    {
      "entropy": 5.513382434844971,
      "epoch": 4.7845410628019325,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002933129977906516,
      "loss": 4.9572,
      "mean_token_accuracy": 0.21684547811746596,
      "num_tokens": 105070220.0,
      "step": 56950
    },
    {
      "entropy": 5.463869667053222,
      "epoch": 4.784961142617098,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029328315043479454,
      "loss": 4.9394,
      "mean_token_accuracy": 0.2220925733447075,
      "num_tokens": 105079250.0,
      "step": 56955
    },
    {
      "entropy": 5.4665491580963135,
      "epoch": 4.785381222432262,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002932533027550609,
      "loss": 4.8819,
      "mean_token_accuracy": 0.21302997767925264,
      "num_tokens": 105088906.0,
      "step": 56960
    },
    {
      "entropy": 5.414087533950806,
      "epoch": 4.785801302247427,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002932234547519794,
      "loss": 4.8532,
      "mean_token_accuracy": 0.21866913437843322,
      "num_tokens": 105097085.0,
      "step": 56965
    },
    {
      "entropy": 5.397283124923706,
      "epoch": 4.786221382062592,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002931936064260788,
      "loss": 4.7979,
      "mean_token_accuracy": 0.22989995330572127,
      "num_tokens": 105105525.0,
      "step": 56970
    },
    {
      "entropy": 5.378076028823853,
      "epoch": 4.786641461877757,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029316375777788784,
      "loss": 4.7931,
      "mean_token_accuracy": 0.22549301087856294,
      "num_tokens": 105114045.0,
      "step": 56975
    },
    {
      "entropy": 5.323417806625367,
      "epoch": 4.787061541692921,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029313390880793527,
      "loss": 4.8227,
      "mean_token_accuracy": 0.22167707681655885,
      "num_tokens": 105122925.0,
      "step": 56980
    },
    {
      "entropy": 5.4251940727233885,
      "epoch": 4.787481621508086,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002931040595167498,
      "loss": 4.8709,
      "mean_token_accuracy": 0.21553658097982406,
      "num_tokens": 105132419.0,
      "step": 56985
    },
    {
      "entropy": 5.475021600723267,
      "epoch": 4.787901701323252,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002930742099048603,
      "loss": 4.9302,
      "mean_token_accuracy": 0.2158082902431488,
      "num_tokens": 105141863.0,
      "step": 56990
    },
    {
      "entropy": 5.556888151168823,
      "epoch": 4.788321781138416,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002930443599727954,
      "loss": 4.9889,
      "mean_token_accuracy": 0.20653920471668244,
      "num_tokens": 105151336.0,
      "step": 56995
    },
    {
      "entropy": 5.441900014877319,
      "epoch": 4.788741860953581,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002930145097210839,
      "loss": 4.8654,
      "mean_token_accuracy": 0.22456731498241425,
      "num_tokens": 105161349.0,
      "step": 57000
    },
    {
      "epoch": 4.788741860953581,
      "eval_entropy": 5.187517064359351,
      "eval_loss": 5.007251739501953,
      "eval_mean_token_accuracy": 0.22067279205861912,
      "eval_num_tokens": 105161349.0,
      "eval_runtime": 27.3864,
      "eval_samples_per_second": 1364.399,
      "eval_steps_per_second": 170.559,
      "step": 57000
    }
  ],
  "logging_steps": 5,
  "max_steps": 119020,
  "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": 1.53925980650496e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}