swe-latn-100mb-after-ppt-Dp-100mb-ckpt500_seed3407/checkpoint-30000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.520394875026255,
  "eval_steps": 3000,
  "global_step": 30000,
  "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
    }
  ],
  "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": 8.0934496524288e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}